You are on page 1of 821

Females, Health

&

Sex in Jamaica

Paul A. Bourne $$

Females, Health

&

Sex in Jamaica

Females, Health

&

Sex in Jamaica

PAUL ANDREW BOURNE Director, Socio-Medical Research Institute

ii

Paul A. Bourne, 2011 First Published in Jamaica, 2011 by Paul Andrew Bourne 66 Long Wall Drive Stony Hill, Kingston 9, St. Andrew National Library of Jamaica Cataloguing Data

Females, Health & Sex in Jamaica


Includes index ISBN Bourne, Paul Andrew All rights reserved. Published , 2011 Cover designed by Paul Andrew Bourne

iii

Contents
Preface Acknowledgement vi viii

Part 1
Health, Illness and chronic health conditions Chapter 1 Health of females in Jamaica: using two cross-sectional surveys Chapter 2 The uninsured ill in a developing nation Chapter 3 Self-rated health of the educated and uneducated classes in Jamaica Chapter 4 Health status of patients with self-reported chronic diseases Chapter Chapter Chapter Chapter Chapter Chapter Chapter Chapter 5 6 7 8 9 10 11 12 1
1 21 52

in Jamaica 77 The changing faces of diabetes, hypertension and arthritis in a Caribbean population 103 Self-assessed health of young adults in an English-speaking Caribbean nation 129 Disparities in self-rated health, health care utilization, illness, chronic illness and other socioeconomic characteristics of the Insured and Uninsured 160 Good Health Status of Rural Women in the Reproductive Ages 192 Determinants of Quality of Life of Jamaican Women 225 Examining Health Status of Women in Rural, Peri-urban and Urban Areas in Jamaica 246 Social determinants of self-reported health across the Life Course 288 Modeling social determinants of self-rated health status of hypertensive in a middle-income developing nation 310

Part 2
Sex and reproductive practices of females Chapter 13 Factor Differentials in contraceptive use and demographic profile Chapter Chapter Chapter 14 15 16
among females who had their first coital activity at most 16 years versus those at 16+ years old in a developing nation Reproductive health matters: Women whose first sexual intercourse occurred at 20+ years old On sexual and non-intimate unions among the general reproductive population of women in Jamaica: A cross-sectional survey Sociodemographic correlates of age at sexual debut among women iv 335 368 392

of the reproductive years in a middle-income developing nation

430

Chapter Chapter Chapter Chapter Chapter

17 18 19 20 21

Current use of contraceptive method among women in a middle-income developing country Females with multiple sexual partners and their reproductive health matters: A comprehensive analysis of women aged 15-49 years in a developing nation Sexually assaulted females on their sexual debut: Reproductive health matters Females of the reproductive ages who have never used a condom with a non-steady sexual partner Multiple sexual partnerships among young adults in a tropically developing nation: A public health challenge

458

484 513 538 559

Part 3
Sex Chapter Chapter Chapter and reproductive practices of males 22 Psychosocial correlates of condom usage in a developing country 23 Young males whose first coitus began at most 15 years old 24 Young males who delay first coitus for the statutory age and
beyond in Jamaica 592 621 646

Part 4
Validity and reliability testing of survey data Chapter 25 The image of health status and quality of life in a Caribbean society 671 Chapter 26 Paradoxes in self-evaluated health data in a developing country 691 Chapter 27 The validity of using self-reported illness to measure objective health 716 Chapter 28 Dichotomising poor self-reported health status:
Using secondary cross-sectional survey data for Jamaica 736 758

Chapter

29

The quality of sample surveys in a developing nation

Part 5
Additional chapter Chapter 30 Self-rated health status of young adolescent females in a
middle-income developing country v 791

Preface
Many developing countries as well as developed nations continue to experience sexual explosions, the lowering of the age at first sexual intercourse. It appears that inspite of the inroads of public health practitioners to effectively tackle sanitary issues, water quality, vaccination and countless other reproductive health matters; they have failed in their efforts to adequately address the continuous lowering of the age at first coitus. This is equally the same in the United States. Jamaica like many societies has tried to revert the lowering of the age at first sexual relations, but this is to no avail. With the lowering of the age of sexual consent from 18 to 16 years in Jamaica, this is equally responsible for the continued sexual experimentation at an even younger age among adolescents as well as children (under 16 years). The high diet of sexual relations is not limited to young children, adolescents and young adults, the prevalence and incidence of among Jamaicans are exorbitantly high. Statistics revealed that almost 23 out of every 25 Jamaicans aged 15-74 years old have had have sex, 33 out of every 50 Jamaicans aged 15-74 years old had sex at least once per week, 21 out of every 25 Jamaicans aged 15-24 years had sex and about 41 in every 50 Jamaicans of the early age had sexual intercourse at least once per week (Wilks et al., 2008). The percentage of reported sexual intercourse increased with age in Jamaica (Wilks et al., 2008), suggesting that sexual relations must have some cultural underpinnings. Clearly from the aforementioned findings, Jamaicans are in a highly sexed people. One of the notable irony (or paradox) is that they (Jamaicans) are not openly expressive about sex, dislike public dialogues on the phenomenon, the older adults are speechless to advice young children and young adults about sexual expression, expect to say abstain, wait for adulthood and in time you will know what to do. The silent sexed culture is equally responsible for rape. Dialectic situations arise when females are raped (or sexually assaulted) as some people believe that the female is to be blamed for inviting the peer by her code of dress or common behaviour. Then there is sympathy for the perpetrator (rapist). Some people become empathetic toward the position of the perpetrators, with social expressions for the behaviour. More so, females are victims and sexually assaulted by

vi

powerful males because of poverty, and the economic power disparity protect the economically powerful males. In response to the aforementioned issues, this volume collated some search papers on various issues on health, sex and sexual experiences of females. Even though the issues are primarily on females, any discussion on sexuality must relate to both sexes. Therefore, I added an entire section on males sexual expression as this will broaden the discourse and provide clarity to females sexual issues. This book would have been incomplete if it had not examined the quality of survey data. I believe that scientists cannot accept cosmology, without questioning, testing and verifying that knowledge. It is as a result of questioning, further testing of knowledge and truths that truths are established, modified or changed with more information. Knowledge is not stationary; therefore, I sought to question the validity and reliability of survey data used in this text. The purpose was to provide readers with better understanding of findings, their roles in being skeptics, and how knowledge is created through questioning. I believe strongly in readable and engaging writing style, and so many complex concepts were simplified in keeping with my purpose to engage and connect with the readers. In some instances technical statistical terms and calculations were unavoidable. In those cases, I tried to explain the issues surrounding the technical terms for the readers to be adequately informed on the subject without a thorough knowledge of introductory or advanced statistics. Knowledge of introductory or advanced statistics will be good but not necessarily for the readers. The majority of the chapters are 1) published in peer reviewed journals, and 2) solely written by yours truly. However, a few chapters are co-authored with Caribbean and International scholars. This book will broaden the discourse as it represents a useful contribution to the literature that is Jamaican in scope. Paul Andrew Bourne 2011

vii

Acknowledgements

Many people have contributed to the completion of this book. I would like to extend my sincere gratitude to them. I would like to single out, 1) Ms. Neva South-Bourne for her advice in penning my ideas, 2) Mrs. Evadney Bourne, my wife, for support, understanding and patience when things were difficult and surmountable at times, 3) all my co-authors, 4) God, for his wisdom, 5) the Data Bank in Sir Arthur Lewis Institute of Social and Economic Studies, the University of the West Indies, Mona, Jamaica for making the dataset available for use in this study, and 6) all my associates (including best friends) whose love, support and encouragement provided the impetus that I drew from to complete this project.

viii

1
Health of females in Jamaica: using two cross-sectional surveys
The 21st Century cannot have researchers examining self-rated health status of elderly, population, children and adolescents and not single out females as they continue to be poorer than males; and are exposed to different socioeconomic situation. The current paper 1) examines the health conditions; 2) provides an epidemiological profile of changing health conditions in the last one half decade; 3) evaluates whether self-reported illness is a good measure of self-rated health status; 4) computes the mean age of females having particular health conditions; 5) calculates the mean age of being ill compared with those who are not ill; and 6) assesses the correlation between self-rated health status and income quintile. There is reduction in the mean age of females reported being diagnosed with chronic illness such as diabetes mellitus (60.54 17.14 years); hypertension (60.85 16.93 years) and arthritis 59.72 15.41 years). In 2007 over 2002, the mean age of females with unspecified health conditions fell by 33%. Although healthy life expectancy for females at birth in Jamaica was 66 years which is greater than that for males, improvements in their self-rated health status cannot be neglected as there are shifts in health conditions towards diabetes mellitus and a decline in the mean age at which females are diagnosed with particular chronic illnesses.

Introduction

Life expectancy is among the objective indexes for measuring health for a person, society, or population. In 1880-1882, life expectancy at birth for females in Jamaica was 39.8 years which was 2.79 years more than that for males. One hundred and twenty-two years later, health disparity increased to 5.81 years: in 2002-2004, life expectancy at birth for females was 77.07
1

years [1]. For the world, the difference in life expectancy for the sexes was 4.2 years more for females than males: for 2000-2005, life expectancy at birth for females was 68.1 years [2]. Within the expanded conceptual framework offered by the World Health Organization (WHO) in the late 1940s, health is more than the absence of morbidity as it includes social, psychological and physiological wellbeing [3]. Some scholars [4] opined that using the opposite of ill-health to measure health is a negative approach as health is more than this biomedical approach. Brannon and Feist [4] forwarded a positive approach which is in keeping with the Biopsychosocial framework developed by Engel. Engel coined the term Biopsychosocial when he forwarded the perspective that patient care must integrate the mind, body and social environment [5-8]. He believed that mentally patient care is not merely about the illness, as other factors equally influence the health of the patient. Although this was not new because the WHO had already stated this, it was the application which was different from the traditional biomedical approach to the study and treatment of ill patients. Embedded in Engels works were wellbeing, wellness and quality of life and not merely the removal of the illness, which psychologists like Brannon and Feist called the positive approach to the study and treatment of health. Recognizing the limitation of life expectancy, WHO therefore developed DALE Disability Adjusted Life Expectancy which discounted life expectancy by number of years spent in illness. The emphasis in the 21st Century therefore was healthy life and not length of life (ie life expectancy) [9]. DALE is the years in ill health which is weighted according to severity, which is then subtracted from the expected overall life expectancy to give the equivalent healthy years of life. Using healthy years, statistics revealed that the health disparity between the sexes in
2

Jamaica was 5 years in 2007 [10], indicating that self-rated health status of females on average in Jamaica is better than that for males. This is not atypical to Jamaica as females in many nations had a greater healthy life expectancy than males. The discipline of public health is concerned with more than accepting the health disparity as indicated by life expectancy or healthy life expectancy, as it seeks to improve the quality of life of the populace and the various subgroups that are within a particular geographical border. In order for this mandate to be attained, we cannot exclude the study of females health merely because they are living longer than males and accept this as a given; and that there is not need therefore to examine their self-rated health status. Many empirical studies that have examined health of Caribbean nationals were on the population [11-15]; elderly [16-25]; children [26, 27]; adolescents [28-30] and females have been omitted from the discourse. A comprehensive search of health literature in Caribbean in particular Jamaica revealed no studies. The values for the healthy life expectancy cannot be enough to indicate the self-rated health status of females neither can we use self-rated health status of population, children, elderly and adolescents to measure that of females. WHO [31] forwarded a position that there is a disparity between contracting many diseases and the gender constitution of an individual, suggesting that population health cannot be used to measure female health. Females have a high propensity than males to contract particular conditions such as depression, osteoporosis and osteoarthritis [31]. A study conducted by McDonough and Walters [32] revealed that women had a 23 percent higher distress score than men and were more likely to report chronic diseases compared to males (30%). It was found that men believed their health was better (2% higher) than that self-reported by females.
3

McDonough and Walters used data from a longitudinal study named Canadian National Population Health Survey (NPHS). Those aforementioned realities justify a study on female health in Jamaica. The current paper fills the gap in the health literature by investigating health of females in Jamaica. The objectives of the current paper are 1) to examine the health conditions; 2) provide an epidemiological profile of changing health conditions in the last one half decade (2002-2007); 3) evaluate whether self-reported illness is a good measure of self-rated health status; 4) compute the mean age of females having particular health conditions; 5) calculate the mean age of being ill compared with those who are not ill; and 6) assess the correlation between self-rated health status and income quintile.

Materials and methods


Sample The current paper extracted subsample of females from two secondary cross-sectional data collected by the Planning Institute of Jamaica and the Statistical Institute of Jamaica [33, 34]. In 2002, a subsample of 12,675 females was extracted from the sample of 25,018 respondents and for 2007; a subsample of 3,479 females was extracted from 6,783 respondents. The survey is called the Jamaica Survey of Living Conditions (JSLC) which began in 1989. The JSLC is modification of the World Banks Living Standards Measurement Study (LSMS) household survey. A self-administered questionnaire is used to collect the data from Jamaicans. Trained data collectors are used to gather the data; and these individuals are trained by the Statistical Institute of Jamaica
4

The survey was drawn using stratified random sampling. This design was a two-stage stratified random sampling design where there was a Primary Sampling Unit (PSU) and a selection of dwellings from the primary units. The PSU is an Enumeration District (ED), which constitutes a minimum of 100 residences in rural areas and 150 in urban areas. An ED is an independent geographic unit that shares a common boundary. This means that the country was grouped into strata of equal size based on dwellings (EDs). Based on the PSUs, a listing of all the dwellings was made, and this became the sampling frame from which a Master Sample of dwelling was compiled, which in turn provided the sampling frame for the labour force. One third of the Labour Force Survey (i.e. LFS) was selected for the JSLC. The sample was weighted to reflect the population of the nation. The non-response rate for the survey for 2007 was 26.2% and 27.7%. Measures Self-reported illness (or Health conditions): The question was asked: Is this a diagnosed recurring illness? The answering options are: Yes, Cold; Yes, Diarrhoea; Yes, Asthma; Yes, Diabetes; Yes, Hypertension; Yes, Arthritis; Yes, Other; and No. Self-rated health status (self-rated health status): How is your health in general? And the options were very good; good; fair; poor and very poor. The first time this was collected for Jamaicans, using the JSLC, was in 2007. Social class: This variable was measured based on the income quintiles: The upper classes were those in the wealthy quintiles (quintiles 4 and 5); middle class was quintile 3 and poor those in lower quintiles (quintiles 1 and 2).
5

Health care-seeking behaviour. This is a dichotomous variable which came from the question Has a doctor, nurse, pharmacist, midwife, healer or any other health practitioner been visited? with the option (yes or no). Statistical analysis The data were collected, stored and retrieved in SPSS for Windows 16.0 (SPSS Inc; Chicago, IL, USA). Descriptive statistics were used to provide information on the socio-demographic variables of the sample. Cross Tabulations were employed to examine correlations between nonmetric variables and Analysis of Variance (ANOVA) were utilized to examine statistical associations between a metric and non-metric variable. The level of significance used in this research was 5% (ie 95% confidence interval). Bryman and Cramer [35] correlation coefficient values were used to determine, the strength of a relation between (or among) variables: 0.19 and below, very low; 0.20 to 0.39, low; 0.40 to 0.69, moderate; 0.70 to 0.89, high (strong); and 0.90 to 1 is very high (very strong).

Results
Demographic characteristic of sample In 2002, 14.7% of sample reported an illness and this increased by 19.1% in 2007. Over the same period, health insurance coverage increased by 81.0% (to 21.0% in 2007); those seeking medical care increased to 67.6% (from 66.0%); the mean age in 2007 was 30.621.9 years which marginal increased from 29.4 22.3 years; diabetic cases exponentially increased by 227.7% (in 2007, 15.4%); hypertension decline by 45.5% (to 24.8% in 2007) and arthritic cases fell by 66.1% (to 9.4% in 2007). Urbanization was evident between 2007 and 2002 as the number of
6

females who resided in urban areas increased by 114.7% (to 30.4% in 2007), with a corresponding decline of 19.4% in females zones. Table 1.1 revealed that the increase in self-reported illness was substantially accounted for by increased cases in the rural sample (from 12.9% in 2002 to 20.0% in 2007). The drastic increase in health insurance coverage in 2007 was due to public establishment of public health insurance coverage. The greatest increase was observed in semi-urban areas 17.8%) followed by urban (9.6%) and rural (7.8%) Table 1.1. The increases in self-reported illness can be accounted for by diabetes mellitus, asthma and other dysfunctions. Concurrently, most of the increased cases were diabetic in semi-urban zones (17.1%); other health conditions in semi-urban areas (12.4%) and asthma in urban zones (12.0%) (Table 1.1). Bivariate analyses There was a significant statistical correlation between self-rated health status and self-reported illness - 2 (df = 4) = 700.633, P < 0.001; with there being a negative moderate relation between the variables correlation coefficient = - 0.412(Table 1.2). Based on Table 1.2, 10.7% of those who reported an illness had had very good self-rated health status compared to 40.2% of those who did not indicate an illness. On the other hand, 2.5% of those who did not report a dysfunction had at least poor self-rated health status compared to 19.8% of those who indicated having an illness. Even after controlling self-rated health status and self-reported illness by age, marital status and per capita annual expenditure, a moderate negative correlation was found correlation coefficient = - 0.362. On further examination of the self-reported illness by age, it was found that in 2002 the mean age of individual who reported an illness was 43.97 26.81 years compared to 27.05
7

20.41 years for who without an illness t-test = 30.818, P < 0.001. In 2007, the mean age of reporting an illness was 42.83 26.53 years compared to 28.16 19.95 years for those who did not report an ailment t-test = 15.263, P < 0.001. Based on Figure 1.1, there is an increase in the mean age of females being diagnosed with diarrhoea (32.00 36.2 years) and asthma (21.73 20.51 years). However, there is reduction in the mean age of females reported being diagnosed with chronic illness such as diabetes mellitus (60.54 17.14 years); hypertension (60.85 16.93 years) and arthritis 59.72 15.41 years). The greatest decline in mean age of chronically ill diagnosed females was in arthritic cases (by 7.41 years). Concurrently, the mean age of females with unspecified health conditions fell by (33%, from 54.62 21.77 years in 2002 to 36.42 23.69 years in 2007). A cross tabulation between self-rated health status and income quintile revealed a significant statistical correlation - 2 (df = 16) = 54.044, P < 0.001; with the relationship being a very weak one correlation coefficient = 0.126 (Table 1.3). Based on Table 1.3, the wealthy reported the greatest self-rated health status (ie very good) compared to the wealthiest 20% (36.7%); with the poorest 20% recorded the least very good self-rated health status. No significant statistical correlation was found between diagnosed self-reported illness and income quintile - 2 (df = 28) = 36.161, P > 0.001 (Table 1.4).

Discussion
Self-rated health status of female Jamaicans can be measured using self-reported illness. The current paper found a moderate significant correlation between the two aforementioned variables, suggesting that self-reported illness is a relatively good measure of females health. In
8

this study it was revealed that 60 out of every 100 who reported an illness had at most fair selfrated health status, with 20 out every 100 indicated a least poor health. It is evident from the findings that self-rated health status is wider than illness, which concurs with the literature [35, 36], which is keeping with the propositions of the WHO that health must be more than the absence of illness. Self-rated health status is peoples self-rated perspective on their general selfrated health status [35], which includes a percentage of poor health (or ill-health). The other components of this status include life satisfaction, happiness, and psychosocial wellbeing. Using a sample of elderly Barbadians, Hambleton et al [37] found 33.5% of explanatory power of selfrated health status is accounted for by illness. There is a disparity between the current paper and that of Hambleton et als work as more of self-rated health status of the elderly is explained by current illness with this being less for females in Jamaica. Concomitantly, there is an epidemiological shift in the typology of illnesses affecting females as the change is towards diabetes mellitus. In 2007 over 2002, the 15 out of every 100 females reported being diagnosed with diabetes mellitus compared to 5 in 100 in 2002 indicating the negative effects of life behaviour of females self-rated health status. Another important finding of the current paper is that diagnosed illnesses are not significantly different based on income quintile in which a female is categorized. However, the self-rated health status of females in different social standing (measured using income quintile) is different. Embedded in this finding is the role of income plays in improving self-rated health status [38]. Like Marmot [38], this study found that income is able to buy some improvement in self-rated health status; but this work goes further as it found that income does not reduce the typology in health conditions affecting females. Before this discussion can proceed, the discourse must address the biases in subjective indexes which are found in studies like this one. Any study on subjective indexes in the
9

measurement of health (for example, happiness, life satisfaction; self-rated health status, selfreported illness) needs to address the challenges of biases that are found in self-reported data in particular self-reported health data. The discourse of subjective wellbeing using survey data cannot deny that it is based on the persons judgement, and must be prone to systematic and nonsystematic biases [40]. Diener [36] argued that the subjective measure seemed to contain substantial amounts of valid variance, suggesting that there is validity to the use of this approach in the measurement of health (or wellbeing) like the objective indexes such as life expectancy, mortality or diagnosed morbidity. A study by Finnas et al [41] opined that there are some methodological issues surrounding the use of self-reported (or self-rated) health and that these may result in incorrect inference; but that this measure is useful in understanding health, morbidity and mortality. Using life expectancy and self-reported illness data for Jamaicans, Bourne [42] found a strong significant correlation between the two variables (correlation coefficient, R = - 0.731), and that self-reported illness accounted for 54% of the variance in life expectancy. When Bourne [42] disaggregated the life expectancy and self-reported illness data by sexes, he found a strong correlation between males health (correlation coefficient, R = 0.796) than for females (correlation coefficient, R = 0.684). Self-reported data therefore do have some biases; but that it is good measure for health in Jamaica and more so for males. In spite of this fact, the current research recognized some of the problems in using self-reported health data (read Finnas et al. [41] for more information), while providing empirical findings using peoples perception on their health. Now that the discourse on objective and subjective indexes is out of the way, the next issue of concern is the reduced aged of reported illness and age of being diagnosed with
10

particular chronic illness. In 2002, the mean age recorded for those who self-reported an illness was 44 years and this fell by 1 year in 2007, indicating that on average females are becoming diagnosed with an illness on average 2 months earlier. When self-reported illness was

disaggregated into acute and chronic health conditions, it was revealed that on average females were being diagnosed 7.41 years earlier with arthritis in 2007 over 2002; 4.95 years earlier with hypertension and 1.13 years earlier with diabetes mellitus.

Conclusion
The current paper revealed that rural females recorded the highest percentage of self-reported illness. Concurrently, in 2007, 20 out of every 100 females in rural Jamaica reported an ailment which is a 3.7% increase over 2002 compared to a 3.1% increase in urban and 2.2% increase in semi-urban females. Furthermore, poverty was greatest for rural females. In 2002, poverty among rural females was 2.2 times more than urban poverty; and this increased to 3.3 times in 2007. In addition to the aforementioned issues, there is a shift in chronic illnesses occurring in females in Jamaica. Hypertension and arthritis have seen a decline in 2007 over 2002; however, there were noticeable increases in diabetes mellitus over the same period. The greatest increase in cases of diabetes mellitus occurred in semi-urban females followed by urban and rural females. In summing, the current paper has revealed that, although healthy life expectancy for females at birth in Jamaica is 66 years, improvements in their self-rated health status cannot be neglected as there are shifts in health conditions (to diabetes mellitus) as well as the decline in ages at which females are being diagnosed with particular chronic illnesses. There is an issue
11

which emerged from the current finding, the increasing cases of unspecified illness among females and this must be examined as to classification in order that public health practitioners will be able to address it before it unfolds into a public health challenge in the future.

References
1. Statistical Institute of Jamaica, (STATIN). Demographic statistics, 2005. Kingston: STATIN; 2006. 2. Department of Economic and Social Affairs Population Division, United Nations, (UN). World population ageing 19590-2050. New York: United Nations; 2002. 3. World Health Organization, (WHO). Preamble to the Constitution of the World Health Organization as adopted by the International Health Conference, New York, June 19-22, 1946; signed on July 22, 1946 by the representatives of 61 States (Official Records of the World Health Organization, no. 2, p. 100) and entered into force on April 7, 1948. Constitution of the World Health Organization, 1948. In Basic Documents, 15th ed. Geneva, Switzerland: WHO, 1948. 4. Brannon L, Feist J. Health psychology. An introduction to behavior and health 6th ed. Los Angeles: Thomson Wadsworth; 2007. 5. Engel G. 1960. A unified concept of health and disease. Perspectives in Biology and Medicine 1960;3:459-485. 6. Engel G. The care of the patient: art or science? Johns Hopkins Medical Journal 1977;140:222-232. 7. Engel G. The need for a new medical model: A challenge for biomedicine. Science 1977;196:129-136. 8. Engel G . The biopsychosocial model and the education of health professionals. Annals of the New York Academy of Sciences 1978;310: 169-181. 9 WHO. WHO Issues New Healthy Life Expectancy Rankings: Japan Number One in New Healthy Life System. Washington & Switzerland: WHO; 2000. 10. WHO. World health statistics, 2009. Geneva: WHO; 2009.

12

11. Bourne, P.A. (2009). A theoretical framework of good health status of Jamaicans: using econometric analysis to model good health status over the life course. North American Journal of Medical Sciences, 1(2): 86-95. 12. Bourne PA. Socio-demographic determinants of Health care-seeking behaviour, selfreported illness and Self-evaluated Health status in Jamaica. International Journal of Collaborative Research on Internal Medicine & Public Health, 2009, 1 (4):101-130. 13. Bourne PA. Health Determinants: Using Secondary Data to Model Predictors of Wellbeing of Jamaicans. West Indian Med J. 2008; 57:476-81. 14. Hutchinson G, Simeon DT, Bain BC, Wyatt GE, Tucker MB, LeFranc E. 2004. Social and Health determinants of well-being and life satisfaction in Jamaica. Int J of Soci Psychiatry. 2004;50:43-53. 15. Asnani MR, Reid ME, Ali SB, Lipps G, Williams-Green. Quality of life in patients with sickle cell disease in Jamaica: rural-urban differences. J of Rural and Remote health 2008;8:890. 16. Hambleton IR, Clarke K, Broome HL, Fraser HS, Brathwaite F, Hennis AJ. 2005. Historical and current predictors of self-reported health status among elderly persons in Barbados. Rev Pan Salud Public. 17: 342-352. 17. Brathwaite FS. The elderly in the Commonwealth Caribbean: A review of research findings. Ageing and Society 1989;9:297-304. 18. Brathwaite FS. The elderly in Barbados: problem and policies. Bulletin of the Pan American Health Organization 1990;23:314-29. 19.Eldemire D. The Jamaican elderly: A socioeconomic perspective and policy implications. Social and Economic Studies 1997;46: 175-193. 20.Eldemire D. Older women: A situational analysis, Jamaica 1996. New York: United Nations Division for the Advancement of Women; 1996. 21. Palloni A, Pinto-Aguirre G, Pelaez M. Demographic and health conditions of ageing in Latin America and the Caribbean. Int J of Epidemiology 2002;31:762-771. 22. Eldemire D. The elderly and the family: The Jamaican experience. Caribbean Affairs 1994;19:31-46. Bulletin of Eastern

23. Bourne PA, McGrowder DA, Crawford TV. Decomposing Mortality Rates and Examining Health Status of the Elderly in Jamaica. The Open Geriatric Med J. 2009; 2:34-44.

13

24. Bourne PA. Good Health Status of Older and Oldest Elderly in Jamaica: Are there differences between rural and urban areas? Open Geriatric Medicine Journal. 2009; 2:18-27. 25. Bourne PA. 2008. Medical Sociology: Modelling Well-being for elderly People in Jamaica. West Indian Med J 57:596-04. 26. Walker S. Nutrition and child health development. In: Morgan W, editor. Health issues in the Caribbean. Kingston: Ian Randle; 2005: p. 15-25. 27. Samms-Vaughn M, Jackson M, Ashley D. School achievement and behaviour in Jamaican children. In: Morgan W, editor. Health issues in the Caribbean. Kingston: Ian Randle; 2005: p. 26-37. 28. Frederick J, Hamilton P, Jackson J, Frederick C, Wynter S, DaCosta V, Wynter H. Issues affecting reproductive health in the Caribbean. In: Morgan W, editor. Health issues in the Caribbean. Kingston: Ian Randle; 2005: p. 41-50. 29. Bourne PA. Demographic shifts in health conditions of adolescents 10-19 years, Jamaica: Using cross-sectional data for 2002 and 2007. North American Journal of Medical Sciences 2009; 1:125-133. 30. Blum RW, Halcon L, Beuhring T, Pate E, Campbell-Forrester S, Venema A. Adolescent heath in the Caribbean: Risk and protective factors. American Journal of Public Health 2003; 93: 456-460. 31. WHO. Ageing and health, epidemiology. Regional Office in Africa: WHO; 2005. 32. McDonough P, Walters V. Gender and health: reassessing patterns and explanations. Social Science and Medicine 2001; 52:547-559. 33. Statistical Institute Of Jamaica. Jamaica Survey of Living Conditions, 2002 [Computer file]. Kingston, Jamaica: Statistical Institute Of Jamaica [producer], 2002. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors], 2003. 34.Statistical Institute Of Jamaica. Jamaica Survey of Living Conditions, 2007 [Computer file]. Kingston, Jamaica: Statistical Institute Of Jamaica [producer], 2007. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors], 2008. 35. Bryman A, Cramer D. Quantitative data analysis with SPSS 12 and 13: a guide for social scientists. London and New York: Routledge; 2005: p. 214-219.

14

35. Kahneman D, Riis J. Living, and thinking about it, two perspectives. In: Huppert FA, Kaverne B, Baylis N, editors. The science of well-being: Integrating neurobiology, psychology, and social science. London: Oxford University Press; 2005. p. 285-304. 36. Diener E. Subjective well-being. Psychological Bulletin, 1984;95:54275 37. Hambleton IR, Clarke K, Broome HL, Fraser HS, Brathwaite F, Hennis, A.J. Historical and current predictors of self-reported health status among elderly persons in Barbados. Revista Panamericana de salud Pblic 2005; 17, 342-352. 38. Marmot M .The influence of Income on Health: Views of an Epidemiologist. Does money really matter? Or is it a marker for something else? Health Affairs 2002; 21, pp.31-46. 40. Schwarz N, Strack F. Reports of subjective well-being: judgmental processes and their methodological implications. In: Kahneman D, Diener E, Schwarz N, editors. Well-being: The Foundations of Hedonic Psychology. Russell Sage Foundation: New York; 1999;pp 61-84. 41. Finnas F, Nyqvist F, Saarela J. Some methodological remarks on self-rated health. The Open Public Health J 2008;1:32-39. 42. Bourne P. Is self-reported health a good measure of objective health? North American J of Medical Sciences. In print.

15

Table 1.1. Sociodemographic characteristics of sample by area of residence, 2002 and 2007
2002 Variable Rural Marital status Married Never married Divorced Separated Widowed Income quintile Poorest 20% Poor Middle Wealthy Wealthiest 20% Health conditions Diagnosed Acute: Cold Diarrhoea Asthma Diagnosed Chronic: Diabetes mellitus Hypertension Arthritis Other Non-diagnosed Self-reported illness Yes No Health care-seekers Yes No Health insurance Yes, Private Yes, Public No Age Mean (SD) in yrs 1232 (25.7) 3033 (63.3) 25 (0.5) 51 (1.1) 453 (9.4) 1864 (24.8) 1867 (24.8) 1559 (20.7) 1340 (17.8) 894 (11.9) SemiUrban 568 (25.7) 1452 (65.7) 16 (0.7) 27 (1.2) 147 (6.7) 450 (13.5) 511 (15.3) 652 (19.2) 759 (22.7) 965 (28.9) Urban 243 (19.3) 907 (71.9) 18 (1.4) 22 (1.7) 71 (5.6) 206 (11.4) 231 (12.7) 331 (18.2) 441 (24.3) 605 (33.4) Rural 262 (23.9) 723 (65.9) 11 (1.0) 12 (1.1) 89 (8.1) 498 (29.9) 437 (26.2) 342 (20.5) 237 (14.2) 154 (9.2) SemiUrban 111 (21.0) 362 (68.6) 16 (3.0) 5 (0.9) 34 (6.4) 77 (10.2) 146 (19.4) 161 (21.4) 183 (24.3) 185 (75.2) Urban 161 (21.2) 523 (68.9) 16 (2.1) 8 (1.1) 51 (6.7) 97 (9.2) 131 (12.4) 212 (20.0) 265 (25.0) 354 (33.4) 2007

1 (0.7) 3 (2.2) 1 (0.7) 8 (6.0) 57 (42.5) 38 (28.4) 26 (19.4) 1181 (16.3) 6051 (83.7) 791 (66.0) 407 (34.0) 540 (7.4) 6723 (92.6) 29.5 (23.0)

0 (0.0) 1 (3.0) 2 (6.1) 0 (0.0) 20 (60.6) 8 (24.2) 2 (6.1) 384 (12.0) 2811 (88.0) 261 (66.8) 130 (33.2) 539 (16.7) 2690 (83.3) 28.6 (21.2)

0 (0.0) 0 (0.0) 0 (0.0) 1 (4.2) 10 (41.7) 7 (29.2) 6 (25.0) 228 (12.9) 1540 (87.1) 145 (64.7) 79 (35.3) 341 (19.3) 1430 (80.7) 30.0 (21.0)

13 (7.8) 2 (1.2) 20 (12.0) 23 (13.8) 33 (19.8) 9 (5.4) 45 (26.9) 22 (13.2) 324 (20.0) 1298 (80.0) 215 (65.5) 113 (34.5) 114 (7.1) 126 (7.8) 1361 (85.0) 29.9 (22.3)

21 (20.0) 2 (1.9) 6 (5.7) 18 (17.1) 29 (27.6) 7 (6.7) 13 (12.4) 9 (8.6) 104 (14.2) 627 (85.8) 65 (63.1) 38 (36.9) 117 (16.3) 56 (17.8) 547 (76.0) 30.6 (21.1)

13 (7.8) 2 (1.2) 20 (12.0) 23 (13.8) 33 (19.8) 9 (5.4) 45 (26.9) 22 (13.2) 164 (16.0) 864 (84.0) 125 (74.4) 43 (25.6) 191 (18.7) 98 (9.6) 735 (71.8) 31.6 (22.0)

16

Table 1.2. Self-rated health status by self-reported illness, 2007 Self-rated health status Yes Very good 63 (10.7) Good 176 (29.8) Fair 234 (39.7) Poor 104 (17.6) Very poor 13 (2.2) Total 590 2 (df = 4) = 700.633, P < 0.001, correlation coefficient = - 0.412 Self-reported Illness No 1114 (40.2) 1305 (47.1) 281 (10.2) 55 (2.0) 13 (0.5) 2768

17

Figure 1.1. Mean scores for self-reported diagnosed health conditions, 2002 and 2007

18

Table 1.3. Self-rated health status by income quintile, 2007 Income Quintile Self-rated health status Poorest 20% 2.00 3.00 4.00 Very good 196 (30.2) 237 (34.0) 225 (32.4) 282 (42.4)

Wealthiest 20% 243 (36.7)

Good

287 (44.2)

320 (45.9)

326 (46.9)

268 (40.3)

284 (42.8)

Fair (moderate)

105 (16.2)

110 (15.8)

107 (15.4)

87 (13.1)

108 (16.3)

Poor

56 (8.6)

23 (3.3)

30 (4.3)

24 (3.6)

26 (3.9)

Very poor Total

6 (0.9) 650

7 (1.0) 697

7 (1.0) 695

4 (0.6) 665

2 (0.3) 663

2 (df = 16) = 54.044, P < 0.001, correlation coefficient = 0.126

19

Table 1.4. Self-reported diagnosed health condition by per capita income Income Quintile Diagnosed health condition Poorest 20% 2.00 3.00 4.00 Yes, Cold Yes, Diarrhoea Yes, Asthma Yes, Diabetes Yes, Hypertension Yes, Arthritis Yes, Unspecified No Total 2 (df = 28) = 36.161, P < 0.001 14 (11.4) 2 (1.6) 12 (9.8) 17 (13.8) 35 (28.5) 11 (8.9) 25 (20.3) 7 (5.7) 123 20 (17.5) 5 (4.4) 9 (7.9) 14 (12.3) 27 (23.7) 5 (4.4) 27 (23.7) 7 (6.1) 114 21 (15.8) 13 (11.8) 6 (4.5) 11 (8.3) 1 (0.9) 3 (2.7)

Wealthiest 20% 12 (10.3) 2 (1.7) 13 (11.1) 23 (19.7) 24 (20.5) 5 (4.3) 25 (21.4) 13 (11.1) 117

12 (9.0) 26 (23.6) 38 (28.6) 24 (21.8) 6 (4.5) 5 (4.5)

26 (19.5) 29 (26.4) 13 (9.8) 133 9 (8.2) 110

20

2
The uninsured ill in a developing nation
Empirical studies have used a piecemeal approach to the examination of health, health careseeking, uninsured people and the health status of those who are chronically ill, but no study emerged in an extensive literature search, on the developing nations, and in particular Latin America and the Caribbean, that has investigated health and health care-seeking behaviour among uninsured ill people in a single research. The current paper aims to narrow this divide by investigating health, self-reported diagnosed health conditions, and health care-seeking behaviour among uninsured ill Jamaicans, and to model factors which account for their moderate-to-very good health status as well as health care-seeking behaviour. Sixty out of every 100 uninsured ill Jamaicans were females; 43 out of every 100 were poor; 59 out of every 100 uninsured ill persons dwelled in rural areas; 1 of every 2 utilised public health care facilities, two-thirds had chronic health conditions, and 22 out of every 100 reported at least poor health. Moderate-to-very good health status was correlated with age (OR = 0.97, 95% CI = 0.95-0.98); male (OR = 0.60, 95% CI = 0.37-0.97); middle class (OR = 0.45, 95% CI = 0.21-0.95); logged income (OR = 2.87, 95% CI = 1.50-5.49); area of residence (Other Town OR = 2.33, 95^% CI = 1.19-4.54; Urban OR = 2.01, 95% CI = 1.11-3.62), and health care-seeking behaviour (OR = 0.45, 95% CI = 0.27-0.74). Sixty-one of every 100 uninsured respondents with ill health sought medical care. Medical care-seeking behaviour was significantly related to chronic illness (OR = 2.25, 95%CI = 1.31-3.88); age (OR = 1.03, 95%CI = 1.01-1.04); crowding (OR = 1.12, 1.01-1.24); income (OR = 1.00, 95% CI = 1.00-1.00); and married people (OR = 0.48, 95% CI = 0.28-0.82). Uninsured ill Jamaicans who resided in rural areas had the lowest moderate-tovery good health status, but there was no difference in health care-seeking behaviour based on the geographical location of residence. Despite the fact that there is health insurance coverage available for those who are chronically ill and elderly in Jamaica, there are still many such people who are without health insurance coverage. The task of public health specialists and policy makers is to fashion public education and interventions that will address many of the realities which emerged in this research.

Introduction
In all cultures, people desire good health and long life. Ill-health, therefore, is a challenge to the aim of healthy life expectancy, and is the rationale for investments in health options such as exercise, diet, nutrition, science and technology, medical consultation and/or health care
21

utilisation. All living organisms will experience ill-health as well as good health over their life courses; and when ill-health threatens the quality and length of life, it becomes the justification for humans willingness to rectify, address and possibly postpone illnesses. Ill-health (i.e. illness, sickness or ailment) threatens existence, productivity, development, the individual and the wider society, and because of that humans demand the best health care options. Demand for health care must be paid for by (1) a combination of health insurance coverage and out-of-pocket payment, (2) the state, (3) out-of-pocket payments or (4) relatives, associates and/or family members. Illhealth can be a burden to the individual, family, community and the nation, and it is a probability against which people and the society seek to protect themselves. All illnesses require some typology of treatment, and while this does not necessarily have to be a traditional medical practitioner, curing illness means that the individual must forego consuming something in order to restore his/her good health. Some illnesses such as the common cold may not require a trained medical practitioner to cure, but often the individual will be required to spend money on over-the-counter medications, use a home remedy or utilise non-traditional healers in the quest to restore his/her former healthy state. There are other illnesses such as diabetes mellitus, heart disease, kidney problems, hypertension, HIV/AIDS, sexually transmitted infections, and other chronic and noncommunicable diseases, which require the attention of traditional medical experts to address their cure. The traditional medical practitioners require payment in the form of cash and/or health insurance coverage. Because individuals desire to restore their health, they are expected to provide payment for health care, which for particular health conditions can be exorbitantly high.
22

It is this reality which may result in premature mortality if the state does not provide health care coverage for those who are economically challenged and/or vulnerable. The World Health Organization (WHO) [1] opined that 80% of chronic illnesses were in low and middle income countries, suggesting that illness interfaces with poverty. The WHO continued that 60% of global mortality was caused by chronic illness, and this should be understood within the context that four-fifths of chronic dysfunctions are in low-to-middle income countries [1]. It also postulated that In reality, low and middle income countries are at the centre of both old and new public health challenges [1]. Embedded in the realities outlined by the WHO are the incapacity of the poor, the association between poverty and illness, between poverty and premature mortality, poverty and human suffering, and poverty and future retardation of economic growth, and the fact that health insurance provides some cushion against this, for the individual and for society. Other studies have equally found that there is a significant statistical relationship between poverty and illness [2-4] and poverty and chronic illness, [5] which means that illness can make the vulnerable less likely to survive and the wealthy become poor. The high risk of mortality in developing countries is owing to food insecurity, low water quality and low sanitation coupled with inadequate access to material resources. Poverty makes it an insurmountable hurdle for poor people to effectively address illness unless health care services are free. Hence, those in the lower socioeconomic class will be expected to have poorer health, as they are crippled by their material deprivation and low health options. The WHO captures this aptly ... People who are already poor are the most likely to suffer financially from chronic diseases, which often deepen poverty and damage long term economic prospects. [1] Among the challenges for people living in poverty is access to health insurance coverage. Such a
23

possibility means that the burden of health care is an out-of-pocket payment that cannot be provided by the poor, and this will eliminate life in the process. Cass et al. [6] found that infant mortality in Peru for those in the poorest quintile (i.e. poorest 20%) was almost 5 times more than for those in the wealthiest quintile (i.e. wealthiest 20%). This indicates the extent of the health challenge of the poor, and the role that the lack of health insurance and income play in the demise of individuals and even their children. Another research paper revealed that life expectancy between the poorest 20% and the wealthiest 20% was 6.3 years, and this figure rose to 14.3 years for disability-free life expectancy, [7] suggesting that access and lack of access to resources explain health and healthy life expectancy in and among the social classes in a society. Grossman [8] found a positive correlation between income and health status, indicating that money makes a difference in health, health care-seeking behaviour, physical milieu and health care coverage. Smith and Kington, [9] on the other hand, went further than Grossman when they postulated that money buys health. This viewpoint is somewhat deceptive, as money provides access to good physical milieu, the best health care options, nutrition, dietary choices and health information which are not readily available to the poor, but it does not buy health. Health is not a commodity for sale, and so it cannot be purchased, but money allows for access to better health choices and by extension can change health outcomes. Those issues could be the intent of Smith and Kington, when they say that money buys health, and they further exemplify the challenges if an individual does not have access to it. Material deprivation is such that the poor will be far from concerned with health insurance coverage, proper diet and nutrition, health care choices, but more with survivability.
24

This denotes that they will be living on the margins of survivability and the decision to purchase health insurance will be the opportunity cost of food, clothing, shelter, minimal education and health options. Within the context of material and widespread health deprivation for those in the lower socioeconomic strata, the state must play a role in aiding improvements in the healthy life expectancy of those therein. It is through this avenue that public health must act in order to fulfill the aim of the state in improving the quality of life of all residents in the nation. Public health uses information from within and outside the society to improve the health and quality of peoples lives, and this requires continuous research findings. According to the WHO, In Jamaica 59% of people with chronic diseases experience financial difficulties because of their illness... Hence, poverty and illness, poverty and chronic illness, and poverty and low access to material resources are well established in research literature, but a dearth of information existed in Latin America and the Caribbean, and in particular Jamaica, on the sick and uninsured. Can we assume that they are all poor people, and use this to plan for them in a developing nation? An extensive review of the literature in developing nations, and in particular Latin America and the Caribbean, did not produce a single study that has examined health, and health care-seeking behaviour among uninsured ill people. The current paper aims to narrow this divide by investigating health, self-reported diagnosed health conditions and health care-seeking behaviour, at the same time examining who are the unhealthy and uninsured, and modelling factors which account for the moderate-to-very good health status of uninsured ill Jamaicans, in order to provide public health specialists with pertinent information that can be used to address some of the challenges within the society.

Methods and material


25

Data The current paper utilised the latest cross-sectional survey data in Jamaica to examine health, self-reported diagnosed health conditions and health care-seeking behaviour, and to model factors which account for the moderate-to-very good health status of unhealthy and uninsured Jamaicans. The Jamaica Survey of Living Conditions (JSLC) began collecting data from Jamaicans in 1988 and the latest dataset available is for 2007. The JSLC is a modification of the World Banks Living Standard Household Survey [10, 11]. This work extracted a sample of 736 respondents who indicated that they were ill and not insured, from a sample of 6,783 respondents [12]. The cross-sectional survey was conducted between May and August 2002 in the 14 parishes across Jamaica, and included 6,783 respondents of all ages. The JSLC used a stratified random probability sampling technique to draw the original sample of respondents, with a non-response rate of 26.2%. The sample was weighted to reflect the population. The design was a two-stage stratified random sampling design where there was a Primary Sampling Unit (PSU) and a selection of dwellings from the primary units. The PSU is an Enumeration District (ED), which constitutes of a minimum of 100 dwellings in rural areas and 150 in urban areas. An ED is an independent geographical unit that shares a common boundary. This means that the country was grouped into strata of equal size based on dwellings (EDs). Pursuant to the PSUs, a listing of all the dwellings was made, and this became the sampling frame from which a Master Sample of dwellings was compiled, which in turn provided the sampling frame for the labour force. One third of the 2007 Labour Force Survey (i.e. LFS) was selected for the survey. Study instrument
26

The JSLC used an administered questionnaire where respondents were asked to recall detailed information on particular activities. The questionnaire was modelled on the World Banks Living Standards Measurement Study (LSMS) household survey. The questionnaire covered demographic variables, health, education, daily expenses, non-food consumption expenditure, and other variables. Interviewers were trained to collect the data from household members. Statistical methods Descriptive statistics were used to provide socio-demographic characteristics of the sample. Chisquare analyses were used to examine the association between non-metric variables. Analysis of variance was used to test the statistical significance of a metric and non-dichotomous variable. Logistic regression analyses examined 1) the relationship between good health status and some socio-demographic, economic and biological variables; as well as 2) a correlation between medical care-seeking behaviour and some socio-demographic, economic and biological variables. The statistical package SPSS 16.0 was used for the analysis. A p-value less than 5% (2-tailed) was used to indicate statistical significance. The correlation matrix was examined in order to ascertain if autocorrelation and/or multicollinearity existed between variables. Based on Cohen and Holliday [13] correlation can be low (weak) - from 0 to 0.39; moderate 0.4-0.69, and strong 0.7-1.0. Any variable that had at least moderate (r > 0.6) was re-examined in order to address multicollinearity and/or autocorrelation between or among the independent variables [14-16]. Another approach in addressing collinearity (r > 0.6) was to independently enter variables in the model to determine which one should be retained during the final model construction. The method for retaining or excluding a variable from the model was based on its contribution to the predictive power of the
27

model and its goodness of fit [17]. Wald statistics were used to determine the magnitude (or contribution) of each statistically significant variable in comparison with the others, and the Odds Ratio (OR) for the interpreting of each significant variable. Measurement Health status is a binary measure where 1= moderate-to-very good health; 0= otherwise which is determined from Generally, how do you feel about your health? Answers to this question were analyzed on a Likert scale ranging from excellent to poor. Medical care-seeking behaviour was taken from the question Has a health care practitioner, healer, or pharmacist been visited in the last 4 weeks? with there being two options: Yes or No. Medical care-seeking behaviour therefore was coded as a binary measure where 1=Yes and 0= otherwise. Crowding is the total number of individuals in the household divided by the number of rooms (excluding kitchen, verandah and bathroom). Sex: This is a binary variable where 1= male and 0= otherwise. Age is a continuous variable which is the number of years alive since birth (using last birthday). Age group is a non-binary measure: children (aged less than 15 years); young adults (ages 15 to 30 years); other-aged adults (ages 31 to 59 years); young elderly (ages 60 to 74 years); old elderly (ages 75 to 84 years) and oldest elderly (ages 85 years and older). Social hierarchy: This variable was measured based on income quintile: The upper classes were those in the wealthy quintiles (quintiles 4 and 5); middle class was quintile 3 and the poor were those in the lower quintiles (quintiles 1 and 2). Chronic illnesses: These are ailments or diseases that are prolonged, not likely to be resolved spontaneously, and are infrequently cured.

28

Inequity denotes differences that are unnecessary and avoidable, but are also thought to be unfair and unjust, and these are adjudged based on the context of the customs operating in the society in general. Equity in health means (1) equal access to care for equal needs, (2) equal access to utilisation for equal needs, and (3) equal quality of care for all in the society. Inequalities in health mean patterns of socioeconomic disparities in health outcome which are systematic, avoidable and important within a country. Model The multivariate model used in this study is in keeping with wanting to capture the multidimensional concept of health and the health care-seeking behaviour of uninsured ill people. Utilising logistic regression on secondary cross-sectional data, the present study modelled moderate-to-very good health status and the health care-seeking behaviour of uninsured ill Jamaicans. Using a p-value of less than 0.05 to indicate statistical significance, each model reflects only those variables that are statistically significant. Health Model

Hit = f(Ait, Xi, SSit, lnYit, ARit, HSBit, it) . [1]


Health Care-seeking Behaviour Model

Hit = f(Ait, CIit, Hit, lnYit, CRit, MSit, it) . [2]


29

Where Hti is current moderate-to-very good health status of uninsured ill person i in time period t; Ai is age (in years) of person i in time period t; Xi is gender of person i; SSit is social class of person i in time period t; lnYit is logged income of person i in time period t; ARit is area of residence in time period time t; HSBit is health care-seeking behaviour in time period t; CRi is crowding in the household of person i in time period t; CIit is chronic illness of person i in time period t; MSit is marital status of person i in time period t; it is residual error of person i - in time period t.

Results
Table 2.2.1 presents information on the demographic characteristics of the sample. The sample was 736 respondents (i.e. 10.85% of the initial survey) who indicated that they were both sick and uninsured, and of which 40.5% were males. Concurringly, of the sample 95.4% had at most primary level education and 0.8% had tertiary level education. Children constituted 28.7% of the sample; young adults, 10.2%; other adults, 31.3%; young-old, 16.4%; old-old, 10.5%; and oldest-old, 3.0%. The median age was 42.0 years (range = 0 99 years). The median total annual expenditure was USD 5,689.89 (range = USD 261.56 32,780.78; US$ 1.00 = J$ 80.47 - at the time of the survey). The number of visits made to medical practitioner(s) was 1.4 1.0), while the amount of time spent in private care facilities was 3.0 2.8 compared to 5.2 5.0 for public care facilities). The mean cost of public medical care was USD 4.44 USD 16.14 compared to USD 13.64 USD 28.22 for private medical expenditure. Of those who utilised public health care facilities, 22.9% of them purchased the prescribed medication compared to 78.8% who visited private health care facilities.

30

Table 2.2 highlights information on health care-seeking behaviour, health care utilisation, self-reported illness and area of residence by social hierarchy. Based on Table 2.2, there were significant statistical associations between (1) health care-seeking behaviour and social hierarchy; (2) public health care centre utilisation and social hierarchy, and (3) private health care centre utilisation and social hierarchy. Table 2.3 highlights information on monthly food expenditure, per capita consumption, length of illness, number of visits made to health practitioners, medical expenditure and selfreported diagnosed illness by area of residence. Based on Table 2.3, there were significant statistical associations between (1) monthly food expenditure and area of residence and (2) per capita consumption and area of residence P < 0.05. However, there were no significant statistical relationships between the other variables and area of residence P > 0.05. There was a statistical association between health care-seeking behaviour and age group of respondents 2 = 11.1, P = 0.048. As uninsured ill people become older, they are more likely to seek medical care: Children, 54.8%; old-adults, 54.8%; other-age adults, 64.0; young-old, 63.3%; old-old, 73.3%; and oldest old, 66.7%. There was a statistical relationship between having chronic illness and being the household head 2 = 63.3, P < 0.0001. Almost 55% of those with chronic illnesses were household heads, compared to 22.4% who did not have chronic illness but were household heads. A significant statistical association existed between sex and having chronic illness - 2 = 4.7, P < 0.031. More females had chronic illness (69.8%) than males (61.7%). There was a significant statistical association between health status and typology of illnesses (i.e. acute and chronic conditions) - 2 = 62.3, P < 0.0001. Thirty-seven percent of
31

those with chronic illnesses reported at least poor health status compared to 12.2% of those with acute conditions. On the other hand, 61.1% of those with acute conditions reported at least good health status compared to 31.3% of those with chronic conditions. A statistical difference was found between the mean income of those in the different social hierarchies F statistic = 277.50, P < 0.0001. The mean income for those in the poorest 20% was USD 666.07 175.40 followed by the second poor, USD 1,090.68 132.14; middle class, USD 1,489.69 169.07; second wealthy, USD 2,131.55 254.49 and the wealthiest 20%, USD 4,201.39 235.26. Multivariate analysis Table 2.5 shows variables which are correlated (or not) with the moderate-to-very good health status of uninsured ill respondents. Seven variables emerged as significantly associated with moderate-to-very good health status Model 2 = 83.70, P < 0.001, -2 Log likelihood = 482.9 and they accounted for 23% of the variability in health status. The model is a good fit for the data - Hosmer and Lemeshow goodness of fit 2= 3.72, P = 0.88. Table 2.6 presents information on variables and self-reported health care seeking behaviour of uninsured respondents. Six variables emerged as significant statistical correlates of self-reported health care-seeking behaviour - Model 2 = 47.9, P < 0.001, -2 Log likelihood = 486.1. The model is a good fit for the data - Hosmer and Lemeshow goodness of fit 2= 8.11, P = 0.62.

Discussion
The current research used a sample of respondents who indicated both experiencing ill-health and having no health insurance coverage. Of the sample of respondents (i.e. n = 736), 60 out of every 100 were females, 43 out of every 100 were poor, 35 out of every 100 were in the upper
32

social class, 59 out of every 100 dwelled in rural areas, 3 out of every 100 had been injured during the last 4 weeks, 61 out of every 100 sought medical care, 50 out of every 100 utilised public health care, two-thirds reported being diagnosed with a chronic illness, 31 out of every 100 were elderly, and 29 out of every 100 were children. Those in the lower socioeconomic class were more likely to dwell in rural areas. Those in the poorest 20% were more likely to use public health centres, and the wealthiest 20% were more likely to utilise private health care centres. Fifty-four percent of those in the poorest 20% sought medical care in the last 4 weeks compared to 72% of those in the wealthiest 20%. Concurringly, of the sample, 78.4% indicated at least fair health status. Moderate-to-very good health status was explained by age, sex, social class, income, area of residence and health care-seeking behaviour. Rural residents had the least moderate-to-very good health status among uninsured ill Jamaicans. People who dwelled in Other Towns were 2.3 times more likely to indicate moderate-to-very good health compared to those in rural areas, and those in urban areas were 2.0 times more likely to claim moderate-tovery good health status. Those who indicated having a chronic illness were 37% less likely to report moderate-to-very good health. In addition, the present sample represents 70% of those who indicated having an illness in Jamaica for 2007. Statistics from the Planning Institute of Jamaica and the Statistical Institute of Jamaica [10] showed that 15.5% of Jamaicans reported ill-health in 2007. Within the context of the current findings and that of PIOJ and STATIN, it computes that 71% of those who were experiencing illness were without health insurance coverage. Given that 50% of those who claimed to be experiencing ill-health utilised the public health care system and the fact that twothirds of the illnesses were chronic conditions (3 females for every 2 males were uninsured and ill, and 6 out of every 10 uninsured ill people were of the dependent age cohort - less than 15
33

years or 60+ years), the public health care sector in Jamaica needs to recognize the impending challenges of uninsured unhealthy people. Van Agt et al. [5] found that the chronically ill were more likely to be poor, a statement with which this study concurs. In this paper, 43.2% of the chronically ill were poor (25.2% of poorest 20%) compared to 35.2% of the upper class (15.3% of the wealthiest 20%). This study went further than Van Agt et al.s work, as the chronically ill were more likely to be elderly (42.5% of the chronically ill were 60+ years), to seek more medical care, were more likely to utilise public health facilities, more likely to live in rural areas (59.1%), more likely to be household heads (54.8%) and more likely to be females (63%). Clearly the poor are highly vulnerable to chronic illness [1, 5] and material deprivation [4], which accounts for more of them not having health insurance coverage while suffering from ill-health. Hence, those who are uninsured and ill must interface with chronic health conditions as well as income deprivation. Income is well established in the health literature as being associated with health [4, 8, 9], and this explains the fact that those in the lower socioeconomic class have poorer health than those in the upper class [18, 19]. This paper found that uninsured ill people with more income are 2.9 times more likely to report moderate-to-very good health status, and they are also more likely to seek medical care. The challenge for those in the lower class is more than lower health status; it is also being deprived of the health care that they need. Statistics revealed that poverty in Jamaica is substantially a rural phenomenon (prevalence of poverty in rural areas, 15.3%; semi-urban poverty, 4.0%; urban poverty 6.2%) [10]. This study highlights that those who are ill and uninsured are likely to dwell in rural zones, explaining how financial deprivation accounts for lower ownership of health insurance coverage, the worst health being found among those in rural areas compared to city dwellers. Using per capita consumption to measure income in this
34

study, it was revealed that urban residents had 1.7 times more income than rural residents, and that semi-urban residents had 1.3 times more income than rural dwellers, suggesting that the health disparities between the geographical dwellers is explained by this income inequity. It is therefore this access to more income that accommodates the greater health status of the urban and semi-urban respondents, compared to the rural dwellers, and it highlights a real need to correct income inequality among the socioeconomic groups in the nation. A study by Stronks et al. [20] found an interrelationship between income, health and employment status, which further argues for greater health for urban and semi-urban dwellers, as rural residents are more likely to be seasonally employed, self-employed or have low-income employment. While income is related to better health status, which is also the case among uninsured ill people, concurring with the literature on a population [8, 9, 20], the great health disparity between the different social classes is more related to income than place of residence. Such a finding provides clarification for a study done by Vila et al. [21] which stated that great health disparities in the city of Milwaukee were associated with area of residence by different social hierarchy. Income has a greater influence on better health than area of residence, and it even correlates with health care-seeking behaviour among the uninsured ill, unlike area of residence. Money matters in the health of uninsured Jamaicans as well as the general populace, as it offers a better explanation for peoples choices, accounting for the greater health of those who are able to choose, than their place of residence. Lack of access to money, therefore, in any geographical locality, explains health and material deprivation. Hence, it is not the fact of being in a rural area that accounts for poor health, but material and other deprivations are greater in rural areas, a factor which provides an understanding for the massive health disparity between them and city residents.
35

Poverty is associated with premature mortality, and the current research provides some explanation for this established fact. This paper is on uninsured ill Jamaicans, and the findings highlighted that 54% of those in the poorest 20% visited a health care practitioner, 58% of the poor compared to 65% of the second wealthy and 72% of the wealthiest 20%. While the affluent class has access to material and other resources to address health concerns, the poor are not as privileged as the upper class. This research found that 70.1% of those in the poorest 20% had at least one chronic health condition, the second poor, 61.2%; the second wealthy, 72.7% and the wealthiest 20%, 68.7%, which means that non-utilisation of medical care is likely to lead to complications and possible premature mortality. The WHO had stated that 60% of global mortality is caused by chronic illness, but clearly poverty, non-treatment of chronic illnesses and cultural practices are all a part of the rationale for mortality, and not merely the condition. Although those who suffer from chronic conditions in Jamaica are able to access public health insurance which can reduce out-of-pocket payments for treatment and medication, clearly the culture prevents some people from accessing this facility. This work showed that a large percentage of uninsured ill people dwelled in rural areas, where poverty was 2.5 times more than urban poverty and 3.8% more than semi-urban poverty, arguing for the role of the culture in preventing them from accessing assistance from the state. With this preponderance of unwillingness on the part of poor and rural residents to access health insurance, accompanied by their low demand for health services compared to the wealthy, the inference is that many of them will seek health services based only on severity of illness. Chronic illnesses are such that nonmedical practitioners should not interpret when conditions are serious and warrant health care assistance. It is this culture underpinning that accounts for the premature mortality and not the poverty or illness, as those with chronic health conditions in Jamaica are able to access public
36

health care despite their reluctance to access public health insurance coverage. With not having health insurance coverage, poverty and illness are likely to become a burden to individuals and family, and when those social agents are unable to assist with the costing of medical treatment, it will then become the responsibility of the state. This paper did not examine nutrition and health, but a study by Khetarpal and Kochar [22] found a statistical relationship between nutrition and health in rural women, which offers some explanation for the great health disparities in geographical areas of residence. Another study by Foster [23] on low-income rural areas concurs with Khetarpal and Kochar [22] that nutrition accounts for health or ill-health, as the body requires particular nutrients. It can be extrapolated from the aforementioned studies, to that of the current one, that great disparities in health status among the different geographical areas in Jamaica can be explained by the nutritional intake (or lack of intake) based on where people dwell in this nation. There is a question which must be addressed in order to provide some explanation for the seemingly low nutritional intake of rural uninsured residents: Are rural residents less likely to intake the required nutrients compared to residents in other geographical areas in Jamaica? The answer is clearly yes as more of the uninsured ill Jamaicans are poor, and this means that they will be less concerned about the required nutrient intake than food consumption and mere survivability. Poverty is therefore more a factor in insurance, illness, lower health status and health careseeking behaviour than the geographical area of residence, but what about the general health status of the uninsured ill, and is it lower than that of the population of Jamaica? Almost 78 out of every 100 uninsured ill Jamaicans claimed to have at least good health status. A study by Bourne [24] found that 82 in every 100 Jamaicans reported at least good health status, which is greater than that for the uninsured ill people. Furthermore, 3.3 times more
37

Jamaicans indicated very good health compared to the uninsured ill Jamaicans. The health disparities were not only between the good and very good health status of Jamaicans and uninsured ill Jamaicans, but were also evident for poor health status. Comparatively, 4.4 times more uninsured ill Jamaicans claimed at least poor health as compared to the general population (i.e. 4.9%), and 3 times more uninsured chronically ill Jamaicans reported at least poor health status compared to those with acute health conditions. The current paper concurs with (1) Reed and Tus work [25] that uninsured chronically ill people in America reported lower health status (or worse health) and (2) Bourne and McGrowder [26] which stated that 25.3% of chronically ill Jamaicans reported at least poor health. Reed and Tu went on to state that the majority of uninsured people with chronic illnesses delay health care utilisation owing to cost, which explains an aspect of this study, that although 43.2% of the uninsured ill people were living in poverty (i.e. poorest 20% and second poor income quintile), 39% did not seek medical care. Faced with poverty, no health insurance coverage and chronic illness, uninsured ill Jamaicans are highly likely to face all kinds of life challenges such as material deprivation, dietary and nutritional deficiencies, high risk of health complications, high out-of-pocket medical bills, disruptions in family life, future vulnerabilities and premature mortality. When this burden becomes untenable for the individual, family and wider community, it will then become the responsibility of the state [27]. This justifies the need to expand public health insurance to protect the poor, the chronically ill and the vulnerable in a society [28], as chronic illness can erode the economic livelihood of an individual and therefore delay needed health care [29]. One study stated that uninsured households are one illness away from financial catastrophe [30], indicating that if a household was already in poverty this will become the burden of the state or may lead to premature mortality, as the individual will be unable to access needed health care
38

owing to his/her inability to afford medical care. This implies that poverty encapsulates powerlessness, physical weakness, illness, chronic illness, premature mortality, lack of productive assets, emotional distress, constricted freedom and future impoverishment due to the aforementioned conditions, if they are not addressed by policy makers. While impoverishment in urban areas is highly visible in the form of squalor, dilapidated edifices, zinc fencing, improper sanitation, squatting and violence, rural poverty is less easily identifiable and may be overlooked by the naked eye. Clearly, using health disparities between area of residence and the socioeconomic strata, rural poverty in Jamaica is showing signs of depleting the human capital more than urban poverty. According to Harpham and Reichenheim [31], on the disaggregating of rural and urban health indicators, the latter appear to have better health status. This study dispels the notion of appearance and goes to the reality of the health differential using self-reported health among urban, semi-urban and rural uninsured ill Jamaicans. The discipline of public health cannot only use external findings to carry out its mandate, or divorce itself from the realities which emerge from the current paper; poverty is destroying the human capabilities and resilience of the Jamaican people and more so in the case of rural uninsured ill people. Because poverty is strongly associated with illness, and illness can result in poverty [32-34], those who are presently uninsured, ill and poor are highly vulnerable to ill-health and premature mortality, which argues for an immediate health campaign to address the challenges among the socioeconomic strata and area of residence, as these were not alleviated with the introduction of the National Health Fund NHF [35]. The NHF is a statutory company which was established by the NHF Act (2003) with a Chairman and Board of Management appointed by the Minister of Health. It was established in 2003 to provide direct assistance to patients with chronic conditions, to purchase drugs and fund
39

support to private and public companies for approved projects [35]. The NHF is a social health insurance which is geared towards alleviating out-of-pocket payments for medication for those who suffer from chronic illnesses. Fourteen chronic illnesses are covered by the NHF, with respect to pharmaceutical benefits in direct assistance to ill individuals. The chronic health conditions that are covered by the NHF are hypertension, diabetes mellitus, breast cancer, prostate cancer, glaucoma, arthritis, asthma, high cholesterol, rheumatic heart disease, major depression, epilepsy, psychosis, ischemia and vascular diseases. The NHF became operational in August 2003, and has undoubtedly aided many chronically ill, non-poor and poor Jamaicans. With all the investment, the NHF has not failed to have a major coverage of chronically ill respondents using the Fund. The individuals are mostly rural residents, poor, under 60 years of age, and female. Such a reality speaks to the administrative and operational failure of the NHF to improve the lives of its intended population owing to the centralization of its operations in Kingston, which is an urban area in Jamaica. The verdict is in, that merely instituting an agency to carry out a particular task (which is to distribute benefits evenly across the socioeconomic strata, area of residence and sex) will not provide solutions to the inequalities and inequities in health between the particular groups in Jamaica. This study concurs with one in Finland [36] showing that the poor are more vulnerable to illnesses, and research conducted in the United Kingdom [37] found that those in the lower socioeconomic stratum were more likely to die prematurely than those in the upper income groups. Embedded in those findings is the fact that any equitable distribution of NHF benefits to those in the different socioeconomic strata will show further unfairness and injustices in the health outcomes which already exist, owing to income inequalities.

Conclusion
40

Two-thirds of uninsured ill Jamaicans are chronically ill. The uninsured ill are mostly within the dependent age cohort (children and elderly), they are female and are rural respondents who are generally poor people. With one half of the uninsured ill respondents utilising the public health care system, and only 2 in every 10 of them purchasing medications, there are serious future challenges for public health in Jamaica. There is an inverse relationship between the health status of uninsured ill Jamaicans and those in socioeconomic strata. The findings of this study highlight the likely challenge of the state in assisting uninsured ill Jamaicans. Despite the fact that health insurance coverage is freely accessible to those who are chronically ill in Jamaica, there are still many such people who are without health insurance coverage, and some are not even seeking medical care. Another reality which emerged from this paper is that although health care utilisation is free in Jamaica for children 18 years and younger, 45 out of every 100 of those uninsured and ill did not seek medical care, emphasizing peoples interpretation of illnesses that require medical attention, and how this retards health care demand. The task of public health specialists and policy makers, therefore, is to fashion public education and intervention programmes that will address many of the realities which emerged in this research. The great health disparity between the lower socioeconomic strata and those in the upper strata, as well as those who reside in rural areas, cannot be left to resolve itself, as clearly it has not happened in the past and the situation cannot be allowed to continue indefinitely in the future. The Way Forward The variations in health status and health care-seeking behaviour within and between the socioeconomic strata who are uninsured ill people, clearly present information that reveals public health concerns, and highlights many challenges which are still unresolved in Jamaica. The current paper did not examine the emotional distress and mortality patterns of uninsured ill
41

respondents, and this should be the subject of some future study, as it would provide needed information about these individuals. Despite the investments in health, the health sector and poverty alleviation programmes in Latin America and the Caribbean, there is still a need to study the heterogeneity in health outcome between the socioeconomic strata and area of residence, as health disparity between and within countries is still great and not in keeping with health inequality eradication in the region. Another unresolved issue stemming from the present research is how much of the cognitive dimension explains the health differential between the socioeconomic strata and the area of residence. In order to understand how to address policy intervention and health education programmes for people in Jamaica, studies need to examine the breadth and scope of cognitive dimensions in explaining health inequalities. This will allow public health technocrats to understand why 70.3% of those who were ill in Jamaica in 2007 did not have health insurance, and some of the chronically ill people, despite having access to public health insurance, did not possess such insurance, and did not seek medical care. A critical issue which needs to be addressed in the future is the structure of the National Health Fund (the NHF is accessible to, and provides public health insurance coverage for, those experiencing chronic illnesses). Barrett and Lalta [32] wrote that The National Health Fund dealt with these issues by treating the non-poor and the poor as part of the same target beneficiary. Survey data and health officials indicate that the poor suffer as much from chronic diseases as the rich, but are less likely to seek treatment, or are only able to pay for part of their prescription drugs by reducing out-ofpocket payment This study is 4 years after the operational establishment of the NHF, and new findings are coming in, which show that the NHF cannot treat different socioeconomic strata in the same way, neither can it deal equitably with those who reside in different geographical areas. The health disparities will not be addressed by merely offering equal
42

benefits to all within the context of the current findings, as these will only perpetuate health inequalities and inequities. The NHF therefore needs to be restructured in order to provide definitions based on socioeconomic class and area of residence, so as to effectively alleviate some of the challenges which emerged from this research.

Conflict of interest
The author has no conflict of interest to report.

Disclaimer
The researcher would like to note that while this study used secondary data from the Jamaica Survey of Living Conditions, 2007, none of the errors that are within this paper should be ascribed to the Planning Institute of Jamaica or the Statistical Institute of Jamaica as they are not theirs, but are instead owing to the researcher.

References
1. World Health Organization, WHO. Preventing Chronic Diseases a vital investment. Geneva: WHO; 2005: p. 9 2. WHO. Dying for change - Poor peoples experience of health and ill health. Retrieved on 29th October from http://www.who.int/hdp/publications/en/index.html. 3. Wagstaff A Poverty, equity, and health: Some research findings. In: Equity and health: Views from Pan American Sanitary Bureau. Pan American Health Organization, Occasional publication No. 8, Washington DC, US; 2001: pp.56-60. 4. Marmot M. The influence of Income on Health: Views of an Epidemiologist. Does money really matter? Or is it a marker for something else? Health Affairs. 2002; 21: 3146. 5. Van Agt HME, Stronks K, Mackenbach JP. Chronic illness and poverty in the Netherlands. Eur J of Public Health 2000; 10:197-200. 6. Casas JA, Dachs JN, Bambas A. Health disparity in Latin America and the Caribbean: The role of social and economic determinants. In: Pan American Health Organisation. Equity and health: Views from the Pan American Sanitary Bureau, Occasional Publication No. 8. Washington DC; 2001: pp. 22-49. 7. Pate E, Collado C, Solis JA. Health equity and maternal mortality. In: Equity and health: Views from Pan American Sanitary Bureau. Pan American Health Organization, Occasional publication No. 8, Washington DC, US, 2001: pp.85-98. 8. Grossman M. The demand for health - a theoretical and empirical investigation. New York: National Bureau of Economic Research, 1972. 9. Smith JP, Kington R. Demographic and Economic Correlates of Health in Old Age. Demography 1997; 34:159-70.
43

10. Planning Institute of Jamaica (PIOJ), Statistical Institute of Jamaica (STATIN). Jamaica Survey of Living Conditions, 1988-2007. Kingston: PIOJ & STATIN; 1989-2008. 11. World Bank, Development Research Group, Poverty and human resources. Jamaica Survey of Living Conditions (LSLC) 1988-2000: Basic Information. Washington DC; 2002. Retrieved on August 14, 2009, from, http://www.siteresources.worldbank.org/INTLSMS/Resources/.../binfo2000.pdf 12. Statistical Institute Of Jamaica. Jamaica Survey of Living Conditions, 2007 [Computer file]. Kingston, Jamaica: Statistical Institute of Jamaica [producer], 2007. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors], 2008. 13. Cohen L, Holliday M. Statistics for Social Sciences. London: Harper & Row; 1982. 14. Hair JF, Black B, Babin BJ, Anderson RE, Tatham RL. Multivariate data analysis, 6th ed. New Jersey: Prentice Hall; 2005. 15. Mamingi N. Theoretical and empirical exercises in econometrics. Kingston: University of the West Indies Press; 2005. 16. Cohen J, Cohen P. Applied regression/correlation analysis for the behavioral sciences, 2nd ed. New Jersey: Lawrence Erlbaum Associates; 1983. 17. Cohen J, Cohen P. Applied regression/correlation analysis for the behavioral sciences, 2nd ed. New Jersey: Lawrence Erlbaum Associates; 1983. 18. Fox J, ed: Health inequalities in European Countries. Aldershot: Gower Publishing Company Limited; 1989. 19. Illsley R, Svensson PG, eds: Health inequities in Europe. Soc Sci Med 1990; 31(special issue):223-420. 20. Stronks K, Van De Mheen H, Van Den Bos J, MacKenbach JP. The interrelationship between income, health and employment status. Int J of Epidemiol 26:592-600. 21. Vila PM, Swain GR, Baumgardner DJ, Halsmer SE, Remington PL, Cisler RA. Health disparities in Milwaukee by socioeconomic status. Wisconsin Med J 2007; 106:366-372. 22. Khetarpal A, Kochar GK. Health and well-being of rural women. The Internet Journal of Nutrition and Wellness 2007; 3. 23. Foster AD. Poverty and illness in Low-Income Rural Areas. The Am Economic Review 1994; 84:216-220. 24. Bourne PA. Dichotomising poor self-reported health status: Using secondary crosssectional survey data for Jamaica. North American J of Med Sci. 2009; 1(6): 295-302. 25. Reed MC, Tu HT. Triple jeopardy: low income, chronically ill and uninsured in America. Issue Brief Cent Stud Health Syst Change 2002 ;( 49):1-4. 26. Bourne PA, McGrowder DA. Health status of patients with self-reported chronic diseases in Jamaica. North American J of Med Sci. 2009; 1(7): 356-364. 27. Becker G. The uninsured and the politics of containment in U.S. health care. Med Anthropol. 2007; 26(4):293-8. 28. Tu HT, Reed MC. Options for expanding health insurance for people with chronic conditions. Issue Brie Cent Stud Health Syst Change 2002; 50:1-4.

44

29. Tu HT, Cohen GR. Financial and health burdens of chronic conditions grow. Track Rep 2009; 24:1-6. 30. Cook K, Dranove D, Sfekas A. Does major illness cause financial catastrophe? Health Serv Res 2009. [Epub]. 31. Harpam T, Reichenmeim M. Urbanisation and health. In: Lankinen KS, Bergstrom S, Makela PH, Peltomaa M. Health and disease in developing countries. London and Oxford: MacMillan; 1994: pp. 85-94. 32. Wagstaff A Poverty, equity, and health: Some research findings. In: Equity and health: Views from Pan American Sanitary Bureau. Pan American Health Organization, Occasional publication No. 8, Washington DC, US; 2001: pp.56-60. 33. Bourne PA. Impact of poverty, not seeking medical care, unemployment, inflation, selfreported illness, health insurance on mortality in Jamaica. North American Journal of Medical Sciences 2009; 1(3):99-109. 34. Alleyne GAO. Equity and health. In: Equity and health: Views from Pan American Sanitary Bureau. Pan American Health Organization, Occasional publication No. 8, Washington DC, US; 2001: pp.3-11. 35. Barrett R.D., Lalta S. Health financing innovations in the Caribbean: EHPO and the National Health Fund of Jamaica. New York and Washington DC: Sustainable Development Department, Technical Paper Series, Inter-American Development Bank; 2004.

36. Kalimo E, et al. Need, use and expenses of health services in Finland, 1974-76. Helsinki, Social Insurance Institution; 1983.

37. Phillmore P. Shortened lives: premature death in North Tyneside. Bristol, University of Bristol; 1989 (Briston Papers in Applied Social Studies No. 12).

45

Table 2.2.1. Demographic characteristic of sample, n=736 Characteristic Sex Male Female Marital status Married Never married Divorced Separated Widowed Social hierarchy Poorest 20% Second poor Middle Second wealthy Wealthiest 20% Area of residence Urban Semi-urban Rural Injury in last 4-weeks Yes No Self-reported diagnosed illness Acute conditions Influenza Diarrhoea Asthma Chronic conditions Diabetes mellitus Hypertension Arthritis Other Health care-seeking behaviour Yes No Health care utilization Public hospital (yes) Private hospital (yes) Public health care centres (yes) Private health care centres (yes) Other (yes) Purchased medication Yes

n 298 438 161 276 14 10 62 170 146 165 142 111 176 128 432 23 712 124 26 73 69 147 40 189 446 280 146 27 96 212 8 411

% 40.5 59.5 30.8 52.8 2.7 1.9 11.9 23.1 19.8 22.4 19.3 15.4 23.9 17.4 58.7 3.1 96.9 18.6 3.9 10.9 10.3 22.0 6.8 28.3 61.4 38.6 29.9 5.5 19.6 43.4 1.6 58.6

46

Table 2.2. Health care-seeking behaviour, health care utilization, self-reported illness and area of residence by social hierarchy
Characteristic Poorest 20% n (%) Second poor n (%) Social hierarchy Middle n (%) Second wealthy n (%) Wealthiest 20% n (%)

P 0.046

Health care-seeking behaviour Yes No Health care utilization Public hospitals Yes No Private hospitals Yes No Public health care centres Yes No Private health care centres Yes No
Self-reported diagnosed illness

90(54.2) 76(45.8)

86(59.7) 58(40.3)

98(60.1) 65(39.9)

91(65.0) 49(35.0)

81(71.7) 32(28.3) 0.337

32(37.2) 54(62.8) 5(5.7) 83(94.3) 29(33.3) 59(67.0) 28(31.5) 61(68.5)

30(35.3) 55(64.7) 5(5.9) 80(94.1) 21(24.7) 64(75.3) 35(41.2) 50(58.8)

35(35.7) 63(64.3) 3(3.1) 95(96.9) 21(21.4) 77(78.6) 49(50.0) 49(50.0)

30(33.7) 59(66.3) 6(6.7) 83(93.3) 15(17.0) 73(83.0) 52(58.4) 37(41.6)

19(23.5) 62(76.5) 0.451 8(9.9) 73(90.1) 0.016 10(12.3) 71(87.7) 0.001 48(59.3) 33(40.7) 0.200

Acute conditions Influenza Diarrhoea Asthma Chronic conditions Diabetes mellitus Hypertension Arthritis Other Area of residence Urban Semi-urban Rural Length of illness (i.e. in days) mean SD

24(15.0) 3(1.9) 21(13.1) 15(9.4) 38(23.8) 15(9.4) 44(27.5)

25(19.1) 9(6.9) 17(13.0) 15(11.5) 23(17.6) 8(6.1) 34(26.0)

34(22.7) 7(4.7) 17(11.3) 9(6.0) 37(24.7) 7(4.7) 39(26.0)

26(20.3) 3(2.3) 6(4.7) 15(11.7) 27(21.1) 6(4.7) 45(35.2)

15(15.2) 4(4.0) 12(12.1) 15(15.2) 22(22.2) 4(4.0) 27(27.3) <0.0001 59(52.2) 21(18.6) 33(29.2) 14.921.8

19(11.2) 21(14.4) 35(21.2) 42(29.6) 16(9.4) 25(17.1) 30(18.2) 36(25.4) 135(79.4) 100(68.5) 100(60.6) 64(45.1) 10.611.6 12.922.7 11.115.9 31.5116.3

0.006

47

Table 2.3. Monthly food expenditure, per capita consumption, length of illness, number of visits made to health practitioner, medical expenditure and self-reported diagnosed illness by area of residence Area of residence Characteristic P Urban Semi-urban Rural n (%) n (%) n (%) 280.71192.00 277.45162.97 237.07145.59 Monthly food expenditure 0.002 mean standard deviation 2425.231992.1 1923.621241.6 1441.301179.8 < 0.0001 Per capita consumption mean standard deviation 8 0 5 9.519.1 13.523.0 17.765.4 Length of illness in day 0.256 mean standard deviation 1.40.7 1.41.3 1.41.0 Number of visits made to 0.927 health care practitioner in last 4-weeks mean standard deviation Medical expenditure Public 3.477.07 4.7216.51 4.7818.65 0.787 mean standard deviation Private 13.5813.21 15.3815.60 13.1435.37 0.851 mean standard deviation Self-reported diagnosed 0.162 illness Acute conditions Influenza 19(12.3) 34(28.8) 17(17.9) Diarrhoea 3(1.9) 4(3.4) 19(4.8) Asthma 21(13.6) 9(7.6) 43(10.9) Chronic conditions Diabetes mellitus 16(10.4) 13(11.0) 40(10.1) Hypertension 37(24.0) 24(20.3) 86(21.7) Arthritis 10(6.5) 6(5.1) 24(6.1) Other 48(31.2) 28(23.7) 113(28.5)
Quoted in USD (USD 1.00 = Ja. $ 80.47 at the time of the survey)

48

Table 2.4. Self-reported diagnosed health conditions of uninsured ill respondents by age cohort Age cohort Characteristic Children Young adults Other-aged Young old adults n (%) n (%) n (%) n (%) Self-reported diagnosed illness Acute conditions Influenza 83(45.6) 10(15.6) 19(9.1) 6(5.1) Diarrhoea 13(7.1) 2(3.1) 6(2.9) 2(1.7) Asthma 42(23.1) 11(17.2) 13(6.2) 4(3.4) Chronic conditions Diabetes mellitus 1(0.5) 2(3.1) 32(15.3) 21(17.9) Hypertension 0(0.0) 4(6.3) 55(26.3) 41(35.0) Arthritis 0(0.0) 0(0.0) 12(5.7) 18(15.4) Other 43(23.6) 35(54.7) 72(34.4) 25(21.4)

Old-old n (%)

Oldest-old n (%)

P < 0.0001

6(8.1) 2(2.7) 2(2.7) 10(13.5) 36(48.6) 9(12.2) 9(12.2)

0(0.0) 1(4.5) 1(4.5) 3(13.6) 11(50.0) 1(4.5) 5(22.7)

49

Table 2.5: Logistic regression: Variables of moderate-to-very good health status of uninsured ill respondents Variable Age Average Medical Expenditure Male Middle Class Upper class Lower class Married Divorced, separated or widowed Never married Logged Income Urban area Other town Rural area Head household Dummy health care-seekers Chronic illness
Model 2 (12) = 83.70, P < 0.001 -2 Log likelihood = 482.96 Nagelkerke R2 = 0.23

Coefficien t -0.033 0.000 -0.511 -0.807 -1.029 0.140 -0.421

Std. Wald Error statistic 0.008 18.605 0.000 0.244 0.387 0.553 0.278 0.349 1.668 4.374 4.345 3.465 0.253 1.455

95.0% C.I. Odds ratio 0.967*** 1.00 0.60* 0.45* 0.36 1.00 1.15 0.66 1.00 0.95 - 0.98 1.00 - 1.00 0.37 - 0.97 0.21 - 0.95 0.12 - 1.06 0.67 - 1.98 0.33 - 1.30

1.053 0.696 0.844 0.218 -0.803 -0.456

0.332 0.300 0.342 0.250 0.255 0.351

10.063 5.365 6.092 0.761 9.882 1.696

2.87** 2.01* 2.33* 1.00 1.24 0.45** 0.63*

1.50 - 5.49 1.11 - 3.62 1.19 - 4.54 0.76 - 2.03 0.27 - 0.74 0.32 - 0.86

Hosmer and Lemeshow goodness of fit 2= 3.72, P = 0.88


Overall correct classification = 75.1% Correct classification of cases of self-rated moderate-to-very good health status = 93.4% Correct classification of cases of not self-rated not moderate-to-very good health status = 26.5% Reference group *** P < 0.0001, **P < 01, *P < 0.05

50

Table 2.6: Logistic regression: Variables of self-reported health care-seekers of uninsured ill respondents Variable Chronic illness Age Moderate-to-very good health Secondary education Tertiary education Primary and below education Male Crowding Logged income Length of illness Married Divorced, separated, or widowed Never married Urban area Other town Rural area
Model 2 (13) = 47.85, P < 0.001 -2 Log likelihood = 486.1 Nagelkerke R2 = 0.15

Coefficient 0.812 0.024 -0.857 1.117 1.278

Std. Error 0.277 0.008 0.281 0.762 1.222

Wald statistic Odds ratio 8.609 2.25** 9.593 9.274 2.148 1.094 1.03** 0.42** 3.06 3.59 1.00

95% CI 1.31 - 3.88 1.01 - 1.04 0.24 - 0.74 0.69 - 13.60 0.33 - 39.42

-0.358 0.114 0.000 0.000 -0.733 -0.692

0.244 0.053 0.000 0.002 0.274 0.384

2.154 4.694 4.138 0.013 7.181 3.248

0.70 1.12* 1.00* 1.00 0.48** 0.50 1.00

0.43 - 1.13 1.01 -1.24 1.00 - 1.00 1.00 - 1.00 0.28 - 0.82 0.24 - 1.06

0.171 -0.336

0.286 0.302

0.359 1.238

1.19 0.72 1.00

0.68 - 2.08 0.41 - 1.29

Hosmer and Lemeshow goodness of fit 2= 8.11, P = 0.62


Overall correct classification = 69.0% Correct classification of cases of self-reported health care-seekers = 89.4% Correct classification of cases of self-reported health care-nonseekers = 32.2% Reference group *** P < 0.0001, **P < 01, *P < 0.05

51

3
Self-rated health of the educated and uneducated classes in Jamaica
Education provides choices, opportunities, access to resources and it is associated with an increased likelihood of higher income. Does this holds true in developing nations like Jamaica, and does the educated class experience greater self-rated health status than the uneducated classes? The current paper will identify the socio-demographic correlates of self-rated health status of Jamaicans, examine the effects of these variables, explore self-rated health status and self-reported diagnosed recurring illness among the educated and uneducated classes, compute mean income among the different educational types, and determine whether a significant statistical correlation exists between the different educational cohorts. Self-rated health statuses of respondents are correlated with age, income, crowding, sex, marital status, area of residence, and self-reported illness (es) 2= 1,568.4, P < 0.001. Respondents with tertiary level educations were most likely to be classified in the wealthiest 20% (53.4%) and there was no significant statistical difference between their health status and the lower educated classes. There is a need for a public health care campaign that is specifically geared towards the educated classes as their educational achievement is not translating itself into better health careseeking behaviour and health status than the uneducated classes.

Introduction

Health is imperative for socio-economic and political development of people, a society and a nation. It is within this context that a study of health is critical as it relates to the wider society. Traditionally, the concept of health is measured using life expectancy, mortality, and diagnosed illness. In the social sciences, researchers have used self-rated health status [1-9], and selfreported illness [10-17] to measure health. Apart from those terminologies, other synonyms such
52

as self-assessed health, self-reported health, perceived health, self assessment of health, global health status, and health status have all been used to speak about health. It follows from the aforementioned perspective that all those terms imply the same measurement of health or health status. Self-rated health status is among the subjective indexes used to measure health, and some scholars argue that they are not a good assessment of health when it comes to life expectancy, per capita income, or mortality [18-20]. The subjective/objective indexes of measuring health emerged as scholars sought to ensure that the measurement of health was a reliable and valid one. Some scholars opined that the self-assessment of ones health status was more comprehensive than objective assessment [3, 5, 21] as it included ones health and general life satisfaction. Studies have shown that subjective indexes are a good measurement for mortality [2, 22-24] and life expectancy [25]. Concurringly, a recently conducted study by Bourne [25] found that self-assessed illness was not a good measure of mortality; however, it was was very useful when it came to the subject of life expectancy in Jamaica. The subjective indexes in measuring health open themselves up to systematic and unsystematic biases [26]. Peoples perception can be biased as they may inflate or deflate their status in an interview or on a self-administered instrument (i.e., questionnaire). Another aspect of bias in subjective evaluation of health is the matter of recall. It is well established in research literature that as people age, their mental faculties decline [27-32], suggesting that some people will have difficulties recalling experiences which happened in the past. Within the context of the time recollection, bias can occur in subjective indexes. Kahneman [33] devised a procedure of integrating and reducing the subjective biases when he found that instantaneous subjective
53

evaluations are more reliable than assessments of recollection of experiences. Contrary to Kahmemans work, Bournes [25] results show that self-assessed health for a 4-week period is a good measure of life expectancy (objective index). In spite of the fact that subjective indexes are a good measure of objective health, the former still contains biases, which Diener [34] opines still have valid variance. It is well established in health research that there is a correlation between or among different socio-demographic, psychological and economic variables [4, 6-17, 20] and self-rated health status. The correlates include education, marital status, area of residence, education, income, psychological conditions (i.e., positive and negative psychological affective conditions), and other variables. Freedman & Martin [35], using data from 1984 and 1993s panel survey of Income and Program Participation, noted that there was an association between educational level and physical functioning of people over 65 years. Another study by Koo, Rie & Park [36], using multivariate regression, concluded that education was a predictor of increased subjective wellbeing (t [2523] = 7.83, P<0.001], which means that education was more than associated with health. Concomitantly, another research found that the number of years of school (i.e., the Quantity Theory) was a crucial predictor of health status of an individual [37] which indicates that tertiary level graduates are more likely to be healthier than non-tertiary level educated people. While education provides choices, opportunities, access to resources and is associated with increased likelihood of achieving a higher income, does it hold true in developing nations like Jamaica that the educated class has greater self-rated health status than the uneducated classes? A paucity of information (research literature) exists in Jamaica on the educated and
54

uneducated classes and their self-rated health status, self-reported illness(es), the areas in which the educated and uneducated classes reside, health care-seeking behaviour among the different educational classes and the self-rated health status of Jamaicans and its correlates. The current paper is important, as it uses a statistical technique which accommodates all items in self-rated health status categories as opposed to dichotomising self-rated health. Dichotomising self-rated health status in good and poor health means that some of the original information will be lost; and this explains why some researchers argue for the maintenance of the Likert nature of the measuring tool over dichotomisation [38-40]. Secondly, the study is significant as it included more variables: (1) educational levels and area of residence, (2) educational levels and health care-seeking behaviour, (3) health insurance coverage and educational levels, (4) self-reported illness(es) and educational levels, (5) social standing and educational levels. The objectives of the current paper therefore are to (1) identify the sociodemographic and economic correlates of self-rated health status of Jamaicans, (2) examine the effects of these variables, (3) explore self-rated health status and self-reported diagnosed recurring illness among the educated and uneducated classes, (4) calculate the mean age of respondents in the different educational categories, (5) compute mean income among the different educational types, and (6) determine whether a significant statistical correlation exists between the different educational cohorts.

Materials and methods


Data

55

A joint survey on the living conditions of Jamaicans was conducted between May and August of 2007 by the Planning Institute of Jamaica (PIOJ) and the Statistical Institute of Jamaica (STATIN) [41]. The survey is called the Jamaica Survey of Living Conditions (JSLC) which began in 1988 and is now conducted annually. The JSLC is a modification of the World Banks Living Standards Measurement Study (LSMS) which is a household survey [42]. The current paper used the JSLCs data set for 2007 in order to carry out the analyses of the data [43]. It had a sample size of 6,783 respondents, with a non-response rate of 26.2%. The JSLC is a cross-sectional survey which used stratified random sampling techniques to draw the sample. It is a national probability survey, and data was collected across the 14 parishes of the island. The design for the JSLC was a two-stage stratified random sampling design where there was a Primary Sampling Unit (PSU) and a selection of dwellings from the primary units. The PSU is an Enumeration District (ED), which constitutes a minimum of 100 residences in rural areas and 150 in urban areas. An ED is an independent geographic unit that shares a common boundary. This means that the country was grouped into strata of equal size based on dwellings (EDs). Based on the PSUs, a listing of all the dwellings was made, and this became the sampling frame from which a Master Sample of dwellings was compiled. This, in turn, provided the sampling frame for the labour force. One third of the Labour Force Survey (i.e. LFS) was selected for the JSLC. The sample was weighted to reflect the population of the nation. Instrument A self-administered instrument (i.e., questionnaire) was used to collect the data from respondents. The questionnaire covers socio-demographic variables such as education, age, and
56

consumption, as well as other variables like social security, self-rated health status, self-reported health conditions, medical care, inventory of durable goods, living arrangements, immunisation of children 059 months, and other issues. Many survey teams were sent to each parish according to the sample size. The teams consisted of trained supervisors and field workers from the Statistical Institute of Jamaica. Statistical Analyses The Statistical Packages for the Social Sciences SPSS-PC for Windows version 16.0 (SPSS Inc; Chicago, IL, USA) was used to store, retrieve and analyze the data. Descriptive statistics such as median, mean, percentages, and standard deviation were used to provide background information on the sample. Cross tabulations were used to examine non-metric dependent and independent variables. Analysis of variance was used to evaluate a metric and a nondichotomous variable. Ordinal logistic regression was used to determine socio-demographic, economic and biological correlates of health status of Jamaicans, and identify whether the educated have a greater self-rated health status than uneducated respondents. A 95% confidence interval was used to examine whether a variable is statistically significant or not. There was no selection criterion used for the current paper. On the other hand, for the model, the selection criteria were based on 1) the literature; 2) low correlations, and 3) nonresponse rate. The correlation matrix was examined in order to ascertain if autocorrelation and/or multicollinearity existed between variables. Based on Cohen and Holliday [44] and Cohen and Cohen [45], low (weak) correlation ranges from 0.0 to 0.39, moderate 0.4-0.69, and strong 0.7-1.0. This was used to exclude (or allow) a variable in the model. Any correlation that had at least a moderate value was excluded from the model in order to reduce multicollinearity and/or
57

autocorrelation between or among the independent variables [46-51]. Another approach in addressing and/or reducing autocorrelation was to include in the model all variables that were identified from the literature review with the exception of those where the percentage of missing cases were in excess of 30%. The current paper used the ordinal nature of the dependent variable (self-rated health status or self-rated health) which denotes that none of the original data will be lost as is the case in dichotomising self-rated health. Ordered regression model is written as:

, s = 1, k,

(1)

Where x is the vector of covariates with coefficient to be estimated, k is the number of cut-points for the dependent variable, and s, l stand for the intercepts in the regression models. Anderson [52] opined that 1=1 and k, and that other constraints are possible. In the current paper, the researcher set 1=1 and 0= 1< 2 < < k =1 to correspond to the levels from very good to very poor, and other levels of health are relative to very good. Based on Andersons arguments, the monotone increase of s are dealt with by varying the sign for . Within this context, a positive estimation of coefficient denotes that those with this characteristic would be negatively associated with good health status and those without would positively associated with good health status (or self-rated health status). Simply put, positive estimation of coefficients means poor health and negative estimation of coefficients denotes better self-reported health status. Measurement of variables Dependent variable
58

Self-rated health status (i.e., self-rated health) was derived from the question, Generally, how is your health? with the options being very good, good, fair (or moderate), poor, or very poor. The ordinal nature of this variable was used as was the case in the literature [38-40]. Independent variables Information on self-reported illness was derived from the question, Have you had any illnesses other than injury? The examples given include cold, diarrhoea, asthma attack, hypertension, arthritis, diabetes mellitus or other illness. A further question about illness asked, (Have you been ill) In the past four weeks? The options were yes and no. This variable was re-coded as binary value, 1 = yes and 0 = otherwise. Information about self-reported diagnosed recurring illness was derived from the question, Is this a diagnosed recurring illness? The options were: (1) yes, cold; (2) yes, diarrhoea; (3) yes, asthma; (4) yes, diabetes mellitus; (5) yes, hypertension; (6) yes, arthritis; (7) yes, other; (8) no. Information on medical care-seeking behaviour was taken from the question, Has a health care practitioner, healer, or pharmacist been visited in the last 4 weeks? The options were yes or no. Medical care-seeking behaviour therefore was coded as a binary measure where 1 = yes and 0 = otherwise. The term crowding refers to the average number of person(s) per room excluding the kitchen, bathroom, and veranda (i.e., total number of people in household divided by the total number of rooms excluding kitchen, bathroom and veranda). Total annual expenditure was used to measure income.

59

Income quintile was used to measure social standing. The income quintiles ranged from poorest 20% to wealthiest 20%.

Results
Demographic characteristic of sample and bivariate analyses The sample was 6,783 respondents: 48.7% males and 51.3% females. Eighty-two percent of respondents rated their health status as at least good compared to 4.9% who rated it as poor. Fifteen percent of respondents reported some form of illness within the last 4 weeks. Of those who recorded an ailment, 89% reported that the dysfunction was a diagnosed recurring one. The most frequently recurring illness was unspecified conditions (23.4%) followed by hypertension (20.6%), cold (14.9%), diabetes mellitus (12.3%), and others (Table 3.3.1). The median age of the sample was 29.9 years (range = 99 years). The median annual income was US $7,050.66 (rate in 2007: 1US$ = Ja$80.47; range = US $4,406.20), and median crowding was 4.0 persons per room (range = 16 persons). A cross-tabulation between educational level and area of residence revealed a significant statistical correlation 2(df = 40 = 78.02, P < 0.001 (Table 3.3.2). Based on Table 3.3.2, 0.8% of rural respondents had tertiary level education and 5.4 times more urban residents had tertiary level education compared to rural respondents. No significant statistical correlation existed between educational level and sex of respondents 2 (df = 2) = 5.61, P > 0.05 (Table 3.3). Similarly, no significant statistical association was found between purchased prescribed medication and educational levels of respondents - 2 (df = 10) = 11.9, P > 0.05.

60

A significant statistical difference was found between mean age of respondents who are at different educational levels F statistic [2, 6589] = 214.64, P < 0.001. The mean age of respondents with primary level of education and below was 32.0 years (SD = 22.6, 95% CI = 31.4-32.6) compared to 14.6 years (SD = 1.7, 95% CI = 14.5-14.8) for those with secondary education level and 26.4 years (SD = 10.6, 95% CI = 24.6-28.2) for those with tertiary education level. A cross-tabulation between self-reported illness and educational level revealed a significant statistical association - 2 (df = 2) = 61.33, P < 0.001. Respondents with primary education level and below recorded the greatest percent of people with illness(es) (16.2%) followed in descending order by tertiary level (9.2%) and secondary level respondents (5.4%). The statistical correlation was a weak one correlation coefficient = 0.10. A significant statistical correlation existed between self-reported diagnosed recurring illness and educational level 2 (df = 14) = 42.56, P < 0.001 (Table 3.4). Respondents with secondary level education (37.5%) had the highest percent of unspecified health conditions followed in descending order by tertiary (33.3%) and primary level respondents (22.7%). Hypertension was substantially a phenomenon occurring among those with primary education level and below: 21.6%, compared to 8.3% of tertiary level individuals. Similarly, diabetes mellitus (12.8%) was more prevalent among primary level respondents compared to 5.0% of secondary level respondents. On the other hand, asthma was the greatest among tertiary level respondents (33.3%) compared to secondary level (22.5%) and primary level respondents (8.7%). Respondents with tertiary level education were most likely to be classified in the wealthiest 20% (53.4%) compared to those with secondary education who were more likely to be
61

in the middle class and those with primary level education were either in the poorest 20% (20.3%) or in the wealthiest 20% (20.3) (Table 3.4) 2 (df = 8) = 124.53, P < 0.001. Of the 20.2% of respondents who had health insurance coverage, tertiary level people were more likely to have private coverage (35.9%) followed by primary or below (12.0%) and secondary level individuals (11.6%) 2 (df = 4) = 76.95, P < 0.001 (Table 3.4). Concurringly, a significant statistical difference existed between the mean age among the different educational levels in which respondents were categorised (Table 3.4) F statistic [2, 6589] = 214.6, P < 0.001: mean age for those with at most primary level education was 32.0 years (SD = 22.6) compared to a mean age of 26.4 years (SD = 10.6) for those with tertiary level education. When educational level of respondents was disaggregated into no formal, basic, and primary to tertiary, the mean age of respondents with no formal education was 42.7 years (SD = 18.0), 2.7 years (SD = 1.9) for basic school level respondents, and 9.0 years (SD = 2.2) for those who have primary level education F statistic [4,6587] = 2207.9, P < 0.001 Multivariate analysis Self-rated health statuses of respondents are correlated with (1) age, (2) income, (3) crowding, (4) sex, (5) marital status, (6) area of residence, and (7) self-reported illness(es) 2= 1,568.4, P < 0.001; and that the data is a good fit for the model LL = 9,218.0. The 7 socio-demographic and economic correlates accounted for 33% of the variability in self-rated health status (Table 3.5). Based on the Table 3.5, the older the respondents get, the more likely they are to rate their health status as poor and this was the same for crowding and for those who report an illness (health condition). Urban residents are more likely to report poor self-rated health status than rural residents. However, there was no statistical difference between self-rated health status for rural and semi-urban residents. Married people are more likely to report better self-rated health
62

status than widowed people, people with more income are more likely to report better health status, and males are more likely than females to report better health status. However, no significant statistical difference was found between self-rated health status among the educated and uneducated cohorts.

Discussion
The current paper concurs with the literature in that self-reported illness has the most influence on self-rated health status of people [8]. In a study of elderly Barbadians (ages 60+ years), Hambleton et al. [8] found that current illness accounted for 87.7% of the variance in self-rated health status. In another study on married people in Jamaica, Bourne and Francis [53] found that 73% of self-reported illnesses explains the variability in self-reported health status. Embedded in the current finding is whether self-rated health is examined on elderly or married people. Current self-reported illnesses accounted for a critical proportion of self-rated health and can be used to measure health. Within this context, self-reported illness is a good measure of self-rated health, and this has been established by other studies [10-17, 25]. A recently conducted research found that self-reported illness accounted for 54% (r-square) of the variance in life expectancy of Jamaicans [25], and this increased to 63% for males. Subjective indexes such as self-rated health and self-reported illness can be used to measure health, but the latter is a better measure and this must be taken into consideration in the interpretation of findings using this measurement. The challenges noted by some researchers in using self-rated health are: (1) bias and (2) the dichotomisation of the measure. While bias is synonymous with subjective assessment or evaluation of any construct, the validity of using the measure is high. Diener [34] noted in 1984 that there are still some valid variances, which was validated in a recent study by Bourne [25]. Health literature has long established that subjective indexes such as self-rated health, happiness,
63

and life satisfaction are good measures of health as they are more comprehensive (including social activities and relationships, psychological conditions, emotions, spirituality, life satisfaction) while still incorporating the objective component [3, 21, 34]. This is justified by studies that found strong statistical correlations between subjective health and objective indexes such as life expectancy [25] and mortality [2, 22-24]. It should be noted here that subjective indexes (e.g., self-reported illness) and mortality are lowly correlated in Jamaica [25], which suggests that health literature among regions has revealed different findings. This denotes that the wholesale use of what is obtained in one nation cannot be applied to another without understanding socio-demographic characteristics. However, Jamaica, like other nations, can use subjective indexes to assess health status of its people and by extension its entire population. The issue of the dichotomisation of self-rated health, because some of the original values will be lost, is now resolved by this study as self-rated health was dichotomised and findings were similar to those who had dichotomised the dependent variable (i.e., self-rated health status). What are the similarities and dissimilarities between the two statistical approaches in operationalising subjective health? Studies in the Caribbean found that age, marital status, crowding, sex of respondents, area of residence, income and illnesses were statistically correlated with subjective health [8, 1017, 53], which is validated by the current paper. Even some non-Caribbean studies have found the aforementioned variables to be statistically associated with subjective health [7, 9], indicating that dichotomising self-rated health status does not fundamentally change most of the sociodemographic, economic, and biological variables.

64

Examining data on married people by way of dichotomising self-rated health status, Bourne [25] found that men had a greater self-reported health status than women, and in the current paper (non-dichotomisation of self-rated health status), males had a higher health status than females. On the other hand, in Bournes work [25], he found in descending order selfreported illnesses, age, income and sex to be the only factors of self-reported good health while in the non-dichotomised study more variables accounted for health status. Nevertheless, ranking of the correlates were similar in both studies as in the current. The factors in descending order were self-reported illness, age, crowding, income, sex and the others, indicating the closeness of the statistical approaches. Married people are a component of the general populace and they have socio-demographic and economic experiences which differ from some unmarried people. The literature showed that income is strongly correlated with self-rated health. However, in Jamaica this is clearly not the case. In Jamaica, income plays a secondary role to illness and age and when self-rated health is non-dichotomised, it becomes an even weaker variable. Although income affords one particular choices (or lack thereof), the educated class in Jamaica received more income than uneducated classes, yet the former class is not healthier than the latter. This finding is contrary to the literature that showed the association between higher education and health [7-9]. Education influences social standing and income, but it does not directly influence good health status in Jamaica. Concurringly, the current work found that education is positively correlated with more health insurance coverage. However, health insurance coverage is not significantly associated with better health status. Embedded here is the fact that health insurance coverage in Jamaica is not an indicator of health care-seeking behaviour but a product that is purchased for the eventuality of the onset of illness, as it will lower out-of-pocket medical care expenditure.
65

Education provides its recipients with knowledge, access to knowledge, access to income and other empowerment, but it does not mean that the educated classes are more concerned about their health, and this can be measured using health care-seeking behaviour and knowledge about the illnesses that are affecting the individual. The current paper found that 25 out of every 100 educated Jamaicans are aware of their health condition(s), and this is greater than that for uneducated classes. Jamaicans with the least level of education were most cognizant of their ailments and sought medical care just as much as did educated Jamaicans. Education, therefore, does not denote empowerment to seek medical care, which is embedded in the culture, in particular for men. Education is still unable to break the bondages of the perceptions of society which purport that health is weakness, and that to display weakness as a man removes his masculinity. This continues to shackle Jamaicans, particularly men, who still subscribe to the traditional notion that illness is correlated to weakness and that men should not display weakness. It is this cultural perspective that bars many men from visiting health care facilities, except in cases of severe illness or if they are married [25]. Hence, mortality being greater for men is not surprising [54] as many men will die prematurely because of the fact that they are reluctant to visit health care institutions. This reluctance to seek medical care is not limited to males. In 1988, when Jamaica began collecting data on the living conditions of its people, females sought more medical care than males, but the disparity ranged between -2 to 6%. In 2007, 68% of females sought medical care compared to 63% of males, which means that higher education, which is substantially a female phenomenon in Jamaica, is not fundamentally improving the health status of females or even males. Educated Jamaicans are more likely to live in urban areas and those with primary education levels or below are more likely to live in semi-urban zones. The current findings found
66

that semi-urban respondents were more likely to have better health status, although they are more likely to have at most primary level education. In 2007, statistics revealed that 15.3% of Jamaicans in rural areas were below the poverty line compared to 4% of semi-urban and 6.2% of urban Jamaicans [41], indicating that poverty is more synonymous with rural areas, yet there is no significant statistical difference between the self-rated health status of rural and urban Jamaicans. Income makes a difference in health, as those with more means can access more and greater resources including health care, but clearly income beyond a certain amount is retarding the health status of Jamaicans. This study cannot stipulate a baseline income that people should receive in order to prevent a decline in health status. However, there is clearly a state of contentment among the poor and very poor who were equally as healthy as the wealthy. The health disparity between them and the educated showed no significant statistical difference and this emphasises that wealth does not automatically transfer itself into health. Another issue which is evident in the data is the variability in the measurement of health among the social classes, as the poorest 20% reported less illness than the wealthiest 20% [41], yet the former group still dwells in slums, inner-city neighbourhoods, and violent communities, and they have lower levels of education. Despite Dieners findings [34] that the variance is minimal, Bournes work showed a strong association between subjective health (i.e., self-reported illness) and life expectancy a correlation coefficient between 50 and 60% for a single variable is strong. However, this highlights that there are still some challenges embedded in the use of self-rated health status.

Conclusion
While the dichotomisation of self-rated health status loses some of the original data, when selfrated health is non-dichotomised, socio-demographic and biological variables accounted for 33%
67

of the explanation of the variance and this was 44% using dichotomisation for married Jamaica, suggesting dichotomisation of health status still holds some validity. Another critical finding that emerged from the current work is that education is not improving the health status of Jamaicans. However, it is correlated with better social standing and higher income. Income is significantly associated with better health status and it played a secondary role to self-reported illness and age of respondents. Education is associated with more health insurance coverage, but that health insurance coverage cannot be used to measure health care-seeking behaviour or measure better health status of Jamaicans. In summary, there is a need for a public health care campaign that is specifically geared towards the educated classes as their educational achievement is not translating itself into better health care-seeking behaviour and health status than the uneducated which suggests that societal pressures are barring Jamaicans from better health status choices.

Conflict of interest
The author has no conflict of interest to report.

Acknowledgement Disclaimer
The researcher would like to note that while this study used secondary data from the Jamaica Survey of Living Conditions, 2007, none of the errors that are within this paper should be ascribed to the Planning Institute of Jamaica or the Statistical Institute of Jamaica, but rather to the researcher.

68

References
1.George LK. Social psychology of health. In: Binstock RH, Goerge LK, editors. Handbook of aging and the social sciences, 5th ed. San Diego: Academic Press; 2001; pp. 217-238. 2. Idler EL, Benyamini Y. Self-rated health and mortality: A review of twenty-seven community studies. J of Health and Social Behavior; 1997; 38:21-37. 3. Idler EL, Hudson SV, Leventhal H. The meanings of self-ratings of health: a qualitative and quantitative approach. Research on Aging 1999; 21:458-476. 4. Bourne PA. Self-rated health and health conditions of females in Jamaica. North American Journal of Medical Sciences. In print. 5. Krause NM, Jay GM. What global self-rated health items measure? Medical Care; 1994; 32:930-942. 6. Leinsalu M. Social variation in self-rated health in Estonia: A cross-sectional study. Social Science and Med; 2002; 55:847-861. 7. Grossman M. The demand for health A theoretical and empirical investigation. New York: National Bureau of Economic Research; 1972. 8. Hambleton IR, Clarke K, Broome HL, Fraser HS, Brathwaite F, Hennis AJ. Historical and current predictors of self-reported health status among elderly persons in Barbados. Rev Pan Salud Public; 2005; 17: 342-352. 9.Smith JP, & Kington R. Demographic and Economic Correlates of Health in Old Age. Demography; 1997; 34:159-70. 10. Bourne PA. Childhood Health in Jamaica: changing patterns in health conditions of children 0-14 years. North American Journal of Medical Sciences; 2009; 1:160-168. 11. Bourne PA. A theoretical framework of good health status of Jamaicans: using econometric analysis to model good health status over the life course. North American Journal of Medical Sciences. 2009; 1: 86-95. 12.Bourne PA. Impact of poverty, not seeking medical care, unemployment, inflation, selfreported illness, health insurance on mortality in Jamaica. North American Journal of Medical Sciences 2009; 1:99-109. 13. Bourne PA. (2009). An epidemiological transition of health conditions, and health status of the old-old-to-oldest-old in Jamaica: a comparative analysis. North American Journal of Medical Sciences; 2009; 1:211-219. 14. Bourne PA. Good Health Status of Older and Oldest Elderly in Jamaica: Are there differences between rural and urban areas? Open Geriatric Medicine Journal; 2009; 2:18-27. 15. Bourne, Paul A. A Comparative Analysis of Health Status of men 60 + years and men 73 + years in Jamaica: A Multivariate Analysis. Asian Journal of Gerontology and Geriatrics. In print. 16. Bourne PA, Rhule J. Good Health Status of Rural Women in the Reproductive Ages. International Journal of Collaborative Research on Internal Medicine & Public Health, 1(5):132155. 17. Bourne PA, McGrowder DA. Rural health in Jamaica: Examining and refining the predictive factors of good health status of rural residents. Journal of Rural and Remote Health 9 (2); 2009: 1116. 18. Gaspart F. Objective measures of wellbeing and the cooperation production problem. Social Choice and Welfare; 1998; 15:95-112.
69

19. Caldwell JC. Good health for many: The ESCAP region, 1950-2000. Asia-Pacific Journal; 1999; 14:21-38. 20. Andrews FM, Withey SB. Social indicators of wellbeing: American perceptions of life quality. New York: Plenum Press; 1976. 21. Maddox GL, Douglas EB. Self-assessment health: A longitudinal study of elderly subjects. J of Health and Soc Behaviour; 1973; 14:87-93. 22. Benjamini Y, Leventhal EA, Leventhal H. Self-assessment of health: What do people know that predicts their mortality? Research on Aging; 1999; 221:477-500. 23. Idler EL, Kasl SV, Lemke JH. Self-evaluated health and mortality among the elderly in New Heaven, Connecticut, and Iowa and Washington counties, Iowa, 1982-1986. Am J of Epidemiol; 1990; 131:91-103. 24.Wolinsk FD, Johnson RJ. Perceived health status and mortality among older men and women. J of Gerentology: Social Sci; 1992; 47:S304-S312. 25. Bourne PA. The validity of using self-reported illness to measure objective health. North American Journal of Medical Sciences; 2009; 1(5):232-238. 26. Schwarz N, Strack F. Reports of subjective well-being: Judgmental processes and their methodological implications. In: Kahneman D, Diener E, Schwarz N, editors. Well-being: The Foundations of Hedonic Psychology. Russell Sage Foundation: New York; 1999; 61-84. 27. Gavrilov LA, Gavrilova NS. The reliability theory of aging and longevity. J. theor. Biol; 2001; 213:527-545. 28. Gavrilov LA, Gavrilova NS. The Biology of Lifespan: A Quantitative Approach. New York: Harwood Academic Publisher; 1991. 29. Charlesworth B. Evolution in Age-structured Populations, 2nd ed. Cambridge: Cambridge University Press; 1994. 30. Carnes BA, Olshansky JS. Evolutionary perspectives on human senescence. Population Development Review; 1993; 19: 793-806. 31. Carnes BA, Olshansky SJ, Gavrilov L A, Gavrilova NS, Grahn D. Human longevity: Nature vs. nurture Fact or fiction. Persp. Biol. Med; 1999; 42: 422-441. 32. Medawar PB. Old age and natural death. Mod Q; 1946; 2:30-49. 33. Kahneman D. Objective happiness. In: Kahneman D, Diener E, Schwartz N, editors. Wellbeing: Foundations of hedonic psychology. Russell Sage: Foundation, New York; 1999. 34. Diener E. Subjective well-being. Psychological Bulletin; 1984; 95: 54275. 35. Freedman VA, Martin LG. The role of education in explaining and forecasting trends in functional limitations among older Americans. Demography; 1999; 36:461-473. 36. Koo J, Rie J, Park K. 2004. Age and gender differences in affect and subjective wellbeing. Geriatrics and Gerontology International; 2004; 4:S268-S270. 37. Ross CE, Mirowsky J. Refining the association between education and health: The effects of quantity, credential, and selectivity. Demography; 1999; 36:445-460. 38. Mackenbach JP, van den Bos J, Joung IM, van de Mheen H, Stronks K. The determinants of excellent health: Different from the determinants of ill-health. Int J Epidemiol; 1994; 23:127381. 39. Manderbacka K, Lahelma E, Martikainen P. Examining the continuity of self-rated health. Int J Epidemiol; 1998; 27:208-13. 40. Manor O, Matthews S, Power C. Dichotomous or categorical response: Analyzing self-rated health and lifetime social class. Int J Epidemiol 2000; 29:149-57.
70

41. Planning Institute of Jamaica (PIOJ), Statistical Institute of Jamaica. Jamaica Survey of Living Conditions, 1988-2007. Kingston: PIOJ, STATIN; 1989-2008. 42. World Bank, Development Research Group, Poverty and Human Resources. 2002. Jamaica Survey of Living Conditions, 1988-2000. Basic information. Washington: The World Bank. Retrieved on September 2, 2009 from http://siteresources.worldbank.org/INTLSMS/Resources/3358986-1181743055198/38773191190214215722/binfo2000.pdf 43. Statistical Institute Of Jamaica. Jamaica Survey of Living Conditions, 2007 [Computer file]. Kingston, Jamaica: Statistical Institute Of Jamaica [producer], 2007. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors], 2008. 44. Cohen L, Holliday M. Statistics for Social Sciences. London: Harper & Row; 1982. 45. Cohen J, Cohen P. Applied regression/correlation analysis for the behavioral sciences, 2nd ed. New Jersey: Lawrence Erlbaum Associates; 1983. 46. Hair JF, Black B, Babin BJ, Anderson RE, Tatham RL. Multivariate data analysis, 6th ed. New Jersey: Prentice Hall; 2005. 47. Mamingi N. Theoretical and empirical exercises in econometrics. Kingston: University of the West Indies Press; 2005. 48. Zar JH. Biostatistical analysis, 4th ed. New Jersey: Prentice Hall; 1999. 49. Hamilton JD. Time series analysis. New Jersey: Princeton University Press; 1994. 50. Kleinbaum DG, Kupper LL, Muller KE. Applied regression analysis and other multivariable methods. Boston: PWS-Kent Publishing; 1988. 51. Koutsoyiannis A. Theory of econometrics, 2nd ed. New York: MacMillan Publishing; 1977. 52. Anderson JA. Regression and ordered categorical variables. J of the Royal Statistical Society, Series B (Methodological); 1984; 46:1-30. 53. Bourne PA, Francis C. Self-rated health status of married people in Jamaica: Why do they have better health status? Irish Medical Journal. In print. 54. STATIN. Demographic statistics, 1970-2007. Kingston; 1991-2008.

71

Table 3.3.1. Demographic characteristic of sample, n=6,783 Characteristic Sex Male Female Marital status Married Never married Divorced Separated Widowed Social standing Poorest 20% Poor Middle Wealthy Wealthiest 20% Area of residence Urban Semi-urban Rural Self-reported illness Yes No Self-reported diagnosed recurring illness Cold Diarrhoea Asthma Diabetes mellitus Hypertension Arthritis Unspecified Not reported as diagnosed Health care-seeking behaviour Yes No Self-rated health status Very good Good Moderate Poor Very poor
72

n 3303 3479 1056 3136 77 41 224 1343 1354 1351 1352 1382 2002 1458 3322 980 5609 149 27 95 123 206 56 234 109 658 347 2430 2967 848 270 50

% 48.7 51.3 23.3 69.2 1.7 0.9 4.9 19.8 20.0 19.9 19.9 20.4 29.5 21.5 49.0 14.9 85.1 14.9 2.7 9.5 12.3 20.6 5.6 23.4 10.9 65.5 34.5 37.0 45.2 12.9 4.1 0.8

Table 3.3.2. Educational level by area of residence, n = 6,592 Characteristic Area of residence Total Educational level Urban Semi-urban Rural % % % % Primary and below 84.8 89.0 88.0 87.3 Secondary 10.9 9.6 11.2 10.8 Tertiary 4.3 1.5 0.8 2.0 Total 1952 1421 3219 6592 Chi-square (df = 4) = 78.02, P < 0.001, cc = 0.11

73

Table 3.3. Education level by sex of respondents, n = 6,592 Characteristic Male % Educational level Primary and below Secondary Tertiary Total Chi-square (df = 2) = 5.61, P > 0.05 87.9 10.5 1.6 3207 Sex Female % 86.6 11.0 2.4 3385 Total % 87.3 10.8 2.0 6592

74

Table 3.4. Self-reported diagnosed recurring illness and social standing by educational level Educational Level Total Characteristic Primary or Secondary Tertiary below % % % % Self-reported diagnosed recurring illness1 Cold 15.0 17.5 0.0 14.9 Diarrhoea 2.9 0.0 0.0 2.7 Asthma 8.7 22.5 33.3 9.5 Diabetes mellitus 12.8 5.0 0.0 12.3 Hypertension 21.6 0.0 8.3 20.6 Arthritis 5.9 0.0 0.0 5.6 Unspecified condition 22.7 37.5 33.3 23.4 Not diagnosed 10.5 17.5 25.0 10.9 Total 947 40 12 999 2 Social standing (income quintile) Poorest 20% 20.3 19.7 3.8 19.9 Poor 20.0 21.7 7.6 20.0 Middle 19.4 24.5 16.0 19.9 Wealthy 19.9 20.3 19.1 19.9 Wealthiest 20% 20.3 13.7 53.4 20.2 Total 5752 709 131 6592 Health Insurance coverage3 No 79.8 83.7 57.8 79.8 Private 12.0 11.6 35.9 12.5 Public 8.1 4.6 6.3 7.7 Total 5682 689 128 6499 Age4 Mean (SD) in years 32.0 (22.6) 14.6 (1.7) 26.4 (10.6) 30.0 (21.8) Health care-seeking behaviour5 Yes 65.7 60.0 66.7 65.5 No 34.3 40.0 33.3 34.5 Total 953 40 12 1005 Income6 Mean (SD) in US$7 8,381.88 9,580.20 14,071.67 8,623.84 (6,641.28) (7,712.81) (9,31.10) (6,874.54) 1 Chi-square (df = 14) = 42.56, P < 0.001, cc=0.20 2 Chi-square (df = 8) = 124.53, P < 0.001, cc=0.14 3 Chi-square (df = 4) = 76.95, P < 0.001, cc=0.11 4 F statistic [2,6589] = 214.6, P < 0.001 5 Chi-square (df = 2) = 0.6, P > 0.05 6 F statistic [2,6589] = 52.4, P < 0.001 7 Rate in 2007:1US$= Ja$80.47

75

Table 3.5. Ordinal logistic regression: Socio-demographic and biological differentials of selfrated health status of Jamaicans Std. 95% CI Characteristic Estimate Error Wald Upper Lower P
Excellent self-rated health Good self-rated health (1) Fair self-rated health (2) Poor self-rated (3) Very poor (4) Age Income Crowding Primary or below Secondary Tertiary (=0) Sex (female=0) Married Never married Divorced Separated Widowed (=0) Poorest 20% Poor Middle Wealthy Wealthiest 20% (=0) Urban Semi-urban Rural (=0) Private insurance Public insurance Public insurance other No insurance coverage (=0) Illness 0.0 0.540 3.504 5.935 8.659 0.045 -3.79E-007 0.083 -0.187 0.042 -0.221 -0.554 -0.352 -0.469 -0.109 0.203 0.013 0.028 -0.238 0.217 0.008 -0.175 0.026 0.387 0.0 0.345 0.625 0.985 1.425 0.008 0.000 0.025 0.252 0.267 0.077 0.200 0.192 0.319 0.369 0.163 0.140 0.126 0.122 0.090 0.085 0.110 0.149 0.209 2.456 31.465 36.327 36.909 34.055 10.636 11.130 0.553 0.025 8.290 7.704 3.342 2.171 0.087 1.554 0.009 0.048 3.782 5.789 0.008 2.542 0.032 3.433 0.117 0.000 0.000 0.000 0.000 0.001 0.001 0.457 0.874 0.004 0.006 0.068 0.141 0.768 0.213 0.925 0.826 0.052 0.016 0.927 0.111 0.859 0.064 -0.135 2.279 4.005 5.865 0.030 -6.06E-007 0.034 -0.681 -0.481 -0.372 -0.945 -0.729 -1.094 -0.832 -0.116 -0.262 -0.219 -0.477 0.040 -0.159 -0.389 -0.265 -0.022 1.216 4.728 7.865 11.452 0.060 -1.51E-007 0.132 0.307 0.566 -0.071 -0.163 0.025 0.155 0.615 0.523 0.288 0.274 0.002 0.395 0.174 0.040 0.318 0.796

2.377

0.401

35.152

0.000

1.591

3.163

Nagelkerke r-square = 0.33 Chi-square = 1,568.4, P < 0.001 LL = 9,218.0 n=4,433

76

4
Health status of patients with self-reported chronic diseases in Jamaica
Paul A. Bourne, Donovan A. McGrowder

Developing countries such as Jamaica suffer increasingly from high levels of public health problems related to chronic diseases. To examine the physical health status and use a model to determine the significant predictors of poor health status of Jamaicans who reported being diagnosed with a chronic non-communicable disease. Approximately one-quarter (25.3%) of the sample reported that they had poor health status. Thirty-three percent of the sample indicated unspecified chronic diseases: 7.8% arthritis, 28.9% hypertension, 17.2% diabetes mellitus and 13.3% asthma. Asthma affected 47.2% of children and 23.2% of young adults. Significant predictors of poor health status of Jamaicans who reported being diagnosed with chronic diseases were: age of respondents, area of residence and inability to work. Majority of the respondents in the sample had good health, and adults with poor health status were more likely to report having hypertension followed by diabetes mellitus and arthritis, while asthma was the most prevalent among children. Improvement in chronic disease control and health status can be achieved with improved patient education on the importance of compliance, access to more effective medication and development of support groups among chronic disease patients.

Introduction
The rapidly increasing burden of chronic diseases is a key determinant of global public health. In 2001, chronic diseases contributed to approximately 60% of the 56.5 million total reported
77

deaths in the world and approximately 46% of the global burden of disease. The proportion of the burden of non-communicable diseases is expected to increase to 57% by 2020 [1]. In five out of the six regions of the World Health Organization (WHO), deaths caused by chronic diseases dominate the mortality statistics [2] and there is evidence that 79% of deaths attributable to chronic diseases are occurring in developing countries, such as those in the Caribbean, predominantly in middle-aged men [2]. Most Caribbean countries have experienced a health transition, with decreases in fertility and mortality rates and changing disease patterns. Leading up to the mid-1990s, the mortality pattern changed from deaths being mainly due to communicable diseases to them being mainly due to non-communicable diseases [3, 4]. More recently, these countries have additionally observed the re-emergence of old communicable diseases and the emergence of new communicable diseases, along with an increasing prominence of non-communicable diseases. Furthermore, with 15-20% and 20-25% of the adult population in English and Dutch-speaking Caribbean countries having diabetes and hypertension respectively, these non-communicable diseases account for the single largest expenditure in national drug budgets [5]. Jamaica has undergone a significant demographic transition in the last 5 decades [6, 7]. Some features of this transition include the increase in the median age of the population from 17 years to 25 years between 1970 and 2000, the doubling of the proportion of persons older than 60 years old to over 10%, and the increase in life expectancy at birth from less than 50 years in 1950 to greater than 70 years in 2000 [8]. As a result, the main causes of illness and death in Jamaica and many other Caribbean islands and regions at a similar state of development are the chronic noncommunicable diseases [9]. There is an increased prevalence of diet-related chronic non78

communicable diseases, such as cardio-vascular diseases, diabetes and obesity. Wilks, et al. [10], reporting on a survey of body mass index in an urban population, found that 30.7% of the men were overweight (7.2% were obese) and 64.7% of the women were overweight (31.5% obese). In this same study, it was found that hypertension had a prevalence of 19.1% among the males and 28.2% among the females, while the prevalence of diabetes was 8.9% and 15.3% among the males and females respectively [10]. Chronic diseases such as heart disease, cancer and diabetes negatively affect the general health status and quality of life of individuals [11], and there is an absence in the literature of studies looking at the health status of persons in the Caribbean with chronic non-communicable diseases. It is against this background that this study was undertaken. This study was designed to explore any association between chronic non-communicable disease and health status. The aim of the study was to examine the self-reported health status of Jamaicans in rural, peri-urban and urban areas of residence. A model is used to predict the social determinants of poor health status of Jamaicans who reported at least one chronic non-communicable disease.

Method
The current paper extracted a subsample of 714 people who answered the question of having sought medical care in the last 4-weeks from a larger nationally representative cross-sectional survey of 6,783 Jamaicans (Jamaica Survey of Living Conditions, 2007) [12]. The survey was drawn using stratified random sampling. This design was a two-stage stratified random sampling design where there was a Primary Sampling Unit (PSU) and a selection of dwellings from the primary units. The PSU is an Enumeration District (ED), which constitutes a minimum of 100

79

residents in rural areas and 150 in urban areas. An ED is an independent geographic unit that shares a common boundary. This study made use of the Jamaica Survey of Living Conditions (JSLC) 2007 [12], which was conducted by the Statistical Institute of Jamaica (STATIN) and the Planning Institute of Jamaica (PIOJ) between May and August 2007. The researchers chose this survey based on the fact that it is the latest survey on the national population and that that it has data on self-reported health status of Jamaicans. Self-administered questionnaires were used to collect the data, and these were then stored and analyzed using SPSS for Windows 16.0 (SPSS Inc; Chicago, IL, USA). The questionnaire was modelled pn the World Banks Living Standards Measurement Study (LSMS) household survey. The questionnaire covered areas such as socio-demographic, economic and health variables. The non-response rate for the survey was 26.2%. Descriptive statistics such as mean, standard deviation (SD), frequency and percentage were used to analyze the socio-demographic characteristics of the sample. Chi-square was used to examine the association between non-metric variables, and an Analysis of Variance (ANOVA) was used to test the relationships between metric and non-dichotomous categorical variables. Logistic regression examined the relationship between the dependent variable and some predisposed independent (explanatory) variables, because the dependent variable was a binary one (selfreported health status is 1 if the respondent reported poor health status and 0 if otherwise). The results were presented using unstandardized B-coefficients, Wald statistics, odds ratio and confidence interval (95% CI). The predictive power of the model was tested using the Omnibus Test of Model and Hosmer & Lemeshow [13], which was used to examine goodness of fit of the model. The correlation matrix was examined in order to ascertain whether autocorrelation (or
80

multicollinearity) existed between variables. Based on Cohen & Holliday [14], correlation can be low (weak) from 0 to 0.39; moderate 0.4-0.69, and strong 0.7-1.0. This was used to exclude (or allow) a variable in the model. Wald statistics were used to determine the magnitude (or contribution) of each statistically significant variable in comparison with the others, and the Odds Ratio (OR) was used to interpret each significant variable. Multivariate regression framework [15] was utilized to assess the relative importance of various demographic, socio-economic characteristics, physical environment and psychological characteristics in determining the reported health status of Jamaicans; this has also been employed outside of Jamaica [16, 17]. Having identified the determinants of health status from previous studies, using logistic regression techniques, final models were built for Jamaicans as well as for each of the geographical sub-regions (rural, peri-urban and urban areas of residence) and sex of respondents using only those predictors that independently predict the outcome. A pvalue of 0.05 was used to for all tests of significance. Model The use of multivariate analysis in the study of health and subjective wellbeing (i.e., selfreported health or happiness) is well established [18] and this is equally the case in Jamaica and Barbados [19, 20]. The use of this approach is better than bivariate analyses as many variables can be tested simultaneously for their impact (if any) on a dependent variable. The current paper examined the social determinants of self-reported health status of Jamaicans (Equation 1). Equation 1 was again tested and decomposed by (i) sex of respondents and (ii) area of residence in order to ascertain those social predictors of each sub-group.

81

Ht=f(Ai, Gi,HHi, ARi, It, Ji, lnC, lnDi, EDi, MRi, Si, HIi, lnY, CRi, MCt, SAi, Ti , i)

[1]

where Ht (i.e., self-rated current health status in time t) is a function of age of respondents, Ai ; sex of individual i, Gi; household head of individual i, HHi; area of residence, ARi; current selfreported illness of individual i, It; injuries received in the last 4 weeks by individual i, Ji; logged consumption per person per household member, lnC; logged duration of time that individual i was unable to carry out normal activities, lnDi; education level of individual i, EDi; marital status of person i, MRi; social class of person i, Si; health insurance coverage of person i, HIi; logged income, lnY; crowding of individual i, CRi; medical expenditure of individual i in time period t, MCt; social assistance of individual i, SAi; length of time living in current household by individual i, Ti; and an error term (i.e., residual error). The final model that was derived from the general Equation [1] can be used to predict health status of Jamaicans (Equation [2]). Ht = f(Ai, ARi, lnDUt, i) [2]

Variables that were investigated include age, self-reported illness (diabetes mellitus, hypertension) and social class. Age group is a non-binary measure: children (under 15 years); young adults (15 to 30 years); other-aged adults (31 to 59 years); young elderly (60 to 74 years); old elderly (75 to 84 years) and oldest elderly (85 years and older).

Results
Demographic characteristics of sample The sample constituted 714 respondents (36.7% men and 63.3% women) with a mean age of 49.15 years. The majority of the sample was never married (44.7%), 13.4% were widowed, 1.7% separated, 3.1% divorced and 37.1% married. Some 25.3% of the sample reported that they had
82

poor health status, 31.9% indicated at least good and 42.8% indicated fair. Thirty-three percent of the sample indicated unspecified chronic illness: 7.8% arthritis, 28.9% hypertension, 17.2% diabetes mellitus and 13.3% asthma. Marginally more of the sample was in the upper class (41.6%), 19.7% in the middle class and 38.7% in the lower class (i.e., poor). The majority of the respondents were elderly (ages 60 years and older 41.6%) compared to 33.6% other-aged adults, 9.7% young adults and 15.1% children. Interestingly, the mean number of persons per room was 4.07 (S.D. 2.63 persons) and in rural areas it was 4.38 (S.D. 2.75 persons) compared to 3.9 persons in other town areas (S.D. 2.41) and 3.6 persons in urban areas (S.D. 2.42) F statistic [2, 711] = 6.642, p = 0.001. Table 4.4.1 revealed that there is a statistical correlation between social class, self-evaluated health status, annual income and area of residence (p < 0.001). Just over 50% of the rural residents were in the lower class (i.e., poor) compared to 26.4% of other town residents and 18.0% of urban dwellers. With regards to self-evaluated health and area of residence, most of the residents reported fair health status: urban residents (46.5%); other town residents (50.8%) and rural residents (38.4%). On the other hand, 28.6% of rural residents indicated that they had good self-evaluated health status compared to 31.7% of other town residents and 38.5% of urban dwellers. The mean annual income of rural residents was US$5,873.08 compared to US$8,218.05 for other town residents and US$10,312.41 for urban residents. Most of the rural respondents were in the lower class (52.9%), while 26.4% of the other town residents were in the lower class and 18% of the urban dwellers were in the lower class. Table 4.4.2 revealed that there is a statistical correlation between diagnosed chronic diseases and age group [2 (df = 20) = 297.701, p < 0.001, n = 714]. Asthma was primarily an illness for the
83

younger ages and primarily affects children: 47.2% of children and 23.2% of young adults (Table 4.4.2). The findings revealed that as an individual aged, he/she was more likely to report being diagnosed with hypertension: 0% of children, 8.7% of young adults, 31.7% of other-aged adults, 35.7% of young old, 50.5% of old-elderly and 48.3% of oldest-elderly. Arthritis was more likely to affect older ages than young ages: 0% of children, 1.4% of young adults, 7.1% of other-aged adults, 12.9% of young-old, 14.4% of old-elderly and 6.9% of oldest-elderly. On the other hand, the findings also revealed that as an individual aged, he/she was more likely to be aware of the typology of chronic illness that he/she has than they were at young ages (i.e., under 31 years). Interestingly, 2.8% of children had diabetes compared to 4.3% of young adults, 18.3% of otheraged adults, 28.7% of young old, 19.6% of old-elderly and 17.2% of oldest-elderly. Based on Table 4.4.3, no statistical correlation was found between diagnosed chronic disease and social class [2 (df = 8) = 13.882, p = 0.085, n = 714]. On the other hand, a statistical relationship was found between income, consumption, crowding and chronic disease (p < 0.5; Table 4.4). Furthermore, there is a similarity across the aforementioned variable as asthma was found to be associated with the most income, consumption and persons per room; and unspecified chronic disease was the second leading reported dysfunction. Diabetes mellitus was found to be the third leading reported chronic disease influencing people with more income and consumption. While hypertension was the third most reported chronic disease associated with crowding, it was the fourth most reported dysfunction associated with income and consumption expenditure. Multivariate analyses Using logistic regression analyses, of the 17 variables that were tested for this study, only 3 emerged as statistically significant predictors of poor health status of Jamaicans who reported
84

being diagnosed with chronic diseases (Table 4.5): age of respondents (OR = 1.029, 95% CI = 1.018 1. 040), area of residence (urban areas OR = 0.352, 95% CI = 0.191 0.652; other towns OR = 0.352, 95% CI = 0.173 0.744) and log duration unable to work (OR = 1.711, 95% CI = 1.280 2.271). The model (Equation 2) had statistically significant predictive power [2 (4) =59.76.149, p < 0.001; Hosmer and Lemeshow goodness of fit 2 = 9.956, p = 0.268] and correctly classified 74.4% of the sample (correctly classified 92.6% of those who were in poor health and 31.6% of those who were not in poor health). The logistic regression model can be written as: Log (probability of poor health status/probability of not reporting poor health status) = -0.704 + 0.028 (age) -1.041(urban residents) -1.041 (other towns) + 0.537 (log duration unable to work). Furthermore, the predictors accounted for 24% of the variability in poor health status (Table 4.5).

Discussion
There is an association between chronic disease and health status and the former has a significant negative impact on the physical aspects of health [21]. Self-reported health status has been widely used in censuses, surveys, and observational studies and there is evidence suggesting that self-reported health is an indicator of general health with good construct validity [22] and is a respectably powerful predictor of mortality risks [23], disability [24] and morbidity [25]. The results of this study showed that the majority of those sampled reported to be experiencing at least good or fair health, while approximately one-quarter indicated poor health. These results concur with those by other researchers from Dominica [26] and Trinidad [27].

85

The current paper revealed that hypertension was the most common chronic disease among the respondents, followed by diabetes mellitus and arthritis. Hypertension was highest among the elderly, with the old-elderly recording the most among the elderly cohorts. In a study by Sargeant et al. [28], hypertension is more common among women and the elderly in Jamaica. Studies from developed countries have reported prevalence of raised blood pressure among the elderly to vary from 60% to 80% [29]. Hypertension is one of the most important treatable causes of morbidity and mortality and accounts for a large proportion of cardiovascular diseases in the elderly in Jamaica [28]. The age- and sex-adjusted prevalence in Jamaica is 24% [30] with somewhat higher levels in women than in men. The Jamaican Healthy Lifestyle Survey Report 2000 [31] noted a prevalence of hypertension of 19.9% among males and 21.7% among females; prevalence increased with age in both rural and urban populations and in both sexes. Among persons known to be hypertensive, 42% were on treatment, and of this group, 37.7% had been able to lower and maintain their blood pressure at 140/90 or less. In the Caribbean and the USA, the higher prevalence of hypertension was associated with an increased prevalence of obesity, especially in women, and with greater intake of dietary sodium [32, 33].

Diabetes mellitus is an important cause of morbidity and mortality in Jamaica and represents a significant burden on health services. Diabetes was the second leading cause of chronic disease in this study and was most prevalent among the young-old with just under one-third reporting that they have diabetes mellitus. The prevalence of diabetes mellitus is high in Jamaica and the Caribbean and many patients have poor metabolic control [34]. In Jamaica the prevalence of diabetes among persons 25-74 years old is estimated to be 12% to 16% [35-37], but of which a third is unrecognized [36, 37]. There is also evidence that the diabetes prevalence has increased
86

[38]. In the Jamaican Healthy Lifestyle Survey Report 2000 [31], diabetes mellitus was found in 6.3% of males and 8.2% of females and there was a sharp increase with age. Awareness of diabetes mellitus among those classified as diabetic by the survey was 76.3%. Almost one-third of those classified as diabetic were not being treated, and 60% of those who reported being on medication did not have their condition under control. The average length of stay in hospitals was 8.3 days for diabetes mellitus in 2002, compared to 6.3 days for all conditions [31]. Diabetes mellitus accounts for about 10% of mortality in Jamaica [39] and is ranked fourth as the principal cause of death among Jamaicans during the period 1990 to 1994 [40]. But the impact of diabetes mellitus on mortality is under-reported since the disease may contribute to mortality from such other conditions as cerebrovascular accidents and myocardial infarctions [41]. Furthermore, there is evidence that the high prevalence of diabetes in Jamaica is due to the low rates of awareness, treatment and control among patients with diabetes mellitus [34]. In the Caribbean, there has been growing concern at the apparent increase in asthma in children and young adults. In 2001, hospital morbidity patterns and primary care data indicated that respiratory illnesses dominated the list of childhood infirmities among children 0-14 years. For children aged 0-4 years, asthma was the major condition for which patients were seen in health facilities, a condition mainly attributable to the high incidence of tobacco smoke to which these children are exposed [42]. In this study, asthma was the predominant chronic disease affecting approximately one-half of the children and almost one-quarter of young adults. Asthma is an important public health issue in Jamaica. Exercise-induced asthma has been reported to occur in 20 percent of school age children [43]. In government hospitals in Jamaica, five percent of clinic visits are asthma related and 25 percent of respiratory admissions to hospital are due to asthma [44]. Barnes and colleagues [45] studied asthmatic children in Barbados where treatment was
87

associated with use of inhalers, but no distinction between bronchodilators and corticosteroids was made [46]. Asthma is a significant cause of mortality in Jamaica, resulting in a death rate of approximately 5 per 100,000 [47]. Studies conducted over the last three decades in Third World countries have confirmed that rheumatoid arthritis occurs throughout the world. Rheumatoid arthritis is a chronic systemic inflammatory disorder that may affect many tissues and organs but particularly the joints, often progressing to destruction of the articular cartilage and ankylosis of the joints [48, 49]. Due to its physical, social and psychological burden, patients experience many difficulties in various aspects of their lives can contribute to their self-reported poor health. Rheumatoid arthritis is the third chronic illness among the respondents in the study. In India, the prevalence of rheumatoid arthritis (0.75%) is similar to that in the West [50]. The rarity of rheumatoid arthritis in rural Africa contrasts with the high prevalence of the disease in Jamaica, where over 2% of the adult population is affected [51]. In a study in Latin America, rheumatoid arthritis was the reason for seeking medical advice in 22% of rheumatology clinic patients [52]. Quality of life is significantly low in patients with rheumatoid arthritis, knee osteoarthritis and fibromyalgia syndrome, whose depression and/or anxiety scores are high [53]. Therefore, these patients should be managed using a multidisciplinary approach including psychiatric support. In this study, just over one-third of the respondents indicated an unspecified chronic illness. The unspecified chronic diseases could be other chronic non-communicable diseases such as a malignant neoplasm or a chronic communicable disease. In Jamaica, cancers accounted for 15% of non-communicable diseases and 9% of total disease burden in 1990. Cancers of the breast and cervix are the most common neoplasms in women, with rates in 1991 of 22.6 and 19.2 per 1,000
88

population members, respectively. Prostate cancer was the number one form of cancer found in men [54]. In 2002, there were 3,769 public hospital discharge diagnoses (4% of total discharge diagnoses) for malignant neoplasms with an equal gender distribution. The types of neoplasms involved for males, in order of decreasing frequency, were: trachea, bronchus, and lungs; prostate; leukemia; and non-Hodgkins lymphoma; representing 56% of all cancers. For females, the order was as follows: breast; cervix uteri; other malignant neoplasms of female genital organs; trachea, bronchus, and lungs; leukemia; and non-Hodgkins lymphoma; together representing 56% of all cancers [55]. The unspecified chronic illness may include HIV/AIDS, a communicable disease, which has become a serious public health concern in Jamaica. The national incidence of AIDS in 2000 was 352 per 1,000,000 population members [56]. In addition, the unspecified chronic disease may include depression and there is evidence to suggest that depressive disorders frequently accompany other chronic medical diseases. The 2000 Lifestyle Survey found approximately 25 % depressive symptoms in the general population [31]. Anderson et al. [57] concluded that the presence of diabetes mellitus increases the risk of depression and studies have shown that in persons diagnosed with diabetes mellitus the prevalence of depression ranges from 6.1% - 60.7% [58]. The majority of the respondents resided in rural areas and just over one-half of these were in the lower class. The study found that there was an income differential between respondents in rural compared with urban areas of residence, with those in the rural areas having mean annual income of approximately two-thirds of their urban counterparts. Diabetes mellitus was found to be the third leading reported chronic disease influencing persons with greater income and consumption. While hypertension was the third most reported chronic disease associated with crowding, it was
89

the fourth most reported dysfunction associated with income and consumption expenditure. According to Sobal and Stunkard [59], in developing societies there is a higher likelihood of obesity among men in higher socio-economic strata. These men are at increased risk of developing type 2 diabetes mellitus [60] which is increasing in the adult population. Most of the respondents in this study were female and single women constitute 45% of Jamaican heads of household [61]. In Jamaica, female-headed households are poorer than those headed by males and twice as likely to be unemployed. Male-headed households are smaller and have a per capita expenditure 10 times higher than female-headed households [62, 63]. The 1999 data from STATIN show that individuals who live in rural areas, who are in the poorest quintile, and who are males are less likely to seek health care [64].

Conclusion
The general epidemiological shift from infectious to chronic non-communicable diseases in Jamaica puts the residents at risk. The majority of the respondents in the sample had good health. Adults with poor health status were more likely to report having hypertension followed by diabetes mellitus and arthritis, while asthma was the most prevalent among children. Poor health status was more prevalent among those of lower economic status in rural areas who reported the least annual income. Predictors of poor health status of Jamaicans who reported being diagnosed with a chronic disease were: age, area of residence, and inability to work (therefore being unemployed). Given the high prevalence and poor levels of control, hypertension and diabetes mellitus remain formidable issues for public health care in Jamaica and the Caribbean. Poverty, low education and poor access to health care in rural communities intensify the inertia to the lifestyle modifications that are necessary to bring about greater levels of control. We suggest that
90

further improvement in chronic disease control can be achieved with improved patient education on the importance of compliance, access to more effective medication and development of support groups among patients with chronic disease(s).

References 1. World Health Organization (WHO). The world health report 2002: reducing risks, promoting healthy life. Geneva, World Health Organization; 2002. 2. World Health Organization. Diet, physical activity and health. Geneva: World Health Organization; 2002. 3. Figueroa JP. Health trends in Jamaica. Significant progress and a vision for the 21st century. West Indian Med J 2001; 50 (Suppl 4): 15-22. 4. Guilliford MC. Epidemiological transition in Trinidad and Tobago, West Indies 1953-1992. Int J Epidemiol 1996; 25: 357-365. 5. Caribbean Commission on Health and Development Report; 2004. 6. Forrester TE. Research into policy. Hypertension and diabetes mellitus in the Caribbean. West Indian Med J 2003; 52: 164-169. 7. Wilks R, Bennett F, Forrester T, McFarlane-Anderson N. Chronic diseases: The new epidemic. West Indian Med J 1998; 47(Suppl 4): 40-44. 8. Statistical Institute of Jamaica (STATIN). Demographic Statistics 2001. Kingston: STATIN; 2002.

91

9. Sargeant LA, Wilks RJ, Forrester TE. Chronic diseases-facing a public health challenge. West Indian Med J 2001; 50(Suppl 4): 27-31. 10. Wilks R, Bennett F, Forrester T, McFarlane-Anderson N, Anderson SG, Kaufman JS, Rotimi C, Cooper RS, Cruickshank JK. Chronic disease: the new epidemic. West Indian Medical Journal 1998; 47(suppl 4): 40. 11. Measuring Healthy Days: Population assessment of health-related quality of life. Atlanta, Ga: Centers for Disease Control and Prevention; 2000. 12. Planning Institute of Jamaica and Statistical Institute of Jamaica (PIOJ & STATIN). Jamaica Survey of Living Conditions 2007. Kingston: PIOJ, STATIN; 2008. 13. Homer D, Lemeshow S. Applied logistic regression, 2nd edn. John Wiley & Sons Inc., New York; 2000. 14. Cohen L, Holliday M. Statistics for social sciences. London, England: Harper and Row; 1982. 15. Asnani MR, Reid ME, Ali SB, Lipps G, Williams-Green P. Quality of life in patients with sickle cell disease in Jamaica: Rural-urban differences. Rural and Remote Health 2008; 8(890): 1-9. 16. Grossman M. The demand for health A theoretical and empirical investigation. New York: National Bureau of Economic Research; 1972. 17. Hambleton IR, Clarke K, Broome HL, Fraser HS, Brathwaite F, Hennis AJ. Historical and current predictors of self-reported health status among elderly persons in Barbados. Rev Pan Salud Public 2005; 17(5-6): 342-352. 18. Smith JP, Kington R. Demographic and economic correlates of health in old age. Demography 1997; 34: 159-170.
92

19. Hutchinson G, Simeon DT, Bain BC, Wyatt GE, Tucker MB, LeFranc E. Social and health determinants of wellbeing and life satisfaction in Jamaica. International Journal of Social Psychiatry 2004; 50: 43-53. 20. Bourne P. Using the biopsychosocial model to evaluate the wellbeing of the Jamaican elderly. West Indian Medical J 2007; 56(suppl 3): 39-40. 21. Groothoff JW, Grootenhuis MA, Offringa M, Gruppen MP, Korevaar JC, Heymans HS. Quality of life in adults with end-stage renal disease since childhood is only partially impaired. Nephrol Dial Transplant 2003; 18: 310-317. 22. Smith J. Measuring health and economic status of older adults in developing countries. Gerontologist 1994; 34: 491-496. 23. Idler EL, Benjamin Y. Self-rated health and mortality: A Review of twenty-seven community studies. Journal of Health and Social Behavior 1997; 38: 21-37. 24. Idler EL, Kasl S. Self-ratings of health: Do they also predict change in functional ability? Journal of Gerontology 1995; 50B(6): S344-S353. 25. Schechter S, Beatty P, Willis GB. Asking survey respondents about health status: Judgment and response issues, in N. Schwarz, D. Park, B. Knauper and S. Sudman [ed.]: Cognition, Aging, and Self-Reports. Ann Arbor, Michigan: Taylor and Francis; 1998. 26. Luteijn B. Health status of the elderly in the Marigot Health District, Dominica. West Indian Medical Journal 1996; 45(Suppl.2): 31. 27. Rawlins JM, Simeon DT, Ramdath DD, Chadee DD. The elderly in Trinidad: Health, social and economic status and issues of loneliness. West Indian Medical Journal 2008; 57: 589595.

93

28. Sargeant L, Boyne M, Bennett F, Forrester T, Cooper R, Wilks R. Impaired glucose regulation in adults in Jamaica: Who should have the oral glucose tolerance test? Pan American Journal of Public Health 2004; 16: 35-42. 29. Kalavathy MC, Thankappan KR, Sharma PS, Vasan RS. Prevalence, awareness, treatment and control of hypertension in an elderly community-based sample in Kerala, India. Natl Med J India 2000; 13: 9-15. 30. Cooper R, Rotimi C, Ataman S, McGee D, Osotimehin B, Kadir S, Muna W, Kingue H, Fraser H, Forrester T, Bennett F, Wilks R. The prevalence of hypertension in seven populations of West African origin. Am J Public Health 1997; 7: 160-168. 31. Figueroa JP, Ward E, Walters C, Ashley DE, Wilks RJ. Jamaica Healthy Lifestyle Survey Report; 2000. 32. Ravallion M. Poverty comparisons. Chur, Switzerland: Harwood Academic Publishers; 1994. 33. Sadana R, Mathers CD, Lopez AD, Murray CJL, Iburg K. Comparative analyses of more than 50 household surveys on health status. Geneva: World Health Organization; 2000. 34. Wilks R, Sargeant L, Gulliford M, Reid M, Forrester T. Management of diabetes mellitus in three settings in Jamaica. Pan Am J Public Health 2001; 9: 65-71. 35. Wilks R, Bennett F, Forrester T, McFarlane-Anderson N, Anderson SG, Kaufman JS, Rotimi C, Cooper RS, Cruickshank JK. Diabetes in the Caribbean: Results of a population survey from Spanish Town, Jamaica. British Diabetic Association. Diabetic Medicine 1999; 16: 875-883. 36. Ragoobirsingh D, Lewis-Fuller E, Morrison EY. The Jamaican Diabetes Survey. Diabetes Care 1995; 18(9): 1277-1279.

94

37. Cooper RS, Rotimi CN, Kaufman JS, Owoaje EE, Fraser H, Forrester T, Wilks R, Riste LK, Cruickshank JK. Prevalence of NIDDM among populations of the African diaspora. Diabetes Care 1997; 20(3): 343-348. 38. Florey Cdu V, McDonald H, McDonald J, Miall WE. The prevalence of diabetes in a rural population of Jamaican adults. Int J Epidemiol 1972; 1(2): 157-166. 39. Statistical Institute of Jamaica. Demographic statistics 1995. Kingston, Jamaica: Statistical Institute; 1996. 40. Pan American Health Organization. Caribbean Regional Health Study. Washington, DC: Pan American Health Organization, 1996: 21-22. 41. Alleyne SI, Cruickshank JK, Golding AL, Morrison EY. Mortality from diabetes mellitus in Jamaica. Bull Pan Am Health Organ 1989; 23(3): 306-314. 42. Prendergast K, Ashley D. Report on the Results of the Global Youth Tobacco Survey in Jamaica 2001; 2001. http:/w/wwww.cdc.gov/tobacco/global/gyts/reports/paho/2001/ Jamaica.2001Paho01.htm 43. Nichols DJ, Longsworth, FG. Prevalence of exercise-induced asthma in schoolchildren in Kingston, St Andrew and St Catherine, Jamaica. West Indian Med J 1995; 44: 16-19. 44. Ward E, Grant A. Epidemiological Profile of Selected Health Conditions and Selected Services in Jamaica. A Ten Year Review; 2002. 45. Barnes K C, Brenner R J, Helm R M, Howitt M E, Naidu R P, Roach T. The role of the house dust mite and other household pests in the incidence of allergy among Barbadian asthmatics. West Indian Med J 1992; 41 (suppl 1): 38. 46. Barnes K C, Naidu R P. Plenty children got wheeze these days: Lay knowledge, beliefs and stated behaviours related to asthma in Barbados. West Indian Med J 1993; 42(suppl 1): 37.
95

47. Scott PW, Mullings RL. Bronchial asthma deaths in Jamaica. West Indian Med J 1998; 47: 129-132. 48. Katz WA. Rheumatoid arthritis. In: Katz WA, ed. Diagnosis and Management of Rheumatic Diseases. Philadelphia: Lippincott; 1998; 380-396. 49. Ravinder N, Feldmann MM. Rheumatoid arthritis. In: Maddison PJ, Isenberg DA, Woo P, Glass DN, ed. Oxford textbook of Rheumatology. Second Edition, New York: Second Edition. New York: Oxford University Press. 1998; 1004-1027. 50. Malaviya AW; Kapoor SK, Singh RR, Kumar A, Pande I. Prevalence of Rheumatoid arthritis in the adult Indian population. Rheumatology International 1993; 13(4): 131-134. 51. Mijiyawa M. Epidemiology and Semiology of rheumatoid arthritis in third world countries. Rev Rhum Engl Ed 1995; 62(2): 121-126. 52. Kerr G, Richards J, Harris E. Rheumatic disease in minority population. Medical Clinics of North America 2005; 89(4): 829-868. 53. Ozcetin A, Ataoglus S, Kocer E, Yazici S, Yildiz O, Ataoglul A, Icmeli C. Effects of depression and anxiety on quality of life in patients with rheumatoid arthritis, knee osteoarthritis and fibromyalgia syndrome. West Indian Medical Journal 2007; 56(2): 122129. 54. Pan American Health Organization. Health in the Americas, 1998 Edition, Volume II Jamaica. Washington D.C.; 1998. 55. Jamaica, Ministry of Health. Ministry of Health Annual Report, 2002. Kingston: Ministry of Health; 2003 56. Pan American Health Organization. Jamaica Health situation analysis and trends summary. www.paho.org/English/DD/AIS/cp_388.htm
96

57. Anderson R, Freedland KE, Clouse RE, Lustman PJ. The prevalence of co-morbid depression in adults with diabetes. Diabetes Care 2001; 24: 1069-1078. 58. Friis R, Nanjundappa G. Diabetes, depression and employment status. Soc Sci Med 1986; 23: 471-475. 59. Sobal J, Stunkard AJ. Socioeconomic status and obesity: A review of the literature. Psychol Bull 1989; 105: 260-275.

60. Astrup A, Finer N. Redefining Type 2 diabetes: Diabetes or obesity dependent diabetes. Obesity Reviews 2001; 1(2): 57-59. 61. Dunn LL. Jamaica: Situation of children in prostitution: A rapid assessment. International Labor Organization: International Program on the Elimination of Child Labor, Geneva, Switzerland; 2001. 62. United Nations Nation Children Fund (UNICEF). Situation analysis on excluded children in Jamaica, 2006. Kingston: UNICEF; 2006. 63. World Bank. A review of gender issues in the Dominican Republic, Haiti and Jamaica, 2002. Report No. 21866-LAC. Washington, DC: The World Bank; 2002. 64. The Statistical Institute of Jamaica. Jamaica Survey of Living Conditions: report. The Statistical Institute of Jamaica, Kingston, Jamaica, 1999: 45-47.

97

Table 4.4.1: Socio-demographic characteristics of sample Area of residence Variable Urban Other towns Rural n (%) n (%) n (%) Sex Male 70 (35.0) 48 (39.7) 144 (36.6) Female 130 (65.0) 73 (60.3) 249 (63.4) Injury Yes 4 (2.0) 6 (5.0) 13 (3.3) No 195 (98.0) 115 (95.0) 380 (96.7) Self-reported chronic illness Asthma 33 (16.5) 11 (9.1) 51 (13.0) Diabetes 32 (16.0) 27 (22.3) 64 (16.3) Hypertension 47 (23.5) 41 (33.9) 118 (30.0) Arthritis 16 (8.0) 10 (8.3) 30 (7.6) Unspecified 72 (36.0) 32 (26.4) 130 (33.1) Social class Poor 36 (18.0) 32 (26.4) 208 (52.9) Middle 30 (15.0) 27 (22.3) 84 (21.4) Upper 134 (67.0) 62 (51.3) 101 (25.7) Self-evaluated health status Good 77 (38.5) 38 (31.7) 112 (28.6) Fair 93 (46.5) 61 (50.8) 150 (38.4) Poor 30 (15.0) 21 (17.5) 129 (33.0) Household head Yes 105 (52.5) 72 (59.5) 189 (48.1) No 95 (47.5) 49 (40.5) 204 (51.8) Marital status Married 60 (35.3) 34 (31.8) 130 (39.8) Never married 78 (45.9) 48 (44.9) 144 (44.0) Divorced 7 (4.1) 6 (5.6) 6 (1.8) Separated 4 (2.4) 4 (3.7) 2 (0.6) Widowed 21 (12.4) 15 (14.0) 45 (13.8) Educational level No formal 160 (80.0) 106 (87.6) 319 (81.2) Basic 14 (7.0) 5 (4.1) 33 (8.4) Primary/Preparatory 12 (6.0) 5 (4.1) 25 (6.4) Secondary/High 9 (4.5) 4 (3.3) 13 (3.3) Tertiary 5 (2.5) 1 (0.8) 3 (0.8) Age Mean (SD) 47.5 yrs (25.07 yrs) 53.36 yrs. (23.61) 48.7 (25.79 yr) Annual Income Mean (SD) USD10,312.41 USD8,218.05 USD5,873.08 (USD9,059.70) (USD7,653.84) (US 4,473.51) 1.4 (1.1) 1.5 (1.5) 1.4 (1.1) Number of visits to health care practitioner Mean (SD)
Annual Income is quoted in USD (US$ 1.00 = Ja. $ 80.47 at the time of the survey)

P 0.702 0.347 0.214

< 0.001

< 0.001

0.082

0.166

0.466

0.114 < 0.001 0.842

98

Table 4.4.2: Diagnosed chronic recurring illness by age group Age group Other-aged adults Young old n (%) n (%) 18 (7.5) 7(4.1)

Diagnosed chronic illness Asthma

Children n (%)

Young adults n (%)

Old elderly n (%) 2 (2.1)

Oldest elderly n (%) 1 (3.4)

Total n (%) 95 (13.3)

51 (47.2) 16 (23.2)

Diabetes mellitus Hypertension Arthritis Other (unspecified) Total 2 (df = 20) = 297.701, P < 0.001

3 (2.8) 0 (0.0) 0 (0.0)

3 (4.3) 6 (8.7) 1 (1.4)

44 (18.3) 76 (31.7) 17 (7.1) 85 (35.4) 240

49 (28.7) 61 (35.7) 22 (12.9) 32 (18.7) 171

19 (19.6) 49 (50.5) 14 (14.4) 13 (13.4) 97

5 (17.2) 14 (48.3)

123 (17.2) 206 (28.9) 56 (7.8) 234 (32.8) 714

2 (6.9) 7 (24.1) 29

54 (50.0) 43 (62.3) 108 69

99

Table 4.4.3: Diagnosed chronic illness by social class

Diagnosed chronic illness Asthma Diabetes mellitus Hypertension Arthritis Other (unspecified) Total 2 (df = 8) = 13.882, P = 0.085

Poor n (%) 42 (15.2) 41 (14.9) 82 (29.7) 25 (9.1) 86 (31.2) 276

Social class Middle class n (%) 19 (13.5) 16 (11.3) 48 (34.0) 12 (8.5) 46 (32.6) 141

Upper class n (%) 34 (11.4) 66 (22.2) 76 (25.6) 19 (6.4) 102 (34.3) 297

Total n (%) 95 (13.3) 123 (17.2) 206 (28.9) 56 (7.8) 234 (32.8) 714

100

Table 4.4: Crowding, income and annual consumption expenditure by diagnosed chronic disease 95% CI Std. Error 0.30 0.21 0.18 0.26 0.18 0.10 65321.32 39832.52 37414.43 64740.33 38681.88 20763.10 47337.01 32750.14 31364.78 47093.62 28261.42 16057.14 Lower 4.55 3.14 3.51 2.66 3.95 3.88 606099.94 489952.96 480323.85 330599.37 573083.63 563886.37 561310.85 444432.04 438796.32 309774.41 530832.07 511752.81 Upper 5.72 3.97 4.22 3.69 4.65 4.26 865493.98 647657.86 627856.71 590084.40 725505.52 645414.61 749288.33 574096.38 562474.15 498529.87 642193.25 574802.65

N Crowding Asthma Diabetes mellitus Hypertension Arthritis Other (unspecified) Total Asthma Diabetes mellitus Hypertension Arthritis Other (unspecified) Total Asthma 95 123 206 56 234 714 95 123 206 56 234 714 95

Mean 5.14 3.55 3.86 3.18 4.30 4.07 735796.96 568805.41 554090.28 460341.89 649294.58 604650.49 655299.59

Std. Deviation 2.88 2.33 2.59 1.93 2.69 2.63 636673.50 441764.01 536998.15 484472.24 591718.97 554806.11 461384.09 363216.62 450169.33 352416.40 432316.66 429059.15

Annual income*

Annual consumption expenditure*

Diabetes mellitus 123 509264.21 Hypertension 206 500635.23 Arthritis 56 404152.14 Other (unspecified) 234 586512.66 Total 714 543277.73 Crowding F statistic [4, 709] = 7.778, P < 0.001 Income F statistic [4, 709] = 3.250, P = 0.012, Annual consumption Expenditure F statistic [4, 709] = 4.472, P = 0.001

*Income and Annual Consumption Expenditure were quoted in Jamaican dollars (Ja. $80.47 = USD1.00 at the time of the survey)

101

Table 4.5: Logistic regression: Predictor of poor health status of patients who reported chronic disease Wald statistic 26.131*** 11.061** 7.582**

Predictors Age Urban areas Other towns

Std. error 0.006 0.313 0.372

Odds ratio 1.029 0.353 0.359

95.0% C.I. 1.018 - 1.040 0.191 - 0.652 0.173 - 0.744 1.289 - 2.271

Log duration unable to 0.145 13.803*** 1.711 work 2 (df = 4) =59.76.149, P < 0.001; n = 714) -2 Log likelihood = 332.325 Nagelkerke R2 =0.240 Hosmer and Lemeshow goodness of fit 2 = 9.956, P = 0.268 Reference group rural areas *P < 0.05, **P < 0.01, ***P < 0.001

102

5
The changing faces of diabetes, hypertension and arthritis in a Caribbean population
Paul A. Bourne, Samuel McDaniel, Maxwell S. Williams, Cynthia Francis, Maureen D. Kerr-Campbell, & Orville W. Beckford

Globally, chronic illnesses are the leading cause of mortality, and this is no different in developing countries, particularly in the Caribbean. Little information emerged in the literature on the changing faces of particular self-reported chronic diseases. This study examines the transitions in the demographic characteristics of those with diabetes, hypertension and arthritis, as we hypothesized that there are changing faces of those with these illnesses. The prevalence of particular chronic diseases increased from 8 per 1,000 in 2002 to 56 per 1,000 in 2007. The average annual increase in particular chronic diseases was 17.2%. Diabetes mellitus showed an exponential average annual increase of 185% compared to hypertension (+ 12.7%) and arthritis (- 3.8%). Almost 5 percent of diabetics were less than 30 years of age (2.4% less than 15 years), and 41% less than 59 years. Three percent of hypertensive respondents were 30 years and under as well as 2% of arthritics. The demographic transition in particular chronic conditions now demands that data collection on those illnesses be lowered to < 15 years. This research highlights the urgent need for a diabetes campaign that extends beyond parents to include vendors, confectionary manufacturers and government, in order to address the tsunami of chronic diseases facing the nation.

Introduction
Globally, chronic illnesses are the leading cause of mortality (60%) [1, 2], and this is no different in developing countries, particularly in the Caribbean [2-6]. Statistics indicate that 79% of all mortalities are attributable to chronic diseases, and that they are occurring in developing
103

countries such as those in the Caribbean [3]. Using data for 1989 and 1990, Holder & Lewis [7] showed that hypertension and diabetes mellitus were among the 5 leading causes of mortality in the English-speaking Caribbean and Suriname. The findings from Holder and Lewis indicated that mortality resulting from hypertension was highest in Dominica (over 90 per 100,000 of the population) and diabetes crude death rates per 100,000 of the population were the greatest in Trinidad and Tobago (over 85 per 100,000). The 20th century has brought with it massive changes in the typology of diseases, where deaths have shifted from infectious diseases such as tuberculosis, pneumonia, yellow fever, Black Death (i.e. Bubonic Plague), smallpox and diphtheria to diseases such as cancer, heart complaints and diabetes. Although diseases have moved from infectious to degenerate, chronic non-communicable illnesses have arisen and are still lingering in spite of all the advances in science, medicine and technology. Morrison [8] titled an article Diabetes and Hypertension: Twin Trouble in which he established that diabetes mellitus and hypertension have now become two problems for Jamaicans and people in the wider Caribbean. This situation was corroborated by Callender [9] and Steingo at the 6th International Diabetes and Hypertension Conference, which was held in Jamaica in March 2000. They found that there is a positive association between diabetic and hypertensive patients - 50% of individuals with diabetes had a history of hypertension [9, 10]. Prior to those scholars work, Eldemire [11] found that 34.8% of new cases of diabetes and 39.6% of hypertension were associated with senior citizens (i.e. ages 60 and over). In an article published by Caribbean Food and Nutrition Institute, the prevalence rate of diabetes mellitus affecting Jamaicans is noted to be higher than in North American and many European countries [9].

104

Chronic illnesses have been on the rise in the Caribbean. In a 1996 study conducted by Morrison and colleagues in Trinidad and Tobago [12], they noted that there is an alarming rise in the prevalence rate of diabetes mellitus (15-18%). A study in Barbados found that between 1988 and 1992 the prevalence rate of diabetes mellitus for the population was 17.5%; 12.5% in mixed population (black/white), 6.0% in white/other and 0.3% in the younger population [13]. Another research, in Europe, found that the prevalence among newly diagnosed diabetics in Europeans was 20%; African-Caribbeans, 22%; and in Pakistanis, 33% [14]. They also postulated that there is an association between poverty and diabetes. Van Agt et al. [15] went further when they found that poverty was greater among the chronically ill, with which a later study by the World Health Organization [16] concurred. The WHO [16] stated that 80% of chronic illnesses were in low and middle income countries, emphasizing the association between not only diabetes and poverty, but chronic conditions and poverty. The relationship between poverty and chronic conditions extends to premature mortality [17]. Findings from the WHO [4] showed that 60% of global mortality is caused by chronic illness, which offers an explanation of the face for those with these particular conditions. Within the context of a strong association between poverty and chronic illness, the high prevalence of diabetes mellitus, hypertension and other chronic conditions in developing countries should not be surprising [16, 18]. Yach et al. [18] further opined that the global figure for diabetes is projected to move from 171 million (2.8%) in 2000 to 366 million (6.5%) in 2030. Of this figure 298 million of these persons will be in developing countries, which reinforces the poverty-illness relationship. Chronic diseases can be likened to a tsunami [19] in developing nations [20-22], and it seems to be spiralling because of the unhealthy lifestyle of people. The tsunami of chronic illnesses in the developing countries is equally reflected in the Americas [20, 21], and particularly Jamaica. The
105

face of chronic illness in developing nations is therefore for (1) lower socioeconomic strata, (2) rural residents, (3) adults, (4) gender differences, (5) lower educational level, and (6) married people. A great deal of research exists on the management of chronic illnesses, and rightfully so, as these go to the health status and mortality of a population [23, 24]. The profiles of those with chronic diseases have never been examined in Latin America and the Caribbean, and studies outside of this region have used a piecemeal approach to the investigation of chronic conditions. Hence information is available on one or a few of the aforementioned faces of chronic illness, and some research has examined diabetes mellitus and hypertension but not arthritis. The present gap in the literature will be lowered by this study examining the faces of chronic illness from half a decade of data. Using data for 2002 and 2007, the current paper will investigate the changing faces of chronic diseases in Jamaica. The study will utilize three chronic diseases (i.e. diabetes mellitus, hypertension, and arthritis), and analyze health status, health insurance status, health care utilization, chronic illness and other sociodemographic characteristics in order to ascertain the transition occurring in the population. We hypothesized that there are changing faces of those with diabetes, hypertension and arthritis over the last half a decade (2000-2007). Materials and methods Data The current paper extracted a sample of 592 respondents from the 2002 and 2007 Jamaica Survey of Living Conditions (JSLC). Only respondents who indicated that they were diagnosed with particular chronic conditions were used for this analysis (i.e. diabetes mellitus, hypertension, and arthritis). The present subsample represents 0.8% of the 2002 national sample
106

(25,018) and 5.7% of the 2007 sample (6,783).

The JSLC is an annual and nationally

representative cross-sectional survey that collects information on consumption, education, health status, health conditions, health care utilization, health insurance coverage, non-food consumption expenditure, housing conditions, inventory of durable goods, social assistance, demographic characteristics and other issues [25]. The information is from the civilian and noninstitutionalized population of Jamaica. It is a modification of the World Banks Living Standards Measurement Study (LSMS) household survey [26]. questionnaire was used to collect the data. Overall, the response rate for the 2007 JSLC was 73.8% and 72.3% for 2002. Over 1,994 households of individuals nationwide are included in the entire database of all ages [27]. The residents of a total of 620 households were interviewed from urban areas, 439 from other towns and 935 from rural areas. This sample represents 6,783 non-institutionalized civilians living in Jamaica at the time of the survey. The JSLC used complex sampling design, and it is also weighted to reflect the population of Jamaica. Statistical analysis Statistical analyses were performed using the Statistical Packages for the Social Sciences for Windows 16.0 (SPSS Inc; Chicago, IL, USA). Descriptive statistics such as mean, standard deviation, frequency and percentage were used to analyze the socio-demographic characteristics of the sample. Chi-square was used to examine the association between non-metric variables, and an Analysis of Variance was used to test the equality of means among non-dichotomous categorical variables. Means and frequency distribution were considered significant at P < 0.05 using chi-square, independent sample t-test, and analysis of variance f test.
107

A self-administered

Measures Table 5.1 presents the operational definitions of some of the variables used in this study. Results Health care utilization, health insurance status, particular chronic illness (i.e. diabetes mellitus, hypertension and arthritis), and sociodemographic characteristics are presented in Table 5.2. The findings in Table 5.2 showed that the average annual increase in the particular chronic illness was 17.2% between 2002 and 2007. Arthritis showed an average annual reduction of 3.8%, hypertension, + 12.7% and diabetes mellitus, + 185.0%. Furthermore, the average annual increase in health care utilization (visits to health care institutions) was 11.9% (public hospital, + 8.2%; private hospital, + 10.7%; public health care centre, + 8.4%; private health care centre, + 17.1%). On average the annual increase in health insurance coverage was + 148%; while the health care utilization (health seekers) increased by 11.7%. The particular chronic illnesses have shifted mostly from urban (67.6%) to rural residents (55.1%). This shift could be attributed to cultural factors affecting how and what individuals eat in rural versus urban areas. The sedentary lifestyles of urban areas also added to the overall dramatic increase in chronic illnesses. Table 5.3 presents information on self-reported diagnosed particular chronic illness by sex of respondents for 2002 and 2007. On average, the annual increase in particular chronic illness in males was 19.0% compared to 16.5% in females. Diabetes mellitus showed the highest annual percentage increase (males 186.7% and females 184.4%), while arthritis fell in females (average annual 7.9%) compared to an increase in males (average annual 10.0%). Hypertension increased more in females (average annual 14.0%) compared to 9.7% in males. This could be

108

attributed to the increasing absorption of females into the upper echelons of management in stressful occupations such as banking and finance, law, and the police force. Table 5.4 examines information on health coverage, health status, health care utilization and some sociodemographic characteristics by self-reported diagnosed particular chronic illnesses for 2002 and 2007. Based on Table 5.4, although particular chronic illnesses have decreased in rural respondents, rural dwellers continue to be the face of chronic conditions as well as married, primary, uninsured, private health centres and those in the lower class. The average annual increase in particular chronic illnesses increased by 22.9% for those in the lower strata compared to 11.0% for those in the middle class and 16.0% for those in the wealthy socioeconomic strata. However, the greatest increase occurred in diabetics belonging to the upper class (average annual + 200%) compared to those lower class (116.7%). On the other hand, the highest average annual increase in hypertension occurred in the lower socioeconomic group (26.9%) as compared to those in the middle class (7.4%) and upper socioeconomic strata (7.1%). The massive increase in cases of diabetes within the upper class is clearly not due to the lack of resources for seeking health care. A more detailed analysis of their diet and lifestyle is needed to ascertain the real causes for the drastic increase relative to other socioeconomic groups. Table 5.5 presents information on the age of respondents and particular self-reported chronic conditions for 2002 and 2007. Based on this information, there is a change in the face of particular chronic ailments in Jamaica. The face is changing to reflect the inclusion of those less than 30 years of age (including children) as distinct from the elderly population.

109

Discussion
The present study revealed that the prevalence of particular chronic diseases (i.e. diabetes mellitus, hypertension and arthritis) increased from 8 per 1,000 in 2002 to 56 per 1,000 in 2007. The average annual increase of particular chronic illnesses was 17.2%. Diabetes mellitus

showed an exponential average annual increase of 185% compared to hypertension (+ 12.7%) and arthritis (- 3.8%). While hypertension remained the most prevalent of the particular chronic diseases in this study, diabetes mellitus showed the greatest annual increase. The transitions of particular chronic conditions are accounted for by (1) urban-to-rural shift, (2) female-to-male, (3) aged-to-young people, and (4) lower socioeconomic strata to upper class. The average annual increase in particular chronic diseases was greatest among those in the lower socioeconomic groups. However when the particular chronic ailments were disaggregated, the findings indicated that those in the wealthy socioeconomic group had the largest prevalence increase in diabetes mellitus, hypertension was greatest among those in the lower class and those in the upper class had the greatest reduction in arthritic cases. Particularly of note is the switching from public health care utilization by particular chronically ill respondents to private health care utilization. Similarly, the prevalence of health insurance coverage on average saw an exponential annual increase of 148%, while health care seeking behaviour over the same period showed a marginal increase of 12%. There is an emerging body of literature to support the changing face of people with particular chronic diseases from old ages (30+ years) to younger people including children [2832]. Traditionally chronic conditions such as diabetes mellitus were mostly prevalent among the elderly. This reality supports the large reservoir of literature on elderly diabetic, hypertensive
110

and arthritic patients. With the emergence of epidemiological and population transition, much attention was placed on diseases in middle and later ages as well as those conditions that accounted for most of the mortality and morbidity in a population. Because lifestyle practices were mostly responsible for chronic illness, many researchers limited their investigation to people 30+ years old [8-11, 23, 33 and 34]. The present paper supports the literature that particular self-reported chronic diseases (such as diabetes, hypertension and arthritis) are found mostly among the elderly (60+ years). The findings revealed that the mean ages of those with the specific self-reported chronic ailments have fallen marginally in Jamaica over the period (2002-2007). This is somewhat deceptive as 41% of those with diabetes were less than 60 years of age, compared to 40% of those with hypertension and 31% of arthritic respondents. Two percent of diabetic respondents were less than 15 years of age, but no children had hypertension or arthritis. Similarly, increases were observed in diabetes and arthritis for the young adult (diabetics aged 15 30 years) for the period. This is evidence that self-reported particular chronic diseases are changing face as almost 5% of diabetics were less than 31 years old in 2007 compared to 0% in 2002. Another emerging face of particular self-reported chronic illness is that of those with arthritis, as almost 2% of cases were among people ages 15-30 years of age. The young face of those with diabetes and other chronic diseases can be accounted for by (1) maternal nutrition during pregnancy [31], (2) diet [35] and the environment [30]. The sedentary lifestyles of the youth in the population are further entrenched by the modern electronic games which have removed the young person from the playing field and see him spending longer periods on the couch in front of the television. This hooked-on-egame syndrome
111

has also resulted in the increased consumption of sweet snacks and other so-called junk food. The new face of those with particular chronic diseases is changing, and this reality is therefore a cause for public health concern. This means that policy makers, health care practitioners, educators and the wider community need to recognize that chronic conditions such as diabetes, hypertension and arthritis have begun manifesting in young people as well as children. There is an urgent rationale for an intervention campaign that will sensitize educators, medical practitioners, parents, and children about the current reality of children and young adults being diagnosed with particular chronic illnesses. The intervention programme that should be formulated must include signs of ailments, place of reference, chronic disease management, nutrition, and medical practitioners understanding that testing for diabetes, hypertension and arthritis must be a rudimentary part of medical examinations, even of children, and further, even if their parents are not experiencing those conditions. The emerging young face of diabetics, and hypertensive and arthritis patients requires a new thrust in the study of mortality and morbidity data for health planning. Although diabetes, hypertension and arthritis may not be among the 10 leading causes of mortality in Jamaica [36] or the developing society, the emergence of those conditions requires researchers, demographers, epidemiologists and policy makers to embark on the inclusion of data on those conditions in publications in order that they can be examined. In a recently conducted study by Wilks et al. [37], they used teens of 15+ years to present information on those with particular diseases, but neglected to mention the new reality of children of younger ages with particular chronic illnesses. The new reality means that researchers, policy makers and the general society need to be cognizant of these facts. This will be accommodated by researchers, and in particular the statistical agency, publishing findings on the new reality in order to commence the discourse and
112

intervention campaign. With the absence of information on the matter, this can be construed as a miniscule problem. However, the new findings are reflecting the early onset of diabetes (< 15 years) and the provision of data beginning at 15 years omits 0.8% of infected children or 2.4% of diabetics. The present paper unearths more information on the new faces of those with particular chronic conditions at younger ages. Fifty-four out of every 100 persons with particular chronic diseases (i.e. diabetes, hypertension and arthritis) had hypertension, 32 out of every 100 had diabetes and 15 out of every 100 had arthritis. Despite the majority of those with particular chronic illnesses having hypertension, the prevalence rate for those with diabetes increased exponentially more than the other conditions. Many studies have established a relationship between poverty and illness [1, 2, 16 and 22], and particularly poverty and chronic illness [15]. Van et al.s work [15] revealed that chronic diseases were greater among those in the lower socioeconomic strata than the other social classes, but this study found that more people in the wealthy class had diabetes, while more hypertensive and arthritic respondents were in the lower socioeconomic group. The current findings are providing some clarification for Van et al.s research. Although the prevalence rate of particular chronic illnesses was greater among the wealthy strata for 2002 and 2007, those in the lower socioeconomic group recorded the greatest average annual percentage change. On disaggregating the particular chronic diseases, the present paper showed that the prevalence of diabetes was greater among the upper than the lower class, and the opposite was noted for hypertension and arthritis. This finding does not only clarify Van et al.s research, but provides pertinent information on the unhealthy lifestyle practices among
113

the wealthy, and reinforces the role of material deprivation on health, health conditions and mortality. Two scholars opined that money can buy health [38], implying that health is a transferable commodity, and that unhealthy lifestyle practices by the wealthy can be reversed with money. Clearly Smith and Kingtons claim [38] can be refuted as 42 out of every 100 chronically ill respondents were in the upper class, and more than half of those with diabetes were part of the wealthy income group. For any postulation to hold true about money purchasing health, one of the key axioms that needs to be looked at is the health conditions being lower among the wealthy than those in the lower class. The wealthy will continue to live by their desires, and at the onset of chronic ailments, may be able to reverse this by medical expenditure. It is well established that income is positively correlated with health, as money affords a particular diet, nutrition, medical facilities, safe drinking water, proper sanitation, leisure and good physical milieu, but the reality is that whenever unhealthy lifestyle practices become the choice of an individual, his/her money will not be able to eradicate the onset of diabetes, hypertension, heart disease, or other chronic diseases. Therefore, money enhances the scope of better health, but it cannot buy good health as this is not transferable from one person to the next. The very reason that health is non-transferable is the rationale behind the mortality of the wealthy elderly, and morbidity among the upper class. Socioeconomic status was found to be the strongest determinant of variations in health [39, 40], as wealth allows for particular choices, opportunities, access, resources and privileges that are not available to the poor. While those matters provide a virtual door leading to better health, money or wealth does not reduce the risk of ill-health arising from poor choices. A study by Wilks et al. [37] found that most (71%) of
114

those in the upper socioeconomic strata currently use alcohol which is more than those in the lower class (59%) and the middle class (64%). Twice as many people in the upper class (14%) had heart attacks compared to those in the middle class (7%) and 6% in the lower class [37]. The evidence is in that concretizes and refutes the proposition that money can buy health, and although the association between income and health is well established, unhealthy lifestyle choices cannot be reversed with money. The carbonated soft drink industry is experiencing a boom in the USA and the Caribbean [41, 42]. Recently, research conducted by Ha et al. [41] found that carbonated soft drinks and milk were the two most popular non-alcoholic beverages in the USA. They accounted for 39.1% of total beverage consumption. This explosion in carbonated soft drinks means that added sugar is infesting the dietary intake of young people and children more than in previous decades. Another study showed that among children aged 6 to 19 years there was a positive significant statistical association with soft drink consumption and a negative one with milk intake [43]. A sedentary lifestyle along with the consumption of sugar, salted food and fast food are accounting for the overweight and obesity in the world. According to Bostrom and Eliasson [44], over 50% of men and 33.3% of women between the ages of 16 and 74 years in Sweden are overweight and obese. Wilks et al. [37] found that 73% of Jamaicans aged 15 to 74 years practice a sedentary lifestyle, and obesity was the third most popular disease (5.6% of the population, 8.5% of females and 2.7% of males) behind hypertension (20.2%) and diabetes mellitus (7.6%). The growing global tsunami of chronic diseases in developing countries, and in particular Jamaica, requires urgent policy and public health intervention. The carbonated soft drink industry has infiltrated the consumption intake of young adults and children. Sugar in the form of
115

sweets (lollipops, candies, et cetera) is sold in every shop and supermarket, and at school gates in Jamaica. Children and young adults are fed a diet of more sugar than vegetables, beans, legumes, nuts, protein, diary products, fruits and fibre. Embedded in the increase in diabetes in children and young adults in Jamaica are parents and childrens nutritional intake (or lack thereof), as the dietary habits of Jamaicans have changed to include more fast foods and less nutrient dense diets. This extends beyond Jamaica to Barbados [44] and the USA [41]. With the exponential increase in diabetes over the last 5 years in Jamaica, and the increase in unhealthy lifestyle practices of the people, coupled with the sales explosion of the carbonated soft drink industry and the increase in fast food outlets, Jamaica is experiencing a diabetes epidemic which cannot be resolved without government and policy interventions. As is clearfrom the literature, with the increase in carbonated soft drinks, reduction in milk intake and influx of fast food entities in the Americas, the diabetes epidemic of Jamaica may become a reality across the Americas. This is not just affecting countries in the Americas, as studies have shown that Type 2 diabetes has become a global public health problem [46, 47]. The WHO contextualized the global public health Type 2 Diabetes epidemic when it stated that during 1999-2025 the prevalence of this ailment will be 40% in the developed nations and 170% in the developing countries. Clearly this paper is showing that diabetes has now reached an epidemic state in Jamaica, and may no longer be an epidemic but a pandemic disease. Type 2 diabetes is no longer an adult or later life disease, as was the case a generation ago, as it is now being diagnosed in children in Jamaica and other countries [48, 49]. This study highlights the changing image of those with particular chronic diseases (i.e. diabetes, hypertension and arthritis) in Jamaica. With 2 out of every 100 diabetics being children
116

(< 15 years) and the new image of hypertensive and arthritic patients being 15 30 years, plus the exponential increase in diabetes in the wealthy class, the present research highlights significant public health problems. In the last half a decade (2002-2007), the average annual increase in diabetes mellitus has risen by 185% indicating the unhealthy lifestyle practices of pregnant women, children and other young adults. The image of particular chronic illness in Jamaica continues to be lower class female and rural residents, but the average annual increase in diabetes mellitus was 200% for those in the wealthy class, compared to 117% of those in the lower socioeconomic class. Forty-seven out of every 100 chronically ill people in Jamaica utilize public health care facilities, which denotes that the matter is a public one and not solely individual. The cost of public health care in the next 5-10 years will increase phenomenally, as greater proportions of the population who rely on the public health care system will be afflicted with these chronic diseases. This has serious implications for the sustainable development of developing countries as well as their future achievements regarding the United Nations Millennium developments goals. To act now will not only save lives but will also save the various developing countries billions of dollars that can be spent on other development programmes. The demographic transition, in particular chronic conditions, now demands that data collection on those illnesses be lowered to < 15 years. Apart from the lowering of the ages in the data collection process, public health specialists need to address the massive changes in new diabetic cases. This is an obvious problem, which requires public health intervention as well as lifestyle management of diabetes. This sensitization and lifestyle management campaign must

117

extend to include educators, parents, children, vendors (especially those at schools), and the government. Governments need to regulate the sugar content of products in Jamaica (carbonated soft drinks, confectionary and fast food) as this is contributing to a public health problem which will cost the government and people in the medium to long-term. Diabetes can be likened to a tsunami in Jamaica and one that demands government intervention. Currently, there is a lifestyle campaign dealing with sexual behaviour, condom usage, and cancer in Jamaica; this research highlights the urgent need for a diabetes campaign that extends beyond parents to include vendors, confectionaries, soft drink manufacturers and government, in order to sensitize the public about this new public health problem. The gravity of the situation is that such a programme cannot be delayed for some time in the future as the opportunity costs of delay are (1) higher public expenditure, (2) increasing cost of diabetic care and management, (3) lower production cost, (4) increased unemployment benefits, (5) the imputed cost of ignorance, and (6) an increased mortality rate. Conclusion In summary, the theoretical position that underlines testing for diabetes among other chronic diseases should be abandoned, as the findings show the need to begin rudimentary health examinations of all ages. The new thrust of governments, public health specialists and researchers is to commence a mandate that addresses confectionary products ingredients, and institution guidelines about the sugar and salt components of manufactured commodities. The wider confectionary and food industry cannot be left unregulated as the chronic diseases tsunami is upon us, and it will require a concerted effort from everyone to combat this public health
118

problem as the nation addresses the diabetes epidemic. Diabetes has risen to such epidemic proportions that it now requires a policy initiative aimed at reducing the level of increases in a managed way.

Conflict of interest
The authors have no conflict of interest to report.

Disclaimer
The researchers would like to note that while this study used secondary data from the Jamaica Survey of Living Conditions, none of the errors in this paper should be ascribed to the Planning Institute of Jamaica or the Statistical Institute of Jamaica, but to the researchers.

119

References
65. World Health Organization. The World Health report 2002: reducing risks, promoting healthy life. Geneva: World Health Organization; 2002. 1. Bourne PA, McGrowder DA. Health status of patients with self-reported chronic diseases in Jamaica. North Am J Med Sci. 2009; 1: 356-364. 2. World Health Organization. Diet, physical activity and health. Geneva: World Health Organization; 2002. 3. Figueroa JP. Health trends in Jamaica. Significant progress and a vision for the 21st century. West Indian Med. J 2001; 50 (Suppl 4):15-22. 4. Guilliford MC. Epidemiological transition in Trinidad and Tobago, West Indies 19531992. Int J Epidemiol. 1996; 25:357-65. 5. McGlashen ND. Causes of death in ten English-speaking Caribbean countries and territories. Bull Pan Am Health Organ. 1982; 16: 212-22. 6. Holder Y, Lewis, MJ. Epidemiological overview of morbidity and mortality. In: PanAmerican Health Organization, editor. Health conditions in the Caribbean. Washington: PAHO; 1997. p. 22-45. 7. Morrison E. Diabetes and hypertension: twin trouble. Cajanus 2000; 33: 61-63. 8. Callender J. Lifestyle management in the hypertensive diabetic. Cajanus 2000; 33: 67-70. 9. Steingo B. Neurological consequences of diabetes and hypertension. Cajanus 2000; 33: 70-76 10. Eldemire S. A situational analysis of the Jamaican elderly, 1992. Kingston: Planning Institute of Jamaica; 1995. 11. Morrison EY, Ragoobirsingh D, Thompson H, Fletcher C, Smith-Richardson S, McFarlane S, et al. Phasic insulin dependent diabetes mellitus: manifestations and cellular mechanisms. J Clin Endo & Metabl. 1995; 80: 1996-2001. 12. Hennis A, Wu Suh-Yuh, Nemesure B, Li Xiaowei, Leske MC. Diabetes in a Caribbean population: epidemiological profile and implications. Int J of Epidemiol. 2002; 31: 234239. 13. Riste L, Khan F, Cruickshank K. High prevalence of Type 2 diabetes in all ethnic groups, including Europeans, in a British Inner City. Diabetes Care 2001; 24: 1377-1383.
120

14. Van Agt HME, Stronks K, Mackenbach JP. Chronic illness and poverty in the Netherlands. Eur J of Public Health 2000; 10: 197-200. 15. World Health Organization. Preventing chronic diseases a vital investment. Geneva: WHO; 2005. 16. Barcelo A. Diabetes and hypertension in the Americas. West Indian Med J. 2000; 49: 262-265. 17. Yach D, Hawkes C, Gould CL, Hofman, KJ. The global burden of chronic diseases: overcoming impediments to prevention and control. JAMA 2004; 291: 2616-2622. 18. Catford J. Chronic disease: preventing the worlds next tidal wave the challenge for Canada 2007? Health Promotion International 2007: 22:1-3. 19. Huicho L, Trelles M, Gonzales F, Mendoza W, Miranda J. Mortality profile in a country facing epidemiological transition: an analysis of registered data. BMC Public Health 2009; 9: 47. 20. Choi BC, Corber SJ, McQueen DV, Bonita R, Zevallos JC, Douglas KA. Enhancing regional capacity in chronic disease surveillance in the Americas. Rev Panam Salud Publica. 2005; 17: 130-141. 21. Foster AD. Poverty and illness in low-income rural areas. the Am Economic Review 1994; 84 (2): pp. 216-220. 22. Swaby P, Wilson E, Swaby S, Sue-Ho R, Pierre R. Chronic diseases management in Jamaican setting: HOPE worldwide Jamaicas experience. PNG Med J. 2001; 44: 171175. 23. Eliasson M, Bostrom G. Major public health problems diabetes. Scandinavian J of Public Health 2006; 34(Suppl 67): 59-68. 24. Planning Institute of Jamaica, Statistical Institute of Jamaica. Jamaica survey of living conditions, 1989-2007. Kingston: PIOJ, STATIN; 2008. 25. World Bank. Jamaica Survey of living conditions, 1988-2000: basic information [Internet]. Washington: The World Bank; 2002. [cited 2009 Sept 2]. Available from: http://siteresources.worldbank.org/INTLSMS/Resources/33589861181743055198/3877319-1190214215722/binfo2000.pdf). 26. Statistical Institute Of Jamaica. Jamaica survey of living conditions, 2007. Kingston, Jamaica: Statistical Institute of Jamaica. [Unpublished manuscript].

121

27. Hagley KE. Diabetes mellitus the deluge. In: Morgan O, editor. Health issues in the Caribbean. Kingston: Ian Randle; 2005. p. 115-121. 28. Murphy MJ, Metcalf BS, Voss LD, Jeffery AN, Kirby J, Mallam KM et al. Girls at five are intrinsically more insulin resistant than boys: the programming hypotheses revisited the early bird study (Early Bird 6). Pediatrics 2004; 113: 82-86. 29. Fogel RW. Changes in the disparities in chronic diseases during the course of the 20th century. Perspectives in Biology and Medicine 2005; 48: S150-S165. 30. Martin-Gronert MS, Ozanne SE. Maternal nutrition during pregnancy and health of the offspring. Biochemical Society Transactions 2006; 34: 779-782. 31. Marvicsin D. School-age children with diabetes: role of maternal self-efficacy, environment, and management behaviors. The Diabetes Educator 2008; 34: 477-483. 32. Middelkoop B JC, Kesarlal-Sadhoeram SM, Ramsaransing GN, Struben HWA. Diabetes mellitus among South Asian inhabitants of The Hague: high prevalence and age-specific socioeconomic gradient. Int J of Epidemiol. 1999; 28: 1119-1123. 33. Barreto SM, Figueirredo RC. Chronic diseases, self-perceived health status and health risk behaviors: gender differences. Rev Saude Publica. 2009; 43: 1-9. 34. Merchant AT, Dehghan M, Behnke-Cook D, Anand SS. Diet, physical activity, and adiposity in children in poor and rich neighbourhoods: a cross-sectional comparison. Nutrition Journal 2007; 6: 1. 35. Statistical Institute of Jamaica. Demographic statistics, 2007. Kingston: STATIN; 2008. 36. Wilks R, Younger N, Tulloch-Reid M, McFarlane S, Francis D: Jamaica health and lifestyle survey 2007-8. Kingston: Tropical Medicine Research Institute, University of the West Indies, Mona; 2008. 37. Smith JP, Kington R. Demographic and economic correlates of health in old age. Demography 1997; 34: 159-170. 38. Adler NE, Boyce T, Chesney MA, Folkman S, Syme SL. Socioeconomic inequalities in health: no easy solution. JAMA 1993; 269: 3140-3145. 39. Williams DR, Collins C. US socioeconomic and racial differences in health. Annu Rev Sociol. 1995; 21: 349-386. 40. Ha E-J, Caine-Bish N, Holloman C, Lowry-Gordon K. Evaluation of effectiveness of class-based nutrition intervention on changes in soft drink and milk consumption among young adults. Nutrition J. 2009; 8: 50.
122

41. Jacoby E. The obesity epidemic in the Americas: making healthy choices the easiest choices. PAHO Journal 2004; 15: 278-84. 42. Forshee RA, Storey ML. Total beverage consumption and beverage choices among children and adolescents. Int J Food Sci Nutr. 2003; 54: 297-307. 43. Bostrom G, Eliasson M. Major public health problems overweight and obesity. Scandinavian J of Public Health 2006; 34 (Suppl 67): 69-77. 44. Sharma S, Cao X, Harris R, Hennis AJM, Wu S, Leske MC. Assessing dietary patterns in Barbados highlights the need for nutritional intervention to reduce risk of chronic disease. J Hum Nutr Diet 2008; 21: 150-158. 45. Zimmet P. Globalization, coca-colonization and the chronic disease epidemic: can the doomsday scenario be averted. J Intern Med. 2000; 247: 301-310. 46. World Health Organization. Obesity: preventing and managing the global epidemic. Geneva: WHO; 1998. 47. American Diabetes Association. Type 2 diabetes in children and adolescents. Diabetes Care 2000; 23: 381-389. 48. Fagot-Campagna A, Pettitt DJ, Engelgau, MM, et al. Type 2 diabetes among North American children and adolescents: an epidemiological review and a public health perspective. J Pediatr. 2000; 136: 11-16.

123

Table 5.1: Operational definitions of particular variables Variable Self-evaluated health status (or health status) Sex Age group Operational definition Coding This is taken from the question In general, would you say your health is excellent, good, moderate, poor or very poor? Being male or female Age group is classified into 4 categories. Children - ages < 15 years old Young adults - 15 to 30 years old Other age adults 31- 59 years old Young old 60 74 years old Old old 75 84 years old Oldest old 85+ years old Low = poorest 20% to poor; middle = middle quintile and upper = wealthy to wealthiest 20% 1 = visits to health care professionals, 0=otherwise

Social hierarchy

Health careseeking behaviour (health seeking behaviour) Self-reported illness Chronic illness

Income quintiles were used to measure social class, and these range from quintile 1 (poorest 20%) to 5 (wealthiest 20%) Visits to pharmacies, medical practitioners, nurses in the last 4-weeks

Have you had any illness or injury during the past four weeks? For example, have you had a cold, diarrhoea, asthma, diabetes, hypertension, arthritis or other? These can be broadly defined as conditions which prolonged, do not resolved spontaneously, and are infrequently curable. This is taken from the question What are the illnesses that you have been diagnosed with Cold, diarrhoea, asthma, diabetes mellitus, hypertension, arthritis, other chronic conditions (unspecified)? The chronic conditions were diabetes mellitus, hypertension and arthritis.

124

Table 5.2. Demographic characteristic of sample, 2002 and 2007 2002 Characteristic n % Chronic illness Diabetes mellitus 12 5.8 Hypertension 126 60.9 Arthritis 69 33.3 Sex Male 58 28.0 Female 149 72.0 Marital status Married 95 46.1 Never married 50 24.3 Divorced 1 0.5 Separated 3 1.5 Widowed 57 27.7 Income quintile Poorest 20% 29 14.0 Poor 40 19.3 Middle 49 23.7 Wealthy 39 18.8 Wealthiest 20% 50 24.2 Health care utilization Public hospital 51 28.8 Private hospital 15 8.5 Public health centre 43 24.3 Private health centre 68 38.4 Health care utilization Sought medical care 163 79.1 Did not seek care 43 20.9 Health insurance status Insured 15 7.2 Uninsured 192 92.8 Age cohort Children 0 0.0 Young adults 2 1.0 Other age adults 49 23.7 Young-old 90 43.5 Old-old 58 28.0 Oldest-old 8 3.9 Area of residence Urban 24 11.6 Semi-urban 43 20.8 Rural 140 67.6
125

2007 n 123 206 56 113 272 163 130 14 10 64 83 65 76 79 82 72 23 61 126 258 125 126 258 3 10 137 132 82 21 95 78 212

% 31.9 53.5 14.5 29.4 70.6 42.8 34.1 3.7 2.6 16.8 21.6 16.9 19.7 20.5 21.3 25.5 8.2 21.6 44.7 67.4 32.6 32.8 67.2 0.8 2.6 35.6 34.3 21.3 5.5 24.7 20.3 55.1

Table 5.3. Self-reported diagnosed chronic illness by sex of respondents, 2002 and 2007 20021 20072 Characteristic Sex of respondents Sex of respondents Male Female Male Female n (%) n (%) n (%) n (%) Chronic illness Diabetes mellitus 3 (5.2) 9 (6.0) 31 (24.7) 92 (33.8) Hypertension 39 (67.2) 87 (58.4) 58 (51.3) 148(54.4) Arthritis 16 (27.6) 53 (35.6) 24 (21.2) 32 (11.8) 58 149 113 272 1 2 = 1.39, P = 0.499 2 2 = 6.09, P = 0.048

126

Table 5.4: Particular demographic and health variable by diagnosed chronic illness, 2002 and 2007 2002 2007 Chronic illness Chronic illness Characteristic Diabetes Hypertension Arthritis Diabetes Hypertension Arthritis mellitus mellitus n (%) n (%) n (%) n (%) n (%) n (%) Area of residence Urban 1 (8.3) 15 (11.9) 8 (11.6) 32 (26.0) 47 (22.8) 16 (28.6) Semi-urban 1 (8.3) 29 (23.0) 13 (18.8) 27 (22.0) 41 (19.9) 10(17.9) Rural 10 (83.3) 82 (65.1) 48 (69.6) 64 (52.0) 118 (57.3) 30 (53.6) Marital status Married 4 (33.3) 61 (48.4) 30 (44.1) 48 (40.0) 91 (44.4) 24 (42.9) Never married 4 (33.3) 30 (23.8) 16 (23.5) 39 (32.5) 69 (33.7) 22 (39.3) Divorced 0 (0.0) 0 (0.0) 1 (1.5) 10 (8.3) 3 (1.5) 1 (1.8) Separated 0 (0.0) 2 (1.6) 1 (1.5) 4 (3.3) 5 (2.4) 1 (1.8) Widowed 4 (33.4) 33 (26.2) 20 (29.4) 19 (15.8) 37 (18.0) 8 (14.3) Health utilization Public hospital 3 (30.0) 31 (29.5) 17 (23.0) 27 (32.9) 35 (25.5) 10 (32.3) Private hospital 1 (10.0) 9 (8.6) 5 (6.8) 11 (13.4) 7 (5.2) 5 (16.1) Public centre 2 (20.0) 21 (20.0) 20 (27.0) 23 (28.1) 34 (24.8) 4 (12.9) Private centre 4 (40.0) 44 (41.9) 32 (43.2) 21 (25.6) 61 (44.5) 12 (38.7) Health seekers Did not 1 (9.1) 26 (20.6) 16 (23.2) 34 (27.6) 66 (32.0) 27 (48.2) Sought 10 (90.9) 100 (79.4) 53 (76.8) 89 (72.4) 140 (68.0) 29 (51.8) Education Primary 8 (66.7) 73 (59.8) 43 (63.2) 121 (98.4) 205 (99.5) 56 (100.0) Secondary 4 (33.3) 47 (38.5) 24 (35.3) 2 (1.6) 0 (0.0) 0 (0.0) Tertiary 0 (0.0) 2 (1.6) 1 (1.5) 0 (0.0) 1 (0.5) 0 (0.0) Health coverage Uninsured 11 (91.7) 114 (90.5) 67 (97.1) 69 (56.1) 148 (71.8) 41 (74.5) Insured 1 (8.3) 12 (9.5) 2 (2.9) 54 (43.9) 58 (28.2) 14 (25.5) Social class Lower 6 (50.0) 35 (27.8) 28 (40.6) 41 (33.3) 82 (39.8) 25 (44.6) Middle 0 (0.0) 35 (27.8) 14 (20.3) 16 (13.0) 48 (23.3) 12 (21.4) Upper 6 (50.0) 56 (44.4) 27 (39.1) 66 (53.7) 76 (36.9) 19 (33.9) Health status Very good NI NI NI 5 (4.1) 10 (4.9) 1 (1.8) Good NI NI NI 21 (17.1) 45 (21.8) 12 (21.4) Fair NI NI NI 67 (54.5) 91 (44.2) 25 (44.6) Poor NI NI NI 26 (21.1) 52 (25.2) 18 (32.1) Very poor NI NI NI 4 (3.3) 8 (3.9) 0 (0.0)
NI No information Significant (P < 0.05)

127

Table 5.5. Age of respondent by particular chronic illness, 2002 and 2007 2002 2007 Chronic illness Chronic illness Characteristic Diabetes Hypertension Arthritis Diabetes Hypertension Arthritis mellitus mellitus n (%) n (%) n (%) n (%) n (%) n (%) Age cohort Children 0 (0.0) 0 (0.0) 0 (0.0) 3 (2.4) 0 (0.0) 0 (0.0) Young adult 0 (0.0) 2 (1.6) 0 (0.0) 3 (2.4) 6 (2.9) 1 (1.8) Other age adult 5 (41.7) 31 (24.6) 13 (18.8) 44 (35.8) 76 (36.9) 17 (30.4) Young-old 5 (41.7) 54 (42.9) 31 (44.9) 49 (39.8) 61 (29.6) 22 (39.3) Old-old 2 (16.7) 32 (25.4) 24 (34.8) 19 (15.4) 49 (23.8) 14 (25.0) Oldest-old 0 (0.0) 7 (5.6) 1 (1.4) 5 (4.1) 14 (6.8) 2 (3.6) Age Mean (SD) 62.1 67.2 (12.8) 68.4 60.9 62.5 (16.8) 64.3 (12.6) (11.5) (16.0) (14.5) Significant (P < 0.05)

128

6
Self-assessed health of young adults in an English-speaking Caribbean nation
Paul Andrew Bourne & Christopher A.D. Charles

Gender differences in self-assessed health in young adults (i.e. ages 15 44 years) are understudied in the English-speaking Caribbean. The aims of the current research are to (1) provide demographic characteristics of young adults; (2) examine self-assessed health of young adults; (3) identify social determinants that explain good health status for young adults; (4) determine the magnitude of each social determinant, and (5) gender differences in self-assessed health. One percent of sample claimed injury and 8% illness. Self-reported diagnosed illnesses were influenza (12.7%); diarrhoea (2.9%); respiratory disease (14.1%); diabetes mellitus (7.8%); hypertension (7.8%); arthritis (2.9%) and unspecified conditions (41.2%). The mean length of illness was 26.0 days (SD = 98.9. Nine social determinants and biological condition explained 19.2% of the variability of self-assessed health. The biological condition accounted for 78.1% of the explanatory model. Injury accounts for a miniscule percentage of illness and so using it to formulate intervention policies would lack depth to effectively address health of this cohort.

Introduction

Gender differences in self-assessed health in young adults (i.e. ages 15 44 years) are understudied in the English-speaking Caribbean. Previous studies that have examined young adults have focused on reproductive health; survivability; teenage pregnancy; substance use and abuse; HIV/AIDS; injuries and impact of injuries on health [1-7]. While studies on injuries have shown
129

that young males 15 to 44 years are mostly affected by violent-injuries [6, 7], in Jamaica, statistics [8] revealed that many of the deaths occurred in this age group can be accounted for by injuries. Injuries are among reasons for ill-health and by extension do not constitute a significant percentage of illness. Injuries accounted for most morbidities and/or mortalities in the world [7], but this is not typical to Jamaica, making studies on injuries germane but lacks extensive coverage on health. Statistics on Jamaica showed that of the 10 leading causes of mortality, in 2002 [8-10], homicides and injuries were the 5th and 10th causes of deaths respectively [10]. In 2004, statistics from the World Health Organization (WHO) showed that injuries were the 4th leading cause of morality in Jamaica [11] and in 2006, statistics from the Jamaica Ministry of Health [9] indicated that injuries was not among the 5 leading cases of hospital utilisation in Jamaica. Policies therefore in Jamaica would not have been formulated using general health status research, but more so from data on injuries, reproductive health, survivability and mortalities. Policy intervention on those issues are pertinent and cannot be neglected from the general pursuit of health, using general health status and health conditions would provide invaluable insights from the individuals perspective on those issues; which would add value to addressing health concerns that waiting for particular outcomes such as pregnancies, mortality, injuries or crime, violence and victimization by young adults. A study by Hambleton et al. [12] identified that illness constituted significant percentage of the explanatory power of self-assessed health of older Barbadians (ages 60+ years) and while this provides some understanding of the role of illness on general health status of which may be caused by injuries, the research identified other factors (i.e. social determinants) that played roles in health status determination.

130

Injuries therefore do account for a percentage of ill-health, indicating that a study of their typologies is imperative but this cannot abate or replace a study on general health of young adults. An extensive reveal of health literature in the English-speaking Caribbean nations found a lack of studies on the general health status of young adults. Empirical literature showed that any study of health must coalesce biological and social determinants [13- 25], which is also lacking for young adults. Recently a study by Bourne [26] provided invaluable insights into the typology of health conditions and the demographic shifts in these between 2002 and 2007. Tables 1-3 highlight hospital utilisation for gunshot wounds and suicides, and victim prolife of individuals in Jamaica for 2005. The data highlights the crime and hospital utilisation profile, which indicates that health care utilisation and victims of crimes are substantially between 14 and 45 years. Age 15 45 years does not only represent most of the victims of crime, mortality and hospital utilization in Jamaica, it also denotes the group which constitutes arrest for major crimes. Some of the issues are social and do affect mortality, but what about those persons of this group who are alive and fear being a victim of violence as well as those who reside in those communities in which such incidences are perpetuated each day. In addition what about their general health as well as those members of this age group who are not likely victims owing to other social conditions such as social hierarchy, area of residence or those who do not reside in inner-cities communities. It is within this context that the current paper chose to examine selfreported health of this group in order to provide insights into the health of young adults and the social determinants that explain their health status. The aims of the current research are to (1) provide demographic characteristics of young adults; (2) examine self-assessed health of young adults; (3) identify social determinants that

131

explain good health status for young adults; (4) determine the magnitude of each social determinant, and (5) gender differences in self-assessed health.

Materials and Methods


The Jamaica Survey of Living Conditions (JSLC) was commissioned by the Planning Institute of Jamaica (PIOJ) and the Statistical Institute of Jamaica (STATIN) in 1988 [27]. These two organizations are responsible for planning, data collection and policy guidelines for Jamaica, and have been conducting the JSLC annually since 1989 [28]. The JSLC is an administered questionnaire where respondents are asked to recall detailed information on particular activities. The questionnaire was modelled from the World Banks Living Standards Measurement Study (LSMS) household survey [28]. There are some modifications to the LSMS, as JSLC is more focused on policy impacts. The questionnaire covers demographic variables, health,

immunization of children 059 months, education, daily expenses, non-food consumption expenditure, housing conditions, inventory of durable goods and social assistance. Interviewers are trained to collect the data from household members. The survey is conducted between April and July annually. The current paper extracted a sub-sample of 3,024 respondents (i.e. ages15 44 years) from a larger nationally cross-sectional survey of 6,782 Jamaicans. This study used the dataset of the JSLC for 2007 [29]. Measures An explanation of some of the variables in the model is provided here. Self-reported is a dummy variable, where 1 (good health) = not reporting an ailment or dysfunction or illness in the last 4 weeks, which was the survey period; 0 (poor health) if there were no self-reported ailments,
132

injuries or illnesses. While self-reported ill-health is not an ideal indicator of actual health conditions because people may underreport, it is still an accurate proxy of ill-health and mortality. Social supports (or networks) denote different social networks with which the individual is involved (1 = membership of and/or visits to civic organizations or having friends who visit ones home or with whom one is able to network, 0 = otherwise). Psychological conditions are the psychological state of an individual, and this is subdivided into positive and negative affective psychological conditions. Positive affective psychological condition is the number of responses with regard to being hopeful, optimistic about the future and life generally. Negative affective psychological condition is number of responses from a person on having lost a breadwinner and/or family member, having lost property, being made redundant or failing to meet household and other obligations. Health status is a binary measure (1=good to excellent health; 0= otherwise) which is determined from Generally, how do you feel about your health? Answers for this question are in a Likert scale matter ranging from excellent to poor. Health care-seeking behaviour is derived from the question: Have you visited a health care practitioner, pharmacist or healer in the past four 4 weeks, with an option of yes or no. For the purpose of the regression the responses were coded as 1=yes, 0=otherwise. Crowding is the total number of individuals in the household divided by the number of rooms (excluding kitchen, verandah and bathroom). Age is a continuous variable in years. Statistical analysis Statistical analyses were performed using the Statistical Packages for the Social Sciences v 16.0 (SPSS Inc; Chicago, IL, USA) for Widows. Descriptive statistics such as mean, standard deviation (SD), frequency and percentage were used to analyze the socio-demographic
133

characteristics of the sample. Chi-square was used to examine the association between nonmetric variables, and an Analysis of Variance (ANOVA) was used to test the relationships between metric and non-dichotomous categorical variables. Logistic regression examined the relationship between the dependent variable and some predisposed independent (explanatory) variables, because the dependent variable was a binary one (self-reported health status: 1 if reported good health status and 0 if reported poor health status). The final model was based on those variables that were statistically significant (p <0.05), and all other variables were removed from the final model (p >0.05). Categorical variables were coded using the dummy coding scheme or a reference category. The predictive power of the model was tested using the omnibus test of model and Hosmer & Lemeshows [30] 3 technique was used to examine the models goodness of fit. The correlation matrix was examined in order to ascertain whether autocorrelation (or multicollinearity) existed between variables. Cohen & Holliday [31] stated that correlation can be low/weak (00.39); moderate (0.40.69), or strong (0.71). This was used in the present study to exclude (or allow) a variable. Finally, forward stepwise technique in logistic regression was used to determine the magnitude (or contribution) of each statistically significant variable in comparison with the others, and the odds ratio (OR) for interpreting each of the significant variables. Model To study the relationship between self-assessed health status and social determinants, biological conditions and welfare, and logistic regression was used to estimate the following regression
134

model. Equation [1] denotes the 20 social, SDHij, 3 welfare variables, Wij, and biological condition, Bi, of self-assessed health status (Hi) and some standard error:

[1]

Table 6.6 presents the results from the econometric exercise, which is captured in Equation [2]. Equation [2] therefore presents only those variables which are significantly correlated with selfassessed health status of young adults:

[2]

where: Hi is the level of self-assessed health status of person i.

SDHij denotes the 9 statistically significant social determinants of person i.

Results
The sample was 3,024 respondents: 47.6% males and 52.4% males. The mean age of the sample was 28.5 years (SD = 8.8 years). Thirty percent of the sample was single; 20.4% common-law; 13% married; and 27.1% in visiting unions. Thirty-six and three-tenth percent of the sample was poor with 17.1% in the poorest 20% compared to 44.1% in the wealthy social hierarchies, of which 23.2% was in the wealthiest 20%. Forty-five and nine tenth percent of the sample dwelled in rural area, 22% in peri-urban and 32.1% in urban areas. Of the sample population, with respect to the questions on injury and illness 97.1% and 97% responded respectively. Of those respondents, 1% claimed injury and 8% mentioned illness. When respondents were asked whether the illness was diagnosed and the typologies of conditions, 100% stated that the health
135

condition was diagnosed by a medical practitioner. The self-reported diagnosed health conditions were influenza (12.7%); diarrhoea (2.9%); respiratory disease (14.1%); diabetes mellitus (7.8%); hypertension (7.8%); arthritis (2.9%) and unspecified conditions (41.2%). The mean length of illness was 26.0 days (SD = 98.9), with 1 visit made to a health care practitioner in the last 4weeks. When respondents were asked if they had visited a health care practitioner (including healer, pharmacist, nurse, and wife) in the last 4-weeks, 64.2% said yes. The health care institutions were public hospitals (34.8%); private hospitals (7.0%); public health care centres (14%); and private health care centre (51.6%). Twenty percent of the sample had health insurance coverage; 89.6% claimed at least good health (including 42.2% very good selfassessed health) compared to 1.9% who stated at least poor health (including 0.3% very poor health). A cross-tabulation of health care-seeking behaviour and illness shows no significant statistical association. Ninety-seven percent of those who seek medical care were ill compared to 94% of those who sought medical care in the last 4-weeks. A cross-tabulation between illness and age group revealed a significant statistical association 2 = 39.4, P < 0.0001. Figure 6.1 provides the information on the age group and percentage of young adults who indicated that they had an illness in the last 4-weeks.

No significant statistical association was found between health care-seeking and age group (P = 0.608): age 15 19 years, 60%; age 20 24 years, 53.1%; age 25 29, 60.0%; age 30 34 years, 67.7%; age 35 39 years, 68.0% and age 40 44 years, 69.%.

136

No significant statistical relationship was found between health care-seeking behaviour and social hierarchy (P = 0.339): poorest 20%, 51.1%; poor, 69.2%; middle class, 67.4%; wealthy, 65.5%, and wealthiest 20%, 67.7%. There is a statistical difference between age of respondents who reported having particular health conditions F-test = 4.5, P < 0.001. The mean ages of particular health conditions were influenza, 29.3 years (SD = 9.2); diarrhea, 32.2 years (SD = 8.7); respiratory, 30.3 years (SD = 9.6); diabetes mellitus, 37.3 years (SD = 5.9); hypertension, 36.8 years (SD = 7.1) and other, 29.9 years (SD = 9.3). Figure 6.2 highlights young adult who reported injury (%) and illness (%) that dwelled in particular area of residence controlled for sex of respondents. Figure 6.2 showed that over 50% of those with illness and injury dwelled in rural areas. However, there was no significant statistical relationship when illness and injury by area of residence was controlled for by sex of respondents (illness male 2 = 2.6, P < 0.271 and female 2 = 2.3, P < 0.323; injury male 2 = 2.5, P < 0.292 and female 2 = 0.93, P < 0.628). Figure 6.3 shows sex composition of those who utilised health care facilities in Jamaica. Most young adult males utilised private hospitals (36.4%) compare to females who visited public health care (72.7%). The least percentage of females visited private hospitals (63.6%) compared to public health care centres for males (27.3%). Multivariate analysis Tables 6 represent the results from the econometric exercise: Of the 24 variables that were tested in an initial model, 9 were social determinants and 1 a biological variable. Biological variable (i.e. self-reported illness) accounted for 78.1% of the explanatory power of the model (i.e.
137

15.3%), indicating that the social determinants accounted for 21.9% of the self-assessed health status of young adults.

Limitations of study
Health is a function of social, psychological, economic, biological and ecological factors. Based on the multi-dimensional nature of health determinants, the present study used secondary survey data and variables such as psychological, ecological and some social issues; such as childhood health history, culture, belief and value system were omitted from the model. Those omissions reduced the explanatory power of the current paper, but provide a platform with which future studies can be launched.

Discussion
In the present study, the prevalence of injury in Jamaica for young adults was 1% compared to 8% in illness. A cross-tabulation between self-reported injury and self-reported illness showed a significant statistical relationship. The association was a very weak one, correlation coefficient = 0.12 (or 12%). Forty-one of every 100 young adults who reported having an injury stated that they had an illness in the last four-weeks, indicating that less than one- half percent of those with an injury had an illness. Concurrently, 2 times more young adult-females sought medical care more than males. On the other hand, males were 2.3 times likely to record injury while females were 2 times more likely to have an illness in the last 4-weeks. Furthermore, the odds ratio of recording better good self-assessed health status for males was 1.5 times more than that of females. Outside of the gender differences in self-assessed health status, medical care-seeking behaviour, and injuries, the odds ratio of recording good health married young adults was 1.6 times more than their single counterparts and this was similar for peri-urban respondents with
138

reference to rural young adults. On the other hand, a young adult who sought medical care was 65% less likely to record good health; young adults with tertiary level education were 47% more likely to record good health and those who spent more on medical care (i.e. medical careexpenditure) were 1% less likely to have good self-assessed health status. Empirically, research has established that any investigation of health must coalesce social, psychological, economic and biological variables [12-25, 32-37]. Hambleton et al. [12] went further when he disaggregated the contribution of biological and non-medical conditions of self-assessed health status. They found that 87.7% of the explanatory power of good health status of elderly Barbadians could be accounted for by current illness. The present study found that current illness accounted for 78.1%, which suggests that illness accounted for less of young adults health status than for elderly people. One of the challenges in effectively comparing the aforementioned issues (which is embedded in the data) is that the perception of people across different nations are not the same, and this as well as the age component could account for some aspects of the disparity.. The present study has not only highlighted the role that social determinants play in health status but also that they play a greater role in the health of younger adults than old people. Statistics seemingly show a large percent of young adults being victims of injuries but the current findings indicate that these represent a small part of ill-health of young adults. The small percent of injuries experienced by young adults denote that using injuries as a guide in health policy intervention would be addressing an even smaller percent of health status than illnesses. From the aforementioned results which show that illness contributes more to health status than social determinants, along with injuries. It is clear that despite the cultural and biological differences rooted in both figures, current illness is a strong determinant of selfassessed health status in each region and if health must combine social, biological, psychological
139

and ecological determinants, public health interventions that are using any one determinant in particular injuries would not be addressing the health concerns of young adults. This empirical evidence concretizes the rationale for social determinants in the discussion and research on health status as well as ill-health. The finding in the present paper showed that social determinants of young adults constituted more explanation than for elderly. Therefore the usage of injuries and/or illness to measure and guide public intervention denotes that1 in 5 of the health status of young adults would have been unaddressed in this effort and as much as 9 out of 10 of injury statistics are used in public policy interventions. Current social determinants of health for elderly Barbadians accounted for 4.1% of health and historical determinants, suggesting the increased role of biological determinant in the health process with ageing. Historical determinants which included education, occupation, children, economic situation, childhood nutrition, childhood health and diseases theoretically is apart of social determinants. Disaggregating social determinants to ascertain a value for historical determinants to compare with Hambleton et al.s finding in this study found that education was the only factor of those identified in the Barbadian health status, and that education accounted for only 0.3% of the explanatory model in this study. Therefore within the limitations of the current paper, meaningful comparison using disaggregated social determinants would be close to impossible as the components are not necessarily the same. Inspite of the limitations of the current work, the study can effective compare selfassessed health status as both studies collected this from its population. The current paper which uses data for 2007 and Hambleton et als work used data for December 1999 to June 2000 showed that young adults health was between 1.5 to 1.9 times more than that for elderly Barbadians. Although there are time differences which cannot be discounted for in this study,
140

there is emerging information in the reduction of health status with ageing. Ageing is a nature event. Imagine purchasing a new car, taking this car home and locking it away in the car porch under cover for 20 years; and on removing the covers although the item was not used, it would have aged. On using the car however increases the deterioration or depreciation on the human structure, and therefore account for illness, health care utilisation and lowered health status The issue of the car symbolizes the natural ageing and progressive depleted state of things and this is similarly the case for humans. The current paper revealed that as young people age, the odds ratio (OR = 0.97) of indicating good health falls by 3% and using the aforementioned statistics would mean that odds ratio of good health for elderly people should fall. A study by Bourne, McGrowder and Crawford [38] showed that illness affecting elderly Jamaicans was more chronic than acute compared to the converse in this study. With the changes in the typology of illnesses from acute to chronic conditions, the elderlys health status must be lower than that for young adults. Hence although homicides accounted for more deaths of young adults that elderly people, the health status of the former is still greater and this is due largely to lower risk of biological conditions. Again the biology of an individual accounts for greater percentage of self-assessed health than external factors such as injuries from accidents. Injuries from accident affect 1 in every 100 young adults, making its effect on health smaller than illnesses which accounts for 8 in every 100 young adults.With biological conditions accounting for more of self-assessed health of older people, this supports lower health status than young adults and greater health care participation for the former as they seek to address the ageing of the organism and the increased depreciation owing to old age. Gompertzs law in Gavriolov and Gavrilova [39] shows that there is a fundamental quantitative theory of ageing and mortality of certain species (the examples here are as follows
141

humans, human lice, mice, fruit flies, and flour beetles. Gompertzs law went further to establish that human mortality increases twofold with every 8 years of an adult life, which means that ageing increases in geometric progression. This phenomenon means that human mortality increases with age of the human adult, but that this becomes less progress in advance ageing. Thus, biological ageing is a process where the human cells degenerate with years (i.e. the cells die with increasing in age), which is explored in evolutionary biology [40-43]. But studies have shown that using evolutionary theory for late-life mortality plateaus, fail because of the arguable unrealistic set of assumptions that the theory uses to establish itself [44-46]. Ageing therefore denotes gradual deterioration in living organisms as well other nonliving items, which accounts for demand in medical care. Medical seeking-behaviour could indicate either preventative or curative care. The present study revealed that the odds ratio of good health of young adults in Jamaica decline by 65% for those who seek medical care. Medical care for young adults therefore is a good measure of curative than preventative care. This also speaks of the cultural impact on health through peoples conceptual perceptions of health; that health is illness and so care is sought for ill-health as against preventative care. The current work revealed that 94 out of every 100 young adults who sought medical care were ill; reinforcing the cultural perception of illness and the reason why young adults seek health-care is curative than preventative for this group. Illness in the current work is substantially a female phenomenon. Young adult females were 2 times more likely to report an illness, and this justifies their greater probability to utilize medical care seeking in order to address ill-health. These findings have a high degree of validity as statistics from the Ministry of Health (Jamaica) showed that females attended health care institutions twice as much as men for curative care since 2000-2007 [9]. Since 1988, statistics
142

obtained from Jamaicans in national cross-sectional surveys revealed that females were approximately more likely to report an illness and utilize medical care than males. This reinforced the cultural biasness of illness and health care facilities. Health care facilities are primarily governed by females for females and this adds to cultural handicap of males afford attending public health care institution on experiencing ill-health. , The feminization of health care facilities and the large percent of people in particular females who utilise public health care institution is another rationale for males use of private health care facilities. Males on the other hand will attend medical care facilities when ill-health interfaces with their economic livelihood and the severity is such that this is the only avenue. This is not atypical to Jamaica as a qualitative study in Pakistan on street children found that boys would attend formal health care if it affects their economic livelihood and health conditions were severe [47]. Another study conducted in Anyigba, North-Central, Nigeria found that [48] found that 85 out of every 100 respondents waited for less than a week after the onset of illness to seek medical, and that 57 out of every 100 indicated that they would recover without treatment. A Caribbean anthropologist [49] stated that the macho socialisation of the Caribbean male accounts for his unwillingness to seek medical care. Caribbean males including Jamaicans are socialised to be strong, do not show weakness, and be involved in particular tasks to exhibit their masculinity as a result illness is a signal of weakness, therefore accounting for the reasons why they are skeptical to visit medical institutions and often times wait for severity. On visiting medical practitioners, it is sometimes so difficult for traditional medical practioner to offer cure. This then offers an explanation for females living longing than males. Although the current findings showed that the odds of recording good health is 1.5 times greater for young adult males, apart of this is owing to the reality that often times males do not see themselves as ill,
143

visit medical practitioner less and justifies the higher mortality among them than females. The social determinants are therefore offering explanation for the biological issues as well, challenges to implement health interventions to improve health of young adults in particular males are great as definition of illness and severity of symptoms reduce the quality of life of people and this finding concurs with a previous study by Williams et al. [50]. Unlike this study, Williams et al. [50] found that medical care-seeking behaviour did not differ significant between the sexes, with this study finding the opposite. Like this paper, Dunlop et al [51] found that African American men had few physician contacts than minority and non-Hispanic white women. The irresponsiveness of young adult males in seeking health care comparable to their female counterparts in Jamaican extends to even older African American men. With the advancement in literacy and numeracy in the world since the 19th century, specifically in Jamaicans since 1960 (i.e. educational levels), empirical findings showed education is among the social determinants that influence health status [12-26]. Education affects health directly and indirectly. A study on twins in USA found that more years in schooling (i.e. education) was associated with healthier patterns of behaviour. [52], which is an example of the direct impact of education on health. In the Fujiwara & Kawachi [52] work on increased schooling was associated with reducing smoking habit and other such healthier practices. The current paper concurs with the literature as the odds ratio of good health status of young adults with tertiary level education are 1.5 times more than those with primary or below education. The indirect way that education affects health can be measured using social hierarchy. The present findings revealed that the middle class who are the educated ones were 1.5 times more likely to report good health status and that wealth or income was not correlated with good health status or for that matter the self-assessed health status of wealthy social hierarchies did not differ from
144

those in the poor social hierarchies. Empirical evidence existed that among the social determinants of health is marital status. Some research showed that married people are healthier than non-married people [12-25, 53-58]. Koo, Rie and Park [54] findings revealed that being married was a good cause for an increase in psychological and subjective wellbeing in old age. Smith and Waitzman [55] offered the explanation that wives found dissuade their husband from particular risky behaviours such as the use of alcohol and drugs, and would ensure that they maintain a strict medical regimen coupled with proper eating habit [53, 56]. In an effort to contextualize the psychosocial and biomedical health status of particular marital status, one demography cited that the death of a spouse meant a closure to daily communicate and shared activities, which sometimes translate into depression that affect the wellbeing more of the elderly who would have had investment must in a partner [57]. They pointed to a paradox within this discourse as this is not observed among men. To provide a holistic base to the argument, the researcher will quote a sentence from the findings of Delbs and Gaymu [57] study that reads The widowed have a less positive attitude towards life than married people, which is not an unexpected result [57]. The present study concurs with the literature that the health status of married young adults is greater than those who are single, but that this was only explained by females. Those findings highlight the value of marriage to females which commences at an early age, and seemingly that the benefits of marriage are not for males. This is clearly not the case as study by Bourne [58], using data on Jamaicans, found that the odds ratio of reported good health was 1.6 times more for married males than their female counterparts.

145

Conclusion and policy recommendations


In sum, statistics for 2007 revealed one in every two Jamaicans was 15-44 years old. This speaks to the importance of a research on this age group. With the demographic reality of young adults in the country, using injury to examine health is grossly inadequate, narrow and fails to understand the matter of health. Health is more that illness as it incorporates social, economic, psychological, ecological and biological determinants. While the biological determinant of selfassessed health of young adult predominates health determinants, injury accounts for a miniscule percentage of illness and so using injury to formulate intervention policies would be lacking in depth to effectively address health of this cohort. Although the health of young adult Jamaicans is very good, there are many health disparities between the sexes which are justifying inequities in health outcomes between males and females. The present study highlights some of the health disparities between the sexes and affords research findings that can be used to refashion health policies and research focus in the future. Health policies must utilize the wide spectrum of health determinants in order to address the multi-dimensional nature of health. The use of injuries to measure and guide policies and programmes because seemingly there are many young adults who are affected is a misnomer and does not capture the gamut of illness or even health of this group of people. The identified health disparities are among reasons for health inequities in health outcome, and should justify a call for a research and policy direction that include avoidabilities such as technical, financial and moral as these would provide additional explanations for health disparities, choices, inequity and/or inequalities in health outcomes among young adults.

146

References
1. Frederick J, Hamilton P, Jackson J, Frederick C, Wynter S, Dacosta V, Wynter H. Issues affecting reproductive health in the Caribbean. In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle; 2005: pp. 41-50. 2. McNeil P. Coping with teenage pregnancy. In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle; 2005: pp. 51-57. 3. Bain B. HIV/AIDS the rude awakening/stemming the tide. In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle; 2005: pp. 62-76. 4. Douglas K-G. Patterns of substance use among adolescents students in Jamaica. In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle; 2005: pp. 77-78. 5. Abel W. Substance use health consequences. In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle; 2005: pp. 79-86. 6. McDonald A. Injuries the broad picture. In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle; 2005: pp. 87-92. 7. McCartney TB. The health impact of injuries (Abridged). In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle; 2005: pp. 93-95. 8. Statistical Institute of Jamaica, (STATIN). Demographic statistics, 1970-2008. Kingston: STATIN; 1971-2009. 9. Jamaica, Ministry of Health (MoH). Annual report, 1990-2007. Kingston: MoH; 19912009. 10. Pan American Health Organization, (PAHO). Health in the Americas 2007, volume II countries. Washington DC: PAHO; 2007. 11. World Health Organization, (WHO). Health statistics, 2009. Geneva: WHO; 2009: p. 50. 12. Hambleton IR, Clarke K, Broome HL, Fraser HS, Brathwaite F, Hennis AJ. Historical and current predictors of self-reported health status among elderly persons in Barbados. Revista Panamericana de Salud Pblica 2005; 17(5-6): 342-352. 13. Graham H. Social Determinants and their Unequal Distribution Clarifying Policy Understanding The MilBank Quarterly 2004; 82 (1), 101-124. 14. Pettigrew M, Whitehead M, McIntyre SJ, Graham H, Egan M. Evidence for Public Health Policy on Inequalities: 1: The Reality According To Policymakers. Journal of Epidemiology and Community Health 2004; 5, 811 816. 15. Asnani MR, Reid ME, Ali SB, Lipps G, Williams-Green P. Quality of life in patients with sickle cell disease in Jamaica: rural-urban differences. Journal of Rural and Remote Health 2008; 8: 890-899. 16. Bourne PA, McGrowder DA. Rural health in Jamaica: examining and refining the predictive factors of good health status of rural residents. Rural and Remote Health 2009; 9 (2), 1116. 17. Bourne PA. Medical Sociology: Modelling Well-being for elderly People in Jamaica. West Indian Med J 2008; 57:596-04. 18. Bourne PA. Health Determinants: Using Secondary Data to Model Predictors of Wellbeing of Jamaicans. West Indian Med J 2008; 57:476-81. 19. Grossman M. The demand for health - a theoretical and empirical investigation. New York: National Bureau of Economic Research, 1972.
147

20. Smith JP, Kington R. Demographic and Economic Correlates of Health in Old Age. Demography 1997; 34:159-70. 21. Wilkinson RG, Marmot M. Social Determinants of Health. The Solid Facts, 2nd ed. Copenhagen: World Health Organization; 2003. 22. Solar O, Irwin A. A Conceptual Framework for Analysis and Action on the Social Determinants of Health. Discussion paper for the Commission on Social Determinants of Health DRAFT April 2007. Available from http://www.who.int/social_determinants/resources/csdh_framework_action_05_07.pdf (Accessed April 29, 2009). 23. Bourne PA, Rhule J. Good Health Status of Rural Women in the Reproductive Ages. International Journal of Collaborative Research on Internal Medicine & Public Health, 2009; 1(5):132-155. (Online). 24. Bourne PA. A theoretical framework of good health status of Jamaicans: using econometric analysis to model good health status over the life course. North American Journal of Medical Sciences, 2009; 1(2): 86-95. 25. Bourne PA. Socio-demographic determinants of Health care-seeking behaviour, selfreported illness and Self-evaluated Health status in Jamaica. International Journal of Collaborative Research on Internal Medicine & Public Health, 2009, 1 (4):101-130. 26. Bourne PA. Demographic shifts in health conditions of adolescents 10-19 years, Jamaica: Using cross-sectional data for 2002 and 2007. North American Journal of Medical Sciences 2009; 1(3):125-133. 27. Planning Institute of Jamaica, (PIOJ), Statistical Institute of Jamaica, (STATIN). Jamaica Survey of Living Conditions, 1989-2007. Kingston: PIOJ, STATIN; 1989-2008. 28. World Bank, Development Research Group, Poverty and Human Resources. Jamaica Survey of Living Conditions, 1988-2000. Basic information. Washington: The World Bank; 2002. (September 2, 2009, at http://siteresources.worldbank.org/INTLSMS/Resources/33589861181743055198/3877319-1190214215722/binfo2000.pdf). 29. Statistical Institute Of Jamaica. Jamaica Survey of Living Conditions, 2007 [Computer file]. Kingston, Jamaica: Statistical Institute Of Jamaica [producer], 2007. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors]; 2008. 30. Homer D, Lemeshow S. Applied Logistic Regression, 2nd edn. John Wiley & Sons Inc., New York, 2000. 31. Cohen L, Holliday M. Statistics for Social Sciences. London, England: Harper and Row, 1982. 32. Casas JA, Dachs JN, Bambas A. Health disparity in Latin America and the Caribbean: The role of social and economic determinants. In: Pan American Health Organisation. Equity and health: Views from the Pan American Sanitary Bureau, Occasional Publication No. 8. Washington DC; 2001: pp. 22-49. 33. Engel G. A unified concept of health and disease. Perspectives in Biology and Med 1960; 3:459-485. 34. Engel G. The care of the patient: art or science? Johns Hopkins Med J 1977; 140:222-232.
148

35. Engel G. The need for a new medical model: A challenge for biomedicine. Sci 1977; 196:129-136. 36. Engel G. The biopsychosocial model and the education of health professionals. Annals of the New York Academy of Sciences 1978; 310: 169-181 37. Engel, GL. The clinical application of the biopsychosocial model. Am J of Psychiatry 1980; 137:535-544. 38. Bourne PA, McGrowder DA, & Crawford TV. Decomposing Mortality Rates and Examining Health Status of the Elderly in Jamaica. The Open Geriatric Medicine Journal 2009; 2:34-44. 39. Gavrilov LA, Gavrilova NS. The reliability theory of aging and longevity. J. theor. Biol 2001; 213:527-545. 40. Medawar PB. Old age and natural death. Mod. Q. 1946; 2:30-49. [Reprinted in the Uniqueness of the Individual (Medawar, P. B., ed. 1958), pp. 17-43. New York: Basic Books. 41. Carnes B, Olshansky SJ. Evolutionary perspectives on human senescence. Population. Development Review 1993; 19: 793-806. 42. Carnes B A, Olshansky SJ, Gavrilov LA, Gavrilova NS, Grahn D. Human longevity: Nature vs. nurture - fact or fiction. Persp. Biol. Med. 1999; 42: 422-441. 43. Charlesworth B. Evolution in Age-structured Populations, 2nd ed. Cambridge: Cambridge University Press; 1994. 44. Mueller L, Rose MR. Evolutionary theory predicts late-life mortality plateaus. Proc. Natl Acad. Sci. S.A. 1996; 93 (15): 249-253. 45. Charlesworth B, Partridge L. Ageing: Leveling of the grim reaper. Curr. Biol. 1997; 7: R440-R442. 46. Pletcher SD, Curtsinger JW. Mortality plateaus and the evolution of senescence: why are old-age mortality rates so low? Evolution 1998; 52: 454-464. 47. Wachter, K.W. 1999. Evolutionary demographic models for mortality plateaus. Proc. Natl Acad. Sci. S.A. 1999; 96 (10): 544-10 547. 48. Ali M, de Muynck A. Illness incidence and health seeking behaviour among street children in Rawalpindi and Islamabad, Pakistan a qualitative study. Child: Care, Health & Development 2005; 31(5):525-532. 49. Chevannes B. Learning to be a man: Culture, socialization and gender identity in five Caribbean communities. Kingston, Jamaica: University of the West Indies Press; 2001. 50. Williams RE, Black CL, Kim H-Y, Andrews EB, Mangel AW, Buda JJ, Cook SF. Determinants of health-care-seeking behaviour among subjects with irritable bowel syndrome. Alimentary Pharmacology & Therapeutics 2006; 23(11):1667-1675. 51. Dunlop DD, Manheim LM, Song J, Chang RW. Gender and ethnic/racial disparities in health care utilization among older adults. J of Gerontology: Soci Sci 2002; 57B (3): S221-S233. 52. Fujiwara T, Kawachi I. Is education causally related to better health? A twin fixed-effect study in the USA. Int J Epidemiol 2009;1-13 53. Gore WR. Sex, marital status, and mortality. American Journal of Sociology 1973; 30:189-208. 54. Koo, J., J. Rie, and K. Park. 2004. Age and gender differences in affect and subjective wellbeing. Geriatrics and Gerontology International, 4:S268-S270.
149

55. Smith KR, Waitzman NJ. Double jeopardy: interaction effects of marital and poverty status on risk of mortality. Demography 1994; 31:487-507. 56. Ross, C. E., J. Mirowsky, and K. Goldsteen. 1990. The impact of the family on health. Journal of Marriage and the Family 52:1059-1078. 57. Delbs, C., and J. Gaymu. 2002. The shock of widowed on the eve of old age: Male and female experience. Demography 3: 885-914. 58. Bourne PA. Self-evaluation of health of married people in Jamaica. (in review)

150

Figure 6.1: Illness (%) by age group

151

Figure 6.2: Area of residence of those with Injury (%) and Illness (%) controlled for by sex

152

Figure 6.3. Sex composition of those who attend health care facilities

153

Table 6.6.1: Treatment for Gunshot wounds at the Accident and Emergency Depts. Of Public Hospitals by Gender and Age cohort (in %): 1999-2002 Age cohort 1999 Male Female < 5 years 5-9 years 10-19 years 20-29 years 30-44 years 45-64 years 65+ years 0.8 0.3 17.9 39.0 30.6 6.6 3.5 1.3 3.0 24.5 32.5 23.6 12.2 3.0 0.0 2000 Male Female 0.2 0.7 16.2 40.5 31.1 6.7 2.3 2.2 3.1 1.9 18.5 30.2 11.1 28.4 11.1 1.2 Year 2001 Male Female 0.2 0.3 10.2 35.8 32.3 10.7 6.7 3.8 0.0 1.1 17.0 19.4 26.9 22.3 12.7 0.7 2002 Male Female 0.0 0.3 13.9 36.6 29.3 8.9 8.8 2.3 0.0 0.6 17.0 35.2 32.1 11.3 3.6 0.6

Not unknown 1.4

Total % 100 100 100 100 100 100 100 100 Calculated by Paul A. Bourne from Annual Report, 2002 published by the Policy, Planning and Development Division, Ministry of Health, Jamaica

154

Table 6.6.2: Visitation to the Accident and Emergency Depts. Of Public Hospitals for attempted suicide by Gender and Age cohort (in %): 2000-2002 Age cohort 2000 Male Female < 5 years 5-9 years 10-19 years 20-29 years 30-44 years 45-64 years 65+ years Not unknown 0.0 0.0 19.0 24.1 34.5 12.1 6.9 3.4 0.0 3.4 39.3 36.0 13.5 2.2 3.4 2.2 Year 2001 Male Female 1.0 2.0 13.0 20.0 13.0 4.0 4.0 0.0 0.0 0.0 49.4 34.8 6.7 3.4 2.2 3.4 2002 Male Female 1.0 2.0 13.0 20.0 13.0 4.0 4.0 1.7 0.9 3.5 38.3 36.5 17.4 0.9 0.0 2.6

Total % 100 100 100 100 100 100 Calculated by Paul A. Bourne from Annual Report, 2002 published by the Policy, Planning and Development Division, Ministry of Health, Jamaica

155

Table 6.6.3: Victims of Major Crimes by Age Cohorts, 2005


Age Group Carnal Abuse Female 3 15 223 103 0 0 0 0 0 0 0 0 2 346 346

Age Group Male 0-4 5-9 10-14 15-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55 & Over Unknown Total Total Reported 2 3 10 122 268 252 223 177 139 72 46 81 91 1486

Murder Female 4 5 8 18 23 33 22 17 12 16 9 16 5 188 Total 6 8 18 140 291 285 245 194 151 88 55 97 96 1674 1674 Male 3 0 4 107 212 192 161 138 107 68 46 50 408 1496

Shooting Female 1 5 11 13 30 22 16 15 15 5 8 6 3 150 Total 4 5 15 120 242 214 177 153 122 73 54 56 411 1646 1646 Male 0 1 16 59 162 233 198 199 171 146 98 152 28 1463

Robbery Female 0 0 11 49 115 130 112 102 77 44 32 66 9 747 Total 0 1 27 108 277 363 310 301 248 190 130 218 37 2210 2210 Male 0 0 0 8 52 81 114 140 116 98 75 171 32 887

Breaking Female 0 0 6 17 75 106 115 104 107 75 47 100 14 766 Total 0 0 6 25 127 187 229 244 223 173 122 271 46 1653 1653

Rape Female 3 27 212 223 122 48 28 23 17 12 7 16 8 746 746

Source: Statistics department, Jamaica Constabulary Force


156

Table 6.6.4: Age Group of Persons Arrested for Major Crimes for 2005
Age Group Murder Age Group 12-15 16-20 21-25 26-30 31-35 36-40 41-45 46-50 51-55 56-60 61& Over Unknown Total Male 6 157 235 160 85 54 15 7 5 1 0 40 765 Female 1 6 8 2 1 3 0 1 1 0 0 0 23 Total 7 163 243 162 86 57 15 8 6 1 0 40 788 Shooting Male 4 167 239 137 74 40 12 2 0 1 2 86 764 Female 0 1 1 0 0 1 0 0 0 0 0 0 3 Total 4 168 240 137 74 41 12 2 0 1 2 86 767 Robbery Male 10 183 214 120 71 36 13 1 2 6 2 23 681 Female 0 0 1 1 1 1 0 0 0 0 0 0 4 Total 10 183 215 121 72 37 13 1 2 6 2 23 685 Breaking Male 54 122 129 105 93 69 44 18 2 1 3 11 651 Female 0 3 3 3 2 0 1 0 0 1 1 0 14 Total 54 125 132 108 95 69 45 18 2 2 4 11 665 Rape Male 12 66 68 73 48 23 18 12 3 1 2 10 336 C/Abuse Male 11 43 52 27 26 19 12 5 2 1 0 0 198 Total 98 748 950 628 401 246 115 46 15 12 10 170 3439

Source: Statistics department, Jamaica Constabulary Force

157

Table 6.6.5. Particular variables by sex of respondents Variable Injury Yes No Illness Yes No Self-assessed health status Very good Good Moderate Poor Very poor Health care-seeking behaviour Yes No Household head Yes No Union status Married Common-law Visiting Single Not stated Self-reported diagnosed health condition Acute: Influenza Diarrhoea Respiratory Chronic: Diabetes Hypertension Arthritis Other (unspecified) Area of residence Urban Peri-urban Rural No. of visits to health care facilities Mean (SD) Age Mean (SD) Medical expenditure Mean (SD) in US $
US$ 1.00 = Ja. $ 80.47

Male (%) n = 1,439

Sex Female (%) n = 1,585 0.6 99.6 10.5 89.5 39.8 47.4 10.4 2.2 0.2

P 0.037

1.4 98.6 5.3 94.7 44.8 47.5 6.3 1.1 0.4 3.5 96.5 34.1 65.9 12.1 19.7 28.1 30.8 9.3 12.9 6.5 16.1 6.5 11.3 1.6 45.2

< 0.0001 < 0.0001

< 0.0001 6.8 93.2 < 0.0001 73.0 27.0 0.103 15.1 21.0 26.2 28.9 8.7 0.289 12.7 1.4 13.4 8.5 21.1 3.5 39.4 0.756 32.2 31.9 21.4 22.5 46.4 45.6 1.2 (0.5) 1.5 (1.3) 28.4 yrs (8.8) 28.5 yrs (8.9) 16.67 (42.01) 16.42 (26.82)

0.144 0.746 0.971

158

Table 6.6. Logistic regression: Explanatory variables of good health status, n = 2, 832 Explanatory variable Social determinants: Age Crowding Tertiary Primary Male MiddleClass Poor classes Married Single Other town Rural Medical expenditure Health care- seeking Biological condition: Self-reported illness Std. Error 0.01 0.03 0.28 0.14 0.18 0.21 0.18 0.00 0.29 0.24 Odds ratio 0.97 0.95 1.47 1.00 1.45 1.45 1.00 1.63 1.00 1.61 1.00 0.99 0.35 0.17 95.0% C.I. 0.96-0.99 0.90-1.00 1.27-1.81 1.11-1.91 1.02-2.07 1.09-2.43 1.12-2.30 0.99-1.00 0.20-0.62 0.11-0.28 P < 0.0001 0.043 0.007 0.007 0.041 0.018 0.009 0.017 < 0.0001 < 0.0001 R2 0.004 0.003 0.003 0.006 0.003 0.004 0.005 0.006 0.009 0.153

Hosmer and Lemeshow goodness of fit 2 = 4.4 (8), P = 0.82 -2LL = 1615.7 Nagelkerke R2 =0.196 Reference group

159

7
Disparities in self-rated health, health care utilization, illness, chronic illness and other socioeconomic characteristics of the Insured and Uninsured
This study examines self-rated health status, health care utilization, income distribution, and health insurance status of Jamaicans, and the disparity by the insured and uninsured. It also models self-rated health status, health care utilization, income distribution, and how these differ between the insured and uninsured. The majority (61.1%) of those who reported being diagnosed with a chronic condition were 60+ years old (diabetes mellitus, 59.3%; hypertension, 60.2%; arthritis, 67.9%) and 2.4% were children. The mean age of those with chronic illness was 62.3 years (SD = 16.2), and this was 61.5 years (SD = 16.5) for the uninsured and 63.8 years (SD = 15.8) for those with insurance coverage. Only 20.2% of respondents had health insurance coverage (private, 12.4%; NI Gold, public, 5.3%; other public, 2.4%). Most of the chronically ill were uninsured (67%). More people with chronic illnesses who had health insurance coverage were elderly, (65.9%), compared to uninsured chronically ill elderly (58.4%). Majority of health insurance was owned by those in the upper class, (65%), and 19%, by those in the lower socioeconomic strata. Insured respondents were 1.5 times (Odds ratio, OR, 95% CI = 1.06 2.15) more likely to rate their health as moderate-to-very good compared to the uninsured, and they were 1.9 times (95% CI = 1.31-2.64) to seek more medical care, 1.6 times (95% CI = 1.022.42) more likely to report having chronic illness, and more likely to have greater income ( = 0.094) than the uninsured. Illness is a strong predictor of why Jamaicans seek medical care (R2 = 71.2% of 71.9%), and health insurance coverage accounted for less than one-half percent of the variance in health care utilization. However, health care utilization is a strong predictor of self-reported illness, but it was weaker than illness explaining health care utilization (61.1% of 66.5%). Public health insurance was mostly had by those with chronic illnesses (76%) compared to 44% private health coverage and 38% had no coverage (2 = 42.62, P < 0.0001). With the health status of the insured being 1.5 times more than the uninsured, their health care utilization being 1.9 times more than the uninsured and illness being a strong predictor of health care seeking, any reduction in the health care budget in developing nations denotes that vulnerable groups (such as elderly, children and the poor) will seek less care, and this will further increase the mortality among those cohorts.

160

Introduction This study examines self-rated health status, health care utilization, income distribution, and health insurance status of Jamaicans, and the disparity between the insured and uninsured. It also models self-rated health status, health care utilization, income distribution, and how these differ between the insured and uninsured. The current findings revealed that 20.2% of Jamaicans had health insurance coverage (i.e. 2,140,316 Jamaicans are uninsured, using end of year population for 2007), suggesting that a large percent of the population are having to use out of pocket payment or governments assistance to pay their medical bills. The health of individuals within a society goes beyond the individual to the socioeconomic development, standard of living, production and productivity of the nation. Individuals health is therefore the crux of humans development, survivability and explains the rationale as to why people seek medical care on the onset of ill-health. In seeking to preserve life, people demand and utilize health care services. Western societies are structured that people meet health care utilization with a combination of approaches. These approaches can be any combination of out of pocket payment, health insurance coverage, government assistance and families aid. In Latin America and the Caribbean, health care is substantially an out of pocket expenditure aided by health insurance policy and governments health care policy. Within the context of the realities in those nations, the health of the populace is primarily based on the choices, decisions, responsibility and burden on the individual. Survival in developing nations are distinct from Developed Western Nations as Latin America and Caribbean peoples willingness, frequency, and demand for health care as well as health choices are based on affordability. Affordability of health care is assisted by health insurance coverage; as the
161

provisions of care offered by the governmental policies mean that the public health care system will be required to meet the needs of many people. Those people will be mostly children, elderly and other vulnerable groups. The public health care system in many societies often time involve long queues, long waiting times, frustrated patients and poor people who are dependent on the service. In order to circumvent the public health care system, people purchase health insurance policies as a means of reducing futuristic health care cost as well as an avoidance of the utilization of public health care. Uninsurance in any society means a dependency on the public health care system, premature mortality and oftentimes public humiliation. The insured on the other hand are able to circumvent many of the experiences of the poor, elderly, children and other vulnerable cohorts who rely on public health care system. Insurance in developing nations, and in particular Jamaica, is private system between the individual and a private insurance company. Because of the nature of health insurance and insurance, people buy into a pool which is usually accommodated through employment. Such a reality excludes retired elderly, unemployed, unemployable, and children of those cohorts. In seeking to understand health care non-utilization and high mortality in developing nations, insurance coverage (or lack of) becomes crucial in any health discourse. There is high proportion of uninsured in the United States and this is equally the reality in many developing nations, particularly in Jamaica [1-6]. According to the World Health Organization (WHO), 80% of chronic illnesses were in low and middle income countries, and 60% of global mortality is caused by chronic illnesses [7]. It can be extrapolated from the WHOs findings that

162

uninsurance is critical in answering some of the health disparities within and among groups and the sexes in the society. The realities of the health inequalities between the poor and the wealthy and the sexes in a society and those in the lower income strata having more illnesses and in particular chronic conditions [7-12] is embedded in financial deprivation. The WHO stated that In reality, low and middle income countries are at the centre of both old and new public health challenges [7]. The high risk of death in low income countries is owing to food insecurity, low water quality, low sanitation coupled with in access to financial resources [11, 13]. Poverty makes it insurmountable for poor people to respond to illness unless health care services are free. Hence, the people who are poor will suffer even more so from chronic diseases. The WHO captures this aptly ...People who are already poor are the most likely to suffer financially from chronic diseases, which often deepens poverty and damage long term economic prospects [7]. This goes back to the inverse correlation between poverty and higher level education, poverty and non-access to financial resources, and now poverty and illness. According to the WHO [7], In Jamaica 59% of people with chronic diseases experienced financial difficulties because of their illnesses... and emphasize the importance of health insurance coverage and the public health care system for vulnerable groups. Previous studies showed that health insurance coverage is associated with health care utilization [1-6], and this provides some understanding of health care demand (or the lack of) in developing countries. Studies have been conducted on the general health of the insured and/or uninsured, health care utilization and other health related issues [1-6] have used a piecemeal approach, which means that there is a gap in the literature that could provides more insight into the insured and uninsured. While the current body of health literature provide pertinent

163

information on health and health care utilization and how these differ based on the insured and uninsured, health choices are complex and requires more than piecemeal inquiry.

Materials and methods Data methods This study is based on data from the 2007 Jamaica Survey of Living Conditions (JSLC), conducted by the Planning Institute of Jamaica (PIOJ) and the Statistical Institute of Jamaica (STATIN). The JSLC is an annual and nationally representative cross-sectional survey that collects information on consumption, education, health status, health conditions, health care utilization, health insurance coverage, non-food consumption expenditure, housing conditions, inventory of durable goods, social assistance, demographic characteristics and other issues [14]. The information is from the civilian and non-institutionalized population of Jamaica. It is a modification of the World Banks Living Standards Measurement Study (LSMS) household survey [15]. Overall, the response rate for the 2007 JSLC was 73.8%. Over 1994 households of individuals nationwide are included in the entire database of all ages [16]. A total of 620 households were interviewed from urban areas, 439 from other towns and 935 from rural areas. This sample represents 6,783 non-institutionalized civilians living in Jamaica at the time of the survey. The JSLC used complex sampling design, and it is also weighted to reflect the population of Jamaica.

164

Statistical analysis Statistical analyses were performed using the Statistical Packages for the Social Sciences v 16.0 (SPSS Inc; Chicago, IL, USA) for Windows. Descriptive statistics such as mean, standard deviation (SD), frequency and percentage were used to analyze the socio-demographic characteristics of the sample. Chi-square was used to examine the association between nonmetric variables, and an Analysis of Variance (ANOVA) was used to test the equality of means among non-dichotomous categorical variables. Means and frequency distribution were considered significant at P < 0.05 using chi-square, independent sample t-test, and analysis of variance f test, multiple logistic and linear regressions. Analytic Models Cross-sectional analyses of the 2007 JSLC were performed to compare within and between subpopulations and frequencies. Logistic regression examined the relationship between the dichotomous binary dependent variable and some predisposed independent (explanatory) variables. A pvalue < 0.05 was selected to established statistical significance. Analytic models, using multiple logistic and linear regressions, were used to ascertain factors which are associated with (1) self-rated health status, (2) health care utilization, (3) selfreported illness, (4) self-reported diagnosed chronic illness, and income. For the regressions, design or dummy variables were for all categorical variables (using the reference group listed last). Overall model fit was determined using log likelihood ratio statistic, odds ration and rsquared. Stepwise regressions were used to determine the contribution of each significant variable. All confidence interval (CIs) for odds rations (ORs) were calculated at 95%.
165

Results Demographic characteristic of sample The sample was 6,783 respondents (48.7% males and 51.3% females). Children constituted 31.3%; other aged adults, 31.3%; young adults, 25.9%; and elderly, 11.9%. The elderly comprised 7.7% young-old, 3.2% old-old and 1.0% oldest-old. Majority of the sample had no formal education (61.8%); primary, 25.5%; secondary, 10.8% and tertiary, 2.0%. Two-thirds of the sample sought health in the last 4-weeks; 69.2% were never married; 23.3% married; 1.7% divorced; 0.9% separated and 4.9% were widowed respondents. Almost 15% reported an illness in the last 4-weeks (43.3% had chronic conditions, 30.4% had acute conditions and 26.3% did not specify the condition). Of those who reported an illness in the last 4- weeks, 87.9% provided information on the typology of conditions: cold, 16.7%; diarrhea, 3.0%; asthma, 10.7%; diabetes mellitus, 13.8%; hypertension, 23.1%; arthritis, 6.3%; and specified conditions, 26.3%. Marginal more people were in the upper class (40.3%) compared to the lower socioeconomic strata (39.8%). Only 20.2% of respondents had health insurance coverage (private, 12.4%; NI Gold, public, 5.3%; other public, 2.4%). Majority of health insurance was owned by those in the upper class (65%) and 19% by those in the lower socioeconomic strata. Bivariate analyses Sixty-one percent of those with chronic conditions were elderly compared to 16.6% of those with other conditions (including acute ailments). Only 39% of those with chronic conditions were non-elderly compared to 83.4% of those with other conditions (2 = 187.32, P < 0.0001). Thirty-three percent of those with chronic illnesses had health insurance coverage compared to 17.8% of those with acute and other conditions - (2 = 26.65, P < 0.0001).
166

Furthermore examination of self-reported health conditions by health insurance status revealed that diabetics recorded the greatest percent of health insurance coverage (43.9%) compared to hypertensive, (28.2%); arthritic (25.5%); acute conditions patients (17.0%) and other health conditions respondents (18.8%). Sixty-seven percent of respondents who reported being diagnosed with chronic conditions sought medical care in the last 4-weeks compared to 60.4% of those with acute and other conditions (2 = 4.12, P < 0.042). Those with primary or below education were more likely to have chronic illnesses (45.0%) compared to secondary level (6.1%) and tertiary level graduants (11.1%) - (2 = 23.50, P < 0.0001). There was no

statistical association between typology of illness and social class - (2 = 0.63, P = 0.730): upper class, 44.6%; middle class, 41.1% and lower class, 43.0%. This study found significant statistical association between health insurance status and (1) educational level (2 = 45.06, P < 0.0001), (2) social class (2 = 441.50, P < 0.0001), and (3) age cohort (2 = 83.13, P < 0.0001). Forty-two percent of those with at most primary level education had health insurance coverage compared to 16.3% of secondary level and 42.2% of tertiary level respondents. Thirty-three percent of upper class respondents had health insurance coverage compared to 16.7% of those in the middle class and 9.4% of those in the lower socioeconomic strata. Almost 33% of the oldest-old had health insurance coverage compared to 15.1% of children; 18.4% of young adults; 23.6% of other aged- adults; 28.6% of young-old and 24.9% of old-old. A significant statistical association was found between health insurance status and area of residence (2 = 138.80, P < 0.0001). Twenty-eight percent of urban dwellers had health insurance coverage compared to 22.1% of semi-urban respondents and 14.5% of rural residents. Furthermore, similarly a significant relationship existed between health care seeking behaviour and health insurance status (2 = 33.61, P < 0.0001). Fourteen percent of those with health
167

insurance sought medical care in the last 4-weeks compared to 9.0% of those who did not have health insurance coverage. Likewise a statistical association was found between health insurance status and typology of illness (2 = 26.65, P < 0.0001). Fifty-eight percent of those with insurance coverage had chronic illnesses compared to 38.3% of those without health insurance. Concurringly, 42% of those with insurance coverage had acute or other conditions compared to 62% of those who did not have health insurance coverage. Further examination revealed that other public health insurance was mostly had by those with chronic illnesses (76%) compared to NI Gold (public, 65%) and 44% private health coverage (2 = 42.62, P < 0.0001). Private health coverage was most had by those with non-chronic illnesses (56%) compared to 35% with NI Gold (public) and 25% other public coverage. No significant statistical difference was found between the average medical expenditure of those who had insurance coverage and non-insured (t = 0.365, P = 0.715) mean average medical expenditure of those without health insurance was USD 10.68 (SD = 33.94) and insured respondents mean average medical expenditure was USD 9.93 (SD = 18.07) - (Ja. $80.47 = US $1.00 at the time of the survey). There was no significant statistical relationship between health care utilization (publicprivate health care visits) and health conditions (acute or chronic illnesses) 2 = 0.001, P = 0.975. 49.2% of those who had chronic illnesses used public health care facilities compared to 49.3% of those with acute conditions. There is a statistical difference between the mean age of respondents with non-chronic and chronic illnesses (t = - 23.1, P < 0.0001). The mean age of some with chronic illnesses was 62.3 years (SD = 16.2) compared to 29.3 years (SD = 26.1) for those with non-chronic illnesses. Furthermore, the mean age of insured respondents with chronic illnesses was 63.8 years (SD =
168

15.8) compared to 32.5 years for those with non-chronic conditions. Concurringly, uninsured chronically ill respondents mean age was 61.5 years (SD = 16.5) compared to 28.6 years (SD = 25.9) for those with non-chronic illnesses. Table1 examines information on crowding index, total annual food expenditure, annual non-food expenditure, income, age, time in household, length of marriage, length of illness and number of visits made to medical practitioner by health insurance status. Self-rated health status, health care seeking behaviour, illness, educational level, social class, area of residence, and health conditions, health care utilization by health insurance status are presented in Table 7.2. Table 7.3 presents information on age cohort of respondents by diagnosed health conditions. A significant statistical association was found between the two variables 2 = 436.8, P < 0.0001. Table 7.4 examines illness by age of respondents controlled for by health insurance status. There existed a significant statistical relationship between illness and age of respondents, but none between the uninsured and insured, P = 0.410. Table 7.5 presents information on the age cohort by diagnosed health conditions, and diagnosed health conditions controlled by health status. There is a statistical difference between the mean age of respondents and the typology of self-reported illnesses (F = 99.9, P < 0.0001). Those with cold, 19.2 years (SD = 23.9); diarrhoea, 30.3 years (SD = 31.4); asthma, 22.9 years (SD = 22.1); diabetes mellitus, 60.9 years (SD = 16.0); hypertension, 62.5 years (SD = 16.8); arthritis, 64.3 years (SD = 14.5), and other conditions, 38.3 years (SD = 25.3).

169

Analytic Models Nine variables account for (Table 7.6), 32.8% of the variance in moderate-to-very good selfrated health status of Jamaicans The variables are medical expenditure, health insurance status, area of residence, household head, age, crowding index, total food expenditure, health care utilization and illness. Self-reported illnesses accounted for 62.2% of the explained variability of moderate-to-very good health status. Table 7.7 shows information on the explanatory factors of self-reported illnesses. Seven factors accounted for 66.5% of the variability in self-reported illnesses. Ninety-two percent of the variability in self-reported illnesses was accounted for by health care utilization (health care seeking behaviour). Three variables emerged as statistically significant correlates of health care utilization. They accounted for 71.9% of the variance in health care utilization. Most of the variability can be explained by self-reported illnesses (71.2%, Table 7.8). Self-reported diagnosed chronic illnesses can be explained by 5 variables (gender, marital status, health insurance status, age and length of illness), and they accounted for 27.7% of the variance in self-reported diagnosed chronic illness (Table 7.9). Sixty-two percent of the variability in income can be explained by crowding index, social class, household head, health insurance status, self-rated health status, health care utilization, area of residence and marital status). Most of the variability in income can be explained by social class (Table 7.10). Table 7.11 presents information on the explanatory variables which account for health insurance coverage. Six variables emerged as significant determinants of health insurance coverage (age, income, chronic illness, health care utilization, marital status and upper
170

socioeconomic class). The explanatory variables accounted for 19.4% of the variability in health insurance coverage. Income was the most significant determinant of health insurance coverage (explained 43% of the explained variance, 19.4%).

Discussion The current paper revealed that 15 out of every 100 Jamaicans reported having an illness in the last 4-weeks, and 57% of those with an illness had chronic conditions. Sixty-one out of every 100 of those with chronic illnesses were 60+ years; 67% of the chronically ill sought medical care when compared to 66% of the population. Most of the chronically ill respondents were uninsured (67%). The chronically ill had mostly primary level education, and there was no statistical association between typology of illness and social class. Almost 2 in every 100 chronically ill Jamaicans were children (less than 19 years), and most of them were uninsured. Nine percent more of the chronically ill who the other aged adult cohort did not have health insurance coverage. Insured respondents were 1.5 times more likely to rate their health as moderate-to-very good compared to the uninsured, and they were 1.9 times more likely to seek more medical care, 1.6 times more likely to report having chronic illnesses, and more likely to have greater income than the uninsured. Illness is a strong predictor of why Jamaicans seek medical care (R2 = 71.2% of 71.9%), and health insurance coverage accounted for less than onehalf percent of the variance in health care utilization. However, health care utilization is a strong predictor of self-reported illness, but it was weaker than illness explaining health care utilization (61.1% of 66.5%). Public health insurance was most common among those with chronic illnesses (76%) compared to 44% private health coverage and 38% had no coverage. Those in the upper

171

income stratas income was significant more than those in the middle and lower socioeconomic group, but chronic illnesses were statistically the same among the social classes. Health disparities in a nation are explained by socioeconomic determinants as well as health insurance status. Previous research showed that health care utilization and health disparities are enveloped in unequal access to insurance coverage and social differences [2, 4, 17-19]. The present paper revealed that health insurance coverage is mostly had by those in the upper class, with less than 20 in every 100 insured being in the lower socioeconomic class. Although this study found that those in the lower class does not have more chronic illness than those in the wealthy class, 86 out of every 100 uninsured respondents indicated that their health status was poor. Health insurance coverage provides valuable economic relief for chronically ill respondents as this allows them to access needed health care. Like Hafner-Eatons research [2], this paper found that health insurance status was the third most powerful predictor of health care utilization. Forty-nine to every 100 chronically ill persons use the public health care facilities. This mean that health insurance coverage appeases the health care burden of its holder, but the insured in Jamaica are mostly wealthy, older, chronically ill, married, and seek more medical care than the uninsured. The uninsured ill are therefore less likely to demand health care, and this economic burden of health care is either going to be the responsibility of the state, the individual or the family. The difficulty here is that the uninsured are more likely to be in the lower-tomiddle class, of working age or children, experienced more acute illness, 38 out of every 100 chronically ill are in the lower class, these provide a comprehensive understanding of the insured and uninsured that will allow for explanations in health disparities between the socioeconomic strata and sexes. With 43 out of every 100 people in the lower socioeconomic strata self-reported
172

being diagnosed with chronic illness, health insurance coverage, public health system and other policy intervention aid in their health, and health care utilization. Among the material deprivation of the poor is uninsurance. Those in the wealthy socioeconomic group in Jamaica were 3.5 times more likely to be holder of health insurance coverage than those in the lower socioeconomic strata. And Gertler and Sturm [3] identified that health insurance cause a switching from public health to the private health system, which indicates that a reduction in public health expenditure and health insurance will significantly influence the health of the poor. This research showed that only 19% of those with health insurance were in the lower class. Therefore issue of uninsurance creates futuristic challenges for the poor in regard to their health and health care utilization. As on the onset of illness, those in the lower income strata without health insurance must first think about their illness and weight this against the cost of losing current income in order to provide for their families as well as parents of ill children must also do the same. The public health care system will relieve the burden of the poor, and while those with health insurance are more likely to utilize health care, this is a futuristic product in enhancing a decision to utilize health care. But outside of those issues, their choices (or lack), the cost of public health care, national insurance scheme and general price index in the society further lowers their quality of life. Although the poor may be dissatisfied with the public health care system (waiting time, crowding, discriminatory practices by medical practitioners), better health for them without health coverage is through this very system. It can be extrapolated therefore from the present data that there are unmet health needs among some people in the lower socioeconomic strata. As those who do not have health insurance, want to avoid the public health care system owing to dissatisfaction or inafffordability, and will only seek health care when their symptoms are severe and sometimes
173

the complications from the delay make it difficult to be addressed on their visits. Among unmet health needs of the poor will be medication. Even if they attend the public health care system and are treated, the system does not have all the medications which is an indication that they are expected to buy some. The challenge of the poor is to forego purchasing medication for food, and this means their conditions would not have been rectified by the health care visitation. By their very nature, the socioeconomic realities of the poor such as lower access to education, proper nutrition, good physical milieu, poor sanitation and lower health coverage, cripple their future health status, this accounts for high premature mortality and hinders health care utilization. It is this lower health care utilization which accounts for their increase risk of mortality as the other deprivations such as proper sanitation and nutrition exposes them to disease causing pathogens which means that their inability to afford health insurance increased their reliance on the public health care system. The present findings showed that the uninsured are mostly poor and within the context of Lasser et al.s work [20] that they receive worse access to care, are less satisfied with the care they receive and medical services than the insured in the US, this is an indication of further resistant of the poor from willingly demanding health care as this rehashes their dissatisfaction and humiliation. Despite the dissatisfaction and humiliation, their choices are substantially the public health care system, abstinence from care, risk of death, and the burden of private health care. Apart of the rationales why those in the lower socioeconomic strata have fewer health coverage than those in the wealthy income group are (1) inafffordability, (2) type of employment (mostly part time, seasonal, low paid and uninsured position) which makes it too difficult for them to be holders of health insurance and this retards the switch from public-to-private health care utilization. Recently a study conducted by Bourne and Eldemire-Shearer [21] found that 74% of those in the poorest income quintile utilized public
174

hospitals compared to 58% of those in the second poor quintile and 31% of those in the wealthiest 20%. Then, if public health becomes privatized or become increasingly more expensive for recipients, the socioeconomically disadvantaged population (poor, elderly, children and other vulnerable groups) will become increasingly exposed to more agents that are likely to result in their deaths, increased utilization of home remedy as well as the widening of the health outcome inequalities among the socioeconomic strata. Illness and particularly chronic condition can easily result in poverty, before mortality sets in. With the World Health Organization (WHO) opined that 80% of chronic illnesses were in low and middle income countries and that 60% of global mortality is caused by chronic illness [7], leveling insurance coverage can reduce burden of care for those in the lower socioeconomic strata. The importance of health insurance to health care utilization, health status, productivity, production, socioeconomic development, life expectancy, poverty reduction strategy and health intervention must include increase health insurance coverage of citizenry within a nation. The economic cost of uninsured people in a society can be measured by the lost of production, payment of sick time, mortality, lowered life expectancy and cost of care for children, orphanage and elderly who become the responsibility of the state from the death of the poor. Therefore the opportunity cost of reduced public health care budget is the economic cost of the aforementioned issues, and goes to the explanation of premature mortality in a society. Particularly the chronically ill, they benefit from health insurance coverage not because of the reduced cost of health care, but the increased health care utilization that result from health coverage. From the findings of Hafner-Eatons work [2], the chronically ill in the United States were 1.5 times more likely to seek medical care and while this is about the same for Jamaicans, health insurance is responsible to their health care utilization and not the condition or illness.
175

According to Andrulis [22], Any truly successful, long-term solution to the health problems of the nation will require attention at many points, especially for low-income populations who have suffered from chronic underservice if not outright neglect Embedded in Andruliss work is the linkage between poverty, poor health care service delivery, differences in health outcomes among the socioeconomic groups, higher mortality among particular social class, deep-seated barriers in health care delivery and the perpetuation of those and how they can increase health differences among the socioeconomic strata. The relationship between poverty and illness is well established in the literature [7, 8, 23] as poverty means deprivation from proper nutrition, safe drinking water, and those issues contribute to lower health, production, productivity, and more illness in the future. Free public health care or lower public health care cost does not mean equal opportunity to access, eliminate the barriers to equal opportunity, neither does it increase health and wellness for the poor and remove lower health disparities among the socioeconomic groups. However, lower-income, increase price indices, removal of government subsidy from public health care, increased uninsurance, lower health care utilization, increase poverty, premature mortality and lower life expectancy of the population and particular subpopulations. Increases in diseases (acute and chronic) are owing to lifestyle practices of people. Lifestyle practices are voluntary lifestyle choices and practices [24]. The poor are less educated, more likely to be unemployed, undernourished, deprived from financial resources, and their voluntary actions will be about survival and not diet, nutrition, exercise and other healthy lifestyle choice. Lifestyle choices such as diet, proper nutrition, and sanitation, safe drinking water are costly, which oftentimes occurs because of poverty, some people can afford to make these choices. It follows therefore that those in the lower socioeconomic stratas voluntary action will be unhealthy choices which are cheaper. Poverty therefore handicaps its people, and
176

predetermines unhealthy lifestyle choices, which further accounts for greater mortality, lower life expectancy, health insurance coverage and private health care utilization. Conclusion Poverty is among the social determinants of health, health care utilization, and health insurance coverage in a society. While the current paper does not support the literature that chronic illnesses were greater among those in the lower socioeconomic strata, they were less likely to have health insurance coverage compared to the upper class. Poverty denotes socioeconomic deprivation of resources which appears in a society, and goes to the crux of health disparities among the socioeconomic groups and sexes. Health care utilization is associated with health insurance coverage as well as governments assistance, and this embodies the challenges of those in the vulnerable groups. Within the current global realities, many governments are seeking to reduce their public financing of health care which would further shift the burden of health care to the individual, and this will even increase premature mortality among those in the lower socioeconomic strata. Governments in developing nations continue to invest in improving public health measures such as safe drinking water, sanitation, mass immunization) and the training of medical personnel, building clinics and hospitals and there is definite a need to include health insurance coverage to their public health measure as this will increase access to health care utilization. Any increase in health care utilization will be able to improve health outcome, reduce health disparities between the socioeconomic groups and the sexes that will see improvements in the quality of life of the poor.

177

In summary, with the health status of the insured being 1.5 times more than the uninsured, their health care utilization being 1.9 times more than the uninsured and illness being a strong predictor of health care seeking, any reduction in the health care budget in developing nations denotes that vulnerable groups (such as elderly, children and poor) will seek less care, and this will further increase the mortality among those cohorts.

Conflict of interest
The authors have no conflict of interest to report.

Disclaimer
The researchers would like to note that while this study used secondary data from the Jamaica Survey of Living Conditions, none of the errors in this paper should be ascribed to the Planning Institute of Jamaica or the Statistical Institute of Jamaica, but to the researchers.

178

References 1. Miles S, Parker K. Men, women, and health insurance. The New England Journal of Medicine 1997; 336:218-221. 2. Hafner-Eaton C. Physical utilization disparities between the uninsured and insured: Comparisons of the chronically ill, acutely ill, and well nonelderly populations. JAMA 1993; 269:787-792. 3. Gertler P, Sturm R. Private health insurance and public expenditures in Jamaica. Journal of Econometrics 1997; 77:237-257. 4. Doty MM, Holmgren AL. Unequal access: insurance instability among low-income workers and minority. Issue Brief (Common Fund) 729:1-6. 5. Bourne PA. Self-reported health and medical care-seeking behaviour of uninsured Jamaicans. North Am J Med Sci 2010; 2: 71-80. 6. Bourne P. Health insurance coverage in Jamaica: Multivariate analyses using two crosssection survey data for 2002 and 2007. Int J of Collaborative Research on Internal Medicine and Public Health 1:195-213. 7. World Health Organization, WHO. Preventing Chronic Diseases a vital investment. Geneva: WHO; 2005. 8. Van Agt HME, Stronks K, Mackenbach JP. Chronic illness and poverty in the Netherlands. Eur J of Public Health 2000; 10:197-200. 9. Fox J ed. Health inequalities in European Countries. Aldershot: Gower Publishing Company Limited; 1989. 10. Illsley R, Svenson PG, ed. Health inequalities in Europe. Soc Sci Med 1990; 31(special issue):223-420. 11. Sen A. Poverty: An ordinal approach to measurement. Econometricia 1979; 44, 219231. 12. Casas JA, Dachs JN, Bambas A. Health disparity in Latin America and the Caribbean: The role of social and economic determinants. In: Pan American Health Organisation. Equity and health: Views from the Pan American Sanitary Bureau, Occasional Publication No. 8. Washington DC; 2001: pp. 22-49. 13. Marmot M .The influence of Income on Health: Views of an Epidemiologist. Does money really matter? Or is it a marker for something else? Health Affairs 2002; 21:3146. 14. Planning Institute of Jamaica, (PIOJ), Statistical Institute of Jamaica, (STATIN). Jamaica Survey of Living Conditions, 1989-2007. Kingston: PIOJ, STATIN; 1989-2008. 15. World Bank, Development Research Group, Poverty and Human Resources. Jamaica Survey of Living Conditions, 1988-2000. Basic information. Washington: The World Bank; 2002. (September 2, 2009, at http://siteresources.worldbank.org/INTLSMS/Resources/33589861181743055198/3877319-1190214215722/binfo2000.pdf). 16. Statistical Institute Of Jamaica. Jamaica Survey of Living Conditions, 2007 [Computer file]. Kingston, Jamaica: Statistical Institute Of Jamaica [producer], 2007. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors]; 2008.
179

17. Hayward RA, Shapiro MF, et al. Inequalities in health services among insured Americans. Do working-age adults have less access to medical care than the elderly. N Engl J Med 1988;318:1507-12. 18. Fiscella K, Williams DR. Health disparities based on socioeconomic inequities: implications for urban health care. Acad Med 2004;79:1139-47. 19. LaVeist TA, Carroll T. Race of physician and satisfaction with care among AfricanAmerican patients. J Natl Med Assoc 2002; 94:937-43. 20. Lasser KE, Himmelstein DU, Woolhandler S. Access to care, health status, and health disparities in the United States and Canada: Results of a Cross-National PopulationBased Survey. Am J Public Health 200696:1300-1307. 21. Bourne PA, Eldemire-Shearer D. Public hospital health care utilization in Jamaica. Australian J of Basic and Applied Scie 2009; 3:3067-3080. 22. Andrulis DP. Access to care is the centerpiece in the elimination of socioeconomic disparities in health. Ann Intern Med 1998; 129:412-416. 23. Foster AD. Poverty and illness in low-income rural areas. The American Economic Review 1994; 84:216-220. 24. Barnekow-Bergkvist M, Hedberg GE, Janlert U, Jansson E. Health status and health behaviour in men and women at the age of 34 years. European J of Public Health 1998; 8:179-182.

180

Table 7.1. Crowding, expenditure, income, age, and other characteristics by health insurance status Health insurance status P Characteristics Non-insured Insured mean SD mean SD Crowding index 4.9 2.6 4.12.1 t = 10.32, < 0.0001 Total annual food expenditure1 3476.092129.97 3948.122257.97 t = - 6.81, < 0.0001 Annual non-food expenditure1 3772.913332.50 6339.405597.60 t = - 21.33, < 0.0001 1 Income 7703.625620.94 12374.899713.00 t = - 22.75, < 0.0001 Age (in year) 28.721.4 35.0 22.7 t = - 9.40, < 0.0001 Time in household (in years) 11.71.6 11.81.3 t = - 1.62, 0.104 Length of marriage 16.914.3 18.313.8 t = - 1.55, 0.122 Length of illness 14.751.1 14.136.2 t = - 0.217, 0.828 No. of visits to medical practitioner 1.41.0 1.51.2 t = - 0.659, 0.511
1

Expenditures and income are quoted in USD (Ja. $80.47 = US $1.00 at the time of the survey)

181

Table 7.2. Health, health care seeking behaviour, illness and particular demographic characteristics by health insurance status Health insurance status P Characteristic Coverage No coverage Private n (%) Public, NI Gold n (%) Other Public n (%) n (%) 2 = 42.62, P < 0.0001 Health conditions Acute and other 53 (56.4) 24 (34.8) 13 (24.5) 415 (61.7) Chronic 41 (43.6) 45 (65.2) 40 (75.5) 258 (38.3) 2 = 70.09, P < 0.0001 Health care seeking behaviour No 724 (89.3) 283 (81.3) 118 (75.2) 4735 (91.0) Yes 87 (10.7) 63 (18.2) 39 (24.8) 468 (9.0) 2 = 67.14, P < 0.0001 Illness No 699 (86.2) 272 (78.6) 101 (64.3) 4453 (85.8) Yes 112 (13.8) 74 (21.4) 56 (35.7) 736 (14.2) 2 = 78.10, P < 0.0001 Education level Primary and below 684 (84.4) 318 (92.2) 144 (91.7) 4536 (87.4) Secondary 80 (9.9) 23 (6.7) 9 (5.7) 577 (11.1) Tertiary 46 (5.7) 4 (1.2) 4 (2.5) 74 (1.4) 2 = 596.08, P < 0.0001 Social class Lower 78 (9.6) 135 (39.0) 31 (19.7) 2345 (45.1) Middle 111 (13.7) 80 (23.1) 27 (17.2) 1085 (20.9) Upper 622 (76.7) 131 (37.9) 99 (63.1) 1773 (34.1) 2 = 190.29, P < 0.0001 Area of residence Urban 373 (46.0) 106 (30.6) 63 (40.1) 1397 (26.8) Semi-urban 212 (26.1) 66 (19.1) 32 (20.4) 1091 (21.0) Rural 226 (27.9) 174 (50.3) 62 (39.5) 2715 (52.2) 2 = 67.14, P < 0.0001 Self-rated health status Poor 699 (86.2) 272 (78.6) 101 (64.3) 4453 (85.8) Moderate-to-excellent 112 (13.8) 74 (21.4) 56 (35.7) 736 (14.2) 2 = 30.06, P < 0.0001 Health care utilization Private 65 (79.3) 29 (47.5) 18 (46.2) 215 (46.8) Public 17 (20.7) 32 (52.5) 21 (53.8) 244 (53.2)
182

Table 7.3. Age cohort by diagnosed illness


Diagnosed illness Acute condition Cold Age cohort n (%) Children n (%) n (%) n (%) n (%) n (%) n (%) n (%) Diarrhoea Asthma Diabetes mellitus Chronic condition Hypertension Arthritis Other Total

97 (65.1)

13 (48.1)

51 (53.7)

3 (2.4)

0 (0.0)

0 (0.0)

54 (23.1)

218 (24.5)

Young adults

14 (94)

2 (7.4)

16 (16.8)

3 (2.4)

6 (2.9)

1 (1.8)

43 (18.4)

85 (9.6)

Other-aged adults

22 (14.8)

6 (22.2)

18 (18.9)

44 (35.8)

76 (36.9)

17 (30.4)

85 (36.3)

268 (30.1)

Young old

8 (5.4)

2 (7.4)

7 (7.4)

49 (39.8)

61 (29.6)

22 (39.3)

32 (13.7)

181 (20.3)

Old Elderly

8 (5.4)

3 (11.1)

2 (2.1)

19 (15.4)

49 (23.8)

14 (25.0)

13 (5.6)

108 (12.1)

Oldest Elderly Total

0 (0.0) 149

1 (3.7) 27

1 (1.1) 95

5 (4.1) 123

14 (6.8) 206

2 (3.6) 56

7 (3.0) 234

30 (3.4) 890

183

Table 7.4. Illness by age of respondents controlled for health insurance status Age of respondents Characteristic Uninsured Insured Mean SD Mean SD Illness Acute condition Cold 18.8 23.5 21.0 26.3 Diarrhoea 28.4 30.3 31.8 13.5 Asthma 21.0 21.7 29.4 22.9 Chronic condition Diabetes mellitus 58.7 16.1 63.8 15.4 Hypertension 62.1 17.3 63.6 15.7 Arthritis 64.0 13.3 65.0 18.7 Other condition 38.1 25.0 39.2 26.8 F statistic 73.1, P < 0.0001 23.3, P < 0.0001

184

Table 7.5. Age cohort by diagnosed health condition, and health insurance status Diagnosed health condition Acute Chronic n (%) Age cohort Children Young adults Other aged-adults Young-old Old-old Oldest-old Total n (%) Diagnosed health condition Acute Chronic Uninsured n (%) n (%) Acute Chronic Insured n (%) n (%)

Characteristic

215 (42.6) 3 (0.8) 75 (14.9) 10 (2.6) 131 (25.9) 137 (2.6) 49 (9.7) 132 (34.3) 26 (5.1) 82 (21.3) 9 (1.8) 21 (5.5) 505 385 2 = 317.5, P < 0.0001

183 (44.1) 1 (0.4) 32 (35.6) 2 (1.6) 58 (14.0) 6 (2.3) 17 (18.9) 4 (3.2) 110 (26.5) 100 (38.6) 21 (23.3) 37 (29.4) 37 (8.9) 82 (31.7) 12 (13.3) 50 (39.7) 20 (4.8) 55 (21.2) 6 (6.7) 27 (21.4) 7 (1.7) 15 (5.8) 2(2.2) 6 (4.8) 415 259 90 126 2 2 = 234.5, P < 0.0001 = 73.6, P < 0.0001

185

Table 7.6. Logistic regression: Explanatory variables of self-rated moderate-to-very good health
Explanatory variable Average medical expenditure Health insurance coverage (1= insured) Urban Other Rural Household head Age Crowding index Total food expenditure Health care seeking (1=yes) Illness Model fit 2 = 574.37, P < 0.0001 -2LL = 1477.76 Nagelkerke R2 = 0.328 Reference group ***P < 0.0001, **P < 0.01, *P < 0.05 Coefficient Std. error Odds ratio 95.0% C.I. R2 0.003 0.005 0.007 0.006

0.000 0.410 0.496 0.462

0.000 0.181 0.180 0.197

1.00* 1.51* 1.64** 1.59* 1.00 1.46* 0.96*** 0.86*** 1.00*** 0.51** 0.24***

1.00 -1.00 1.06 - 2.15 1.15 - 2.34 1.08 - 2.34

0.376 -0.046 -0.156 0.000 -0.671 -1.418

0.154 0.004 0.035 0.000 0.211 0.212

1.08 - 1.97 0.95 - 0.96 0.80 - 0.92 1.00 - 1.00 0.34 - 0.77 0.16 - 0.37

0.004 0.081 0.010 0.003 0.005 0.204

186

Table 7.7. Logistic regression: Explanatory variables of self-reported illness


Explanatory variable Average medical expenditure Male Married Age Coefficient Std Error Odds ratio 95.0% C.I.

R2
0.001 0.003 0.002 0.037 0.002 0.009 0.611

0.000 -0.467 0.527 0.031 0.000 -1.429

0.000 0.137 0.146 0.004 0.000 0.213 0.262

1.00* 0.63** 1.69*** 1.03*** 1.00** 0.24*** 342.11***

1.00 - 1.00 0.48 - 0.82 1.27 - 2.25 1.02 - 1.04 1.00 -1.00 0.16 -0.36 204.71 -571.72

Total food expenditure


Self-rated moderate-to-excellent health

5.835 Health care seeking (1=yes) Model fit 2 = 2197.09, P < 0.0001 -2LL = 1730.41 Hosmer and Lemeshow goodness of fit 2 = 4.53, P = 0.81 Nagelkerke R2 = 0.665 Reference group ***P < 0.0001, **P < 0.01, *P < 0.05

187

Table 7.8. Logistic regression: Explanatory variables of health care seeking behaviour
Explanatory variable Coefficient Std error Odds ratio 95.0% C.I. R2

Health insurance coverage (1= insured)


Self-reported illness Self-rated moderate-to-excellent health

0.620 5.913 -0.680

0.179 0.252 0.198

1.86** 369.92*** 0.51**

1.31 - 2.64 225.74 - 606.17 0.34 - 0.75

0.003 0.712 0.004

Model fit 2 = 1997.86, P < 0.0001 -2LL = 1115.93 Hosmer and Lemeshow goodness of fit 2 = 1.49, P = 0.48 Nagelkerke R2 = 0.719 Reference group ***P < 0.0001, **P < 0.01, *P < 0.05

188

Table 7.9. Logistic regression: Explanatory variables of self-reported diagnosed chronic illness
Explanatory variable Male Married Never married Coefficient -1.037 0.425 Std error 0.205 0.199 Odds ratio 0.36*** 1.53* 1.00 1.58* 1.05*** 1.13* 95.0% C.I. 0.24 - 0.53 1.04 - 2.26

R2 0.048 0.012

Health insurance coverage (1= insured)


Age Logged Length of illness

0.454 0.047 0.125

0.220 0.005 0.059

1.02 - 2.42 1.04 - 1.06 1.01 - 1.27

0.008 0.201 0.008

Model fit 2 = 136.32, P < 0.0001 -2LL = 673.09 Hosmer and Lemeshow goodness of fit 2 = 15.96, P = 0.04 Nagelkerke R2 = 0.277 Reference group ***P < 0.0001, **P < 0.01, *P < 0.05

189

Table 7.10. Multiple regression: Explanatory variables of income


Unstandardized Coefficients B 11.630 0.206 1.265 0.692 Std. Error 0.061 0.008 0.052 0.047 Standardized Coefficients Beta 95% CI 11.511 - 11.750 0.190 - 0.221 1.162 - 1.368 0.599 - 0.784

Explanatory variable Constant Crowding index Upper class Middle Class Lower class Household head

R2

0.625*** 0.649*** 0.347***

0.195 0.320 0.133

-0.181 0.137 0.165 0.109 0.145 0.130 0.075

0.038 0.042 0.040 0.039 0.046 0.049 0.038

-0.108*** 0.075** 0.094*** 0.063** 0.079** 0.063** 0.044*

-0.256 - -0.106 0.054 - 0.220 0.088 - 0.243 0.033 - 0.185 0.055 - 0.235 0.033 - 0.226 0.000 - 0.150

0.012 0.007 0.006 0.003 0.002 0.003 0.001

Health insurance coverage (1= insured)


Self-rated good health status

Health care seeking (1=yes)


Urban Other town Rural area Married Never married

F = 144.15, P < 0.0001 R2 = 0.682 Reference group ***P < 0.0001, **P < 0.01, *P < 0.05

190

Table 7.11. Logistic regression: Explanatory variables of health insurance status (1= insured)
Explanatory variable Age Income Chronic condition Health care seeking (1=yes) Married Never married Upper class Lower class Coefficient 0.014 0.000 0.563 0.463 0.647 Std. error 0.006 0.000 0.210 0.211 0.192 Odds ratio 1.01* 1.00*** 1.7** 1.59* 1.91** 95.0% C.I. 1.00 - 1.03 1.00 - 1.00 1.16 - 2.65 1.05 - 2.40 1.31 - 2.79 R2 0.040 0.082 0.013 0.010 0.024

0.841

0.227

3.46***

1.49 - 3.62

0.025

Model fit 2 = 95.7, P < 0.0001 -2LL = 686.09 Hosmer and Lemeshow goodness of fit 2 = 5.08, P =0.75 Nagelkerke R2 = 0.194 Reference group ***P < 0.0001, **P < 0.01, *P < 0.05

191

8
Good Health Status of Rural Women in the Reproductive Ages
Paul A. Bourne & Joan Rhule

Women are traditionally overrepresented among the poor and therefore in the long run, have less access to remuneration and health resources, including health insurance and social security services. Women are disadvantaged on some fundamental economic indicators such as unemployment and access to economic resources. In 2007 in Jamaica, for instance among the 124 500 unemployed persons in the labour force, 65.4 % were women (Planning Institute of Jamaica, 2008). Thus, women's health and the control that they can exercise over resources are key factors in achieving effectiveness, efficiency, and sustainability in health interventions. This study examined the good health status of rural women in the reproductive ages of 15 to 49 years. Having extensively reviewed the literature, this paper is the first study of its kind in Jamaica and will provide pertinent information on this cohort for the purpose of public health planning. Using logistic regression analyses, 6 variables emerged as statistically significant predictors of current good health status of rural women (i.e. ages 15 to 49 years) in Jamaica. These are social standing (two wealthiest quintile OR=0.524, 95%CI: 0.350,0.785); marital status (separated, divorced or widowed OR=0.382, 95%CI: 0.147, 0.991); health insurance (OR=0.041, 95%CI: 0.024, 0.069); negative affective psychological conditions (OR=0.951, 95%CI:0.704, 1.284); asset ownership (OR=1.089, 95%CI:1.015, 1.168) and age of respondents (OR+0.965, 95%CI:0.949, 0.982). Poverty is synonymous with rural area and women, and inspite of this reality majority of rural women in Jamaica ages 15 to 49 years reported current good health status. Wealth creates more access to financial and other resources, and makes a difference in nutritional intake, water and food quality as well as an explanation for better environmental conditions. In this study, wealth did not mean better health but that poor women had greater health status than their wealthy counterparts. Another interesting finding was that good health is inversely correlated with the ownership of health insurance coverage.

192

Introduction

Many studies have shown that there is a statistical relationship between health status and poverty (Murray, 2006; Marmot, 2002; Muller & Krawinkel, 2005; Bloom & Canning, 2003; Smith & Waitzman, 1994), standard of living (Pacione, 2003; Bourne, 2007a, 2007b), and other socioeconomic determinants (Grossman, 1972; Smith & Kington 1997; Bourne, 2009; Bourne & McGrowder, 2009 Benzeval et al, 2001) . According to Abel-Smith (1994), the influence of income on health decreases as the society shifts from lowers to higher levels of income. And this is in keeping with the findings that show an inverse relationship between income of a country and levels of mortality, and the reverse is equally true (Abel-Smith, 1994; Matsaganis, 1992). Other scholars have refined this association when they opined that it is inequalities of income within a country that explains higher mortality and not mere income (Cochrane et al, 1978). The use of mortality to assess health is primary because this is easily measurable unlike the use of morbidity which is a minimalists approach to the study of health (Grossman, 1972); but the latter still does not capture quality life expectancy and so is the former measure. The emphasis on income to provide explanation for health status without incooperating sanitation, education and lifestyle practices (Bourne, 2007a, 2007b; Hambleton et al, 2005), water and (Abel-Smith, 1994), health care do not provide the core rationale for the health status of a population as the determinants of health covering, social, economic, psychological, environmental, and biological conditions. In many societies across the world, poverty is rural and gender specific. Poverty is more than just the lack of income (ie. low income) as it includes the lack of access to services,
193

resources and skills, vulnerability, insecurity and powerlessness. There is another result of poverty which has a multiple effect on the economy, and that is poor health conditions owing to malnutrition, low water quality, non-access to primary health care and food insecurity. According to the WHO (2005), 80% of chronic illnesses were in low and middle income countries, suggesting that illness interfaces with poverty and vice versa. A study by Bourne, Beckford and McGrowder (2009), using 2-decade of data on unemployment, self-reported and health-care-seeking behaviour of Jamaicans (from 1988-2007), found that there was a positive correlation between poverty and unemployment; poverty and illness; and crime and unemployment. In Jamaica, poverty is substantially a rural and gender phenomena. Statistics from the Planning Institute of Jamaica and the Statistical Institute of Jamaica (PIOJ & STATIN, 2008) revealed that in 1997, 19.9% of Jamaicans were poor. Of this figure, 73.3% were in rural areas; 13.1% in semi-urban zones and 13.6% in urban areas. One decade later (ie 2007), the prevalence of poverty fell to 9.9% of which 71.3% was in rural areas, 8.9% in semi-urban and 19.9% in urban zones. In the same year (ie 2007), 11.1% of persons living in female-headed households were classified as poor compared to 8.6% of those residing in male-headed household. Poverty is not only rural as there has been a rising in its levels in urban areas. The survey determined the poverty line was US$ 1,070.32 per year (US $2.92 per day) for an individual and US$ 4045.29 per year for a family of five (US $2.22 per person per day). The Jamaica Survey of Living Conditions (2002) indicated that the wealthiest 20% of the population accounted for 45.9% of national consumption while the poorest 20% accounted for only 6.1% of national consumption. On average, the wealthiest 10% of the population consumed approximately 12.5 times more than the poorest 10%. This is a mean per capita annual consumption expenditure of US$ 3963.53
194

compared to US$314.48. The Jamaica is not atypical in having poor people or having to address the predominance of this rural phenomenon. The World Bank (1996) estimated that in 1996,

38% of the total population (or 25% including Haiti) in the Caribbean or more than seven million people to be poor. According to Bourne (2008), in 1880 to 1882, life expectancy at birth for men was 37.02 years and 39.80 years for women with the gap between sexes widening to 5.81 years (71.26 for men and 77.07 for women). Despite the high life expectancy of women in Jamaica which is comparable to that of many developed nations (United Nations, 2002), people with lower socioeconomic status have worse health in all adult age groups, including older ages (House et al, 2005). Reduced capacity to generate income and the growing risk of illness increase the vulnerability of the elderly to poverty, regardless of their original economic status, in developing and industrialized countries (Lloyd-Sherlock, 2000). Poverty, therefore, is age, area and gender specific. Women are traditionally overrepresented among the poor and therefore in the long run, have less access to remuneration and health resources, including health insurance and social security services. Women are disadvantaged on some fundamental economic indicators such as unemployment and access to economic resources. In 2007 in Jamaica, for instance among the 124 500 unemployed persons in the labour force, 65.4 % were women (Planning Institute of Jamaica, 2008). Thus, women's health and the control that they can exercise over resources are key factors in achieving effectiveness, efficiency, and sustainability in health interventions. According to Marmot (2002), poverty accounts for poor nutrition and physical milieu, deprivation from material resources and further explains the higher levels of health conditions of
195

those that are therein. The WHO (2005) concurs with Marmot as it opined that poverty explains chronic illness and premature death. Women are more likely to be poor, unemployed and have lower material wealth compared to men. Like the WHO (2005), Marmot (2002) and Abel-Smith (1997) that showed the health challenges of being poor and by extension female, any study of health status and women must include not only poverty but other socio-demographic variables. Poverty is substantially more than income poverty; it is the denial of choices and opportunities for living a tolerable life (UNDP, 1997). Over the past two to three decades, our understanding of poverty has broadened from a narrow focus on income and consumption to a multi-dimensional notion of education, health, social and political participation, personal security and freedom, and environmental quality. Hence, those socio-economic factors not only explain poverty they influence health status for the individual, household, society, country and world. Health which is more than the absence of diseases (WHO, 1948) suggests that people are multi-dimensional and any study of their health status must incorporate the environment (Pacione, 2), income (Grossman, 1972; Smith & Kingston, 1997; Bourne, 2009). The WHO has endorsed the evaluation of social determinants in any examination of health status (WHO, 2008; Kelly et al. 2007). It is the social determinants (ie non-biological factors) which produce the inequality in income, health and regards health development. Hence, addressing those determinants account for a percentage of health status (Hambleton et al. 2005). In a study of elderly Barbadians, Hambleton et al. (2005) found that biological conditions accounted for 67.5% of health status of sample. This indicates that the social determinants are equally

important in the examination of health status (they account for 32.5% of the explanatory power of health status).
196

Concomitantly, Hambleton et al.s work reveals that there was a statistical causal relationship between socioeconomic conditions and the health status of Barbadians. The

findings reveal that 5.2% of the variation in reported health status was explained by the traditional determinants of health. Furthermore, when this was controlled for current

experiences, this percent fell to 3.2% (falling by 2%). When the current set of socioeconomic conditions were used they account for some 4.1% of the variation in health status, while 7.1% were due to lifestyle practices compared to 33.5% that was as a result of current diseases (see Hambleton et al. 2005). It holds that importance place by medical practitioners on the current illnesses as an indicator of health status is not unfounded as people place more value on biomedical conditions as responsible for their current health status. Diener (1984, 2000) and others (Idler & Benyamini 1997; Idler & Kasl, 199) have showed that wellbeing, happiness or health status is equally good to measure health or subjective wellbeing. Economists like Grossman (1972) and Smith & Kington (1997) have used selfreported health status in evaluating health of people. Hence, self-reported health status (health status) is widely accepted in health literature as a measure of health status. In this study, data were not collected on health status but on health conditions. The sample was asked to state whether they have an illness or not, and if they do what were the typology of health conditions. For this paper the researcher used good health status to indicate not reported a health condition and poor health to indicate at least one reported health condition. Self-reported ill-health is not an ideal indicator of actual health conditions because people may underreport; however, it is still an accurate proxy of ill-health and mortality (Idler & Kasl, 1991; Idler & Benyamini, 1997).

197

The reason for the importance of health conditions (illness) is simply that a healthy population holds the key to development. It is within this framework that a study of health is required to examine the factors that determine health status of women in the reproductive years of 15 to 49 years. It is clear from the review of the literature that health is influenced by income and other social factors. A literature search revealed that no study existing in the Caribbean, in particular Jamaica has sought to examine factors that determine the health status of rural women in the reproductive ages of 15 to 49 years. This is the first research of its type in the Caribbean and in particular Jamaica. It provides an insight into the factors that determine self-reported health status of women in ages 15 to 49 years, and this can now be used to guide public health policy. Hence, the purposes of this study are to (i) examine the good health status of women in the reproductive ages, (ii) model socio-economic determinants of good health status of women in the reproductive ages, and (iii) provide public health policy makers with research information on this cohort for better policies design in the future. Methods Participants and questionnaire The current research extracted a sample of 3 450 respondents who indicated that they were rural women ages 15 to 49 years. This sample was taken from a national cross-sectional survey from the 14 parishes in Jamaica. The survey used a stratified random probability sampling technique to drawn the original 25 018 respondents. The non-response rate for the survey was 29.7%. The study used secondary cross-sectional data from the Statistical Institute of Jamaica (2003) (ie Jamaica Survey of Living Conditions or JSLC). The JSLC was commissioned by the Planning

198

Institute of Jamaica and the Statistical Institute of Jamaica. These two organizations are responsible for planning, data collection and policy guidelines for Jamaica.

The JSLC is a self-administered questionnaire where respondents are asked to recall detailed information on particular activities to trained interviewers from the Statistical Institute of Jamaica. The questionnaire covers demographic variables, health, immunization of children 0 59 months, education, daily expenses, non-food consumption expenditure, housing conditions, inventory of durable goods and social assistance. Interviewers are trained to collect the data from household members. The survey is conducted between April and July annually. Model The multivariate model used in this study (a modification of Bourne and McGrowders health status model) captures a multi-dimensional concept of health and health status. It is fundamentally different from that of Bourne and McGrowders model (2009) as it is gender (women) and age specific (15 to 49 years), and a number of new variables were included such as social standing; crime and pregnancy. Hence, the proposed model that this research seeks to evaluate is displayed (Eqn (2)):
Ht = f(lnPmc, EDi, Rt, HIi, HTi, Xi, CRi,(NPi, PPi), Mi, Fi, Ni, Ai, i) [1]

Where the current good health status of a rural resident, Ht, is a function of 12 explanatory variables, where Ht is current good health status of person i, if good or above (ie no reported health conditions in the 4 weeks leading up to the survey period to trained interviewers from the
199

Statistical Institute of Jamaica), 0 if poor (ie at least one health condition reported to trained interviewers from the Statistical Institute of Jamaica); lnPmc is the logged cost of medical care of person i; EDi is the educational level of person i, 1 if secondary, 1 if tertiary and the reference group is primary and below; Rt is the retirement income of person i, 1 if receiving private and/or government pension, 0 if otherwise; HIi is the health insurance coverage of person i, 1 if they have a health insurance policy, 0 if otherwise; HTi is the house tenure of person i, 1 if rent, 0 if squatted; Xi is the gender of person i, 1 if female, 0 if male; CRi is crowding in the household of person i; (2i=1 NPi,PPi) NPi is the sum of all negative affective psychological conditions, and PPi is the sum of all positive affective psychological conditions; Mi is the number of males in the household of person i and Fi is the number of females in the household of person i; Ai is the age of the person i and Ni is the number of children in the household of person i; LLi is the living arrangements, where 1 = living with family members or relatives, and 0 = otherwise. Variables were identified from the literature, using the principle of parsimony. Only those explanatory variables that are statistically significant (p <0.05) were used in the final model to predict current health status of Jamaican women in the reproductive ages of 15 to 49 years. Here, the final model that accounted for self-reported good health of Jamaican women in the reproductive years of 15 to 49 years is expressed in Eqn. [2].
Ht = f(Wi, MRi, HIi, NPi,, Di, Ai, i) [2]

The current good health status of Jamaican women in the reproductive ages of 15 to 49 years, Ht, is a function of social standing of individual i, Wi; marital status of individual i, MRi; health insurance of person i, HIi; NPi is negative affective psychological conditions of person i; Di is

200

total number of durable good owned by individual i (excluding property and land) and Ai is the age of the person i. Measures An explanation of some of the variables in the model is provided here. Health status is a dummy variable, where 1 (good health) = not reporting an ailment or dysfunction or illness in the last 4 weeks, which was the survey period; 0 (poor health) if there were no self-reported ailments, injuries or illnesses. While self-reported ill-health is not an ideal indicator of actual health conditions because people may underreport, it is still an accurate proxy of ill-health and mortality (Idler & Kasl, 1991; Idler & Benyamini, 1997). Social supports (or networks) denote different social networks with which the individual is involved (1 = membership of and/or visits to civic organizations or having friends who visit ones home or with whom one is able to network, 0 = otherwise). Psychological conditions determine the psychological state of an individual, and this is subdivided into positive and negative affective psychological conditions (Diener, 2000; Harris & Lightsey, 2005) Positive affective psychological condition is the number of responses with regard to being hopeful, optimistic about the future and life in general. Negative affective psychological condition is the number of responses from a person on having lost a breadwinner and/or family member, having lost property, being made redundant or failing to meet household and other obligations. Per capita income quintile was used to measure social standing. Poor (ie lower class) were all individuals classified as in poorest and poor quintiles (ie quintiles 1 and 2); middle class were those classified as in quintiles 3 and wealth (upper classes) were those classified in quintiles 4 and 5 ( quintile 5 being the wealthiest income quintile). Statistical analysis
201

Statistical analyses were performed using the Statistical Packages for the Social Sciences v 16.0 (SPSS Inc; Chicago, IL, USA) for Widows. Descriptive statistics included frequency, mean and standard deviation were used to provide background information on the sample. A single hypothesis was tested, which was: the health status of rural residents is a function of demographic, social, psychological and economic variables. The enter method in logistic regression was used to test the hypothesis in order to determine those factors that influence the health status of rural residents. The logistic regression used as dependent variable was binary. The final model was based on those variables that were statistically significant (p <0.05), and all other variables were removed from the final model (p >0.05). Categorical variables were coded using the dummy coding scheme. The predictive power of the model was tested using the omnibus test of model and Hosmer and Lemeshows (2000) technique to examine the models goodness of fit. The correlation matrix was examined in order to ascertain whether autocorrelation (or multi-collinearity) existed between variables. Cohen and Holliday (1982) stated that correlation can be low/weak (00.39); moderate (0.40.69), or strong (0.71). This was used in the present study to exclude (or allow) a variable. Finally, Wald statistics were used to determine the magnitude (or contribution) of each statistically significant variable in comparison with the others, and the odds ratio (OR) for interpreting each of the significant variables. Results: Demographic Characteristics of sample

Of the sampled respondents (n=3,450), 84.7% reported good health; 3.3% were pregnant; 89.6% had secondary level education; 20.1% were married; 78.6% were never married; 5.5% had
202

private health insurance coverage; 58.3% were owners of lands;40.1% had some form of social support; mean age was 29.7 years (SD=9.9 years); 45.7% belonged to the two poorest quintiles compared to 34.1% who were classified in the two wealthiest quintiles and 49.6% visited a public hospital or public health care establishment in the 4-week period of the survey (Table 8.1). On an average, there were 2 persons per household (SD=1 person), with average medical expenditure being US $26.37 (SD= US$40.81). Of the 15.3% of the sample that indicated poor current health status, 69.3% reported being diagnosed with (chronic) recurring illness. Marginally, more of those who reported being diagnosed with a recurring ailment had hypertension (36.4%); 31.8% did not specify the condition; 22.7% indicated arthritis and 9.1% claimed diabetes mellitus. When those who mentioned having a recurring dysfunction were asked about the length of the last attack, the median number of days was 7 days. They also indicated that 3 days were the median number of days that prevented them from carrying out their normal activities.

203

Table 8.1: Demographic characteristic of sample Number Percent Current Health Status: Poor 511 15.3 Good 2832 84.7 Pregnant: No 3143 96.7 Yes 106 3.3 Social Support: No 2065 59.9 Yes 1385 40.1 Educational Level: Primary or below 151 5.3 Secondary or post-secondary 2574 89.6 Tertiary 149 5.2 Visits to: Public hospital or establishment 122 49.6 Private hospital or establishment 124 50.4 Social Standing (ie per capita Income quintile): 1=Poorest 768 22.3 2 808 23.4 3 698 20.2 4 707 20.5 5=Wealthiest 469 13.6 Marital status: Married 665 20.1 Never married 2605 78.6 Divorced/Separated/Widowed 45 1.3 Health Insurance: No 3138 94.5 Yes 183 5.5 Land Ownership: No 1025 41.7 Yes 1432 58.3 Age (Mean SD) 29.7 9.9 Crowding (Mean SD) 2.1 1.3 Average Annual Consumption per household (Mean SD): Ja. $30,216.64 Ja.$39,095.35; (Minimum: Ja.$1,546 to maximum: Ja.$1,876,821) Medical Expenditure (Mean SD) Ja.$1,344.22 Ja.$2,079.87 Ja $50.97 = 1 US$

204

Disaggregating current good health status of the sample by pregnancy or no pregnancy revealed that there is no statistical difference between the two groups (p=0.356). Approximately 85% of the sample reported good current health status compared to 83% of the women who were pregnant and 85% for those who were not pregnant (Table 8.2).

Table 8.2: Current Health Status by Pregnancy Status

Pregnancy Status

Health status

Not pregnant n (%)

Pregnant n (%)

Total n (%)

Poor

480 (15.3)

18 (17.0)

498 (15.3)

Good

2663 (84.7)

88 (83.0)

2751 (84.7)

Total

3143

106

3249

2 (1) = 0.231, p=0.356

205

A cross tabulation between reported recurring illness and per capita population quintile revealed a statistical correlation (p=0.030) (Table 8.3). Self-reported diabetes mellitus was reported as illness of wealthy rural women in the reproductive ages of 15 to 49 years (24% for quintile 4 and 25% for quintile 5). Table 8.3 showed that 42% of those in quintile 2 who reported a recurring illness had hypertension, 50% of those in quintile 3 and 75% of the wealthiest quintile. Self-reported arthritis was greater in the wealthy quintile (76%) compared to 28.6% for those in quintile 2. Substantially, more rural women in the reproductive ages of 15 to 49 years reported an unspecified illness (100%) compared to 28.6% of those in the poor quintile and 50% of those in the middle income quintile. Table 8.3: Recurring Illness by Per capita Population Quintile Per Capita Population Quintile 1=poorest n (%) 0 (0.0) 2 n (%) 0 (0.0) 3 n (%) 0 (0.0) 33 (50.0) 0 (0.0) 33 (50.0) 66 4 n (%) 16 (24.0) 0 (0.0) 50 (76.0) 0 (0.0) 66 5=wealthiest n (%) 17 (25.0) 50 (75.0) 0 (0.0) 0 (0.0) 67 Total n (%) 33 (9.1) 132 (36.4) 83 (22.7) 116 (31.8) 22

Recurring Illness Diabetes mellitus Hypertension Arthritis Unspecified

0 (0.0) 49 (42.0) 0 (0.0) 33 (28.6) 50(100.0) 33 (28.6) 50 115

Total 2 (12) =22.755, p=0.030

206

There is a statistical correlation between visits to the type of health care facilities and social standing of rural women in the reproductive ages of 15 to 49 years (2 (4) =22.993, p<0.001). Three times more of the poorest respondents visited public health care establishment than private health care facilities in comparison to 3 times more of the wealthiest who attended private than public health care establishment for health care visits (Table 8.4). Here Table 8.4 showed that as ones social standing increases from poorest to wealthiest, they switch from the usage of public to private health care facilities. Table 8.4: Visits to Private or Public Health Care Establishment by Social Standing Per Capita Population Quintile Visits to health care establishment Private Public 5.00= Wealthiest n (%) 27 (73.0) 10 (27.0) 37

1=Poorest 2.00 n (%) n (%) 13 (26.0) 37 (74.0) 50 28 (45.9) 33 (54.1) 61

3.00 n (%) 19 (50.0) 19 (50.0) 38

4.00 n (%) 37 (61.7) 23 (38.3) 60

Total n (%) 124 (50.4) 122 (49.6) 246

Count 2 (4) = 22.993, p < 0.001

Results: Multivariate Regression Using logistic regression analyses, 6 variables emerged as statistically significant predictors of current good health status of rural women (ie. ages 15 to 49 years) in Jamaica (Table 8.5). These are social standing (two wealthiest quintile OR=0.524, 95%CI: 0.350,0.785); marital status (separated, divorced or widowed OR=0.382, 95%CI: 0.147, 0.991); health insurance (OR=0.041, 95%CI: 0.024, 0.069); negative affective psychological conditions (OR=0.951,
207

95%CI:0.704, 1.284); asset ownership (OR=1.089, 95%CI:1.015, 1.168) and age of respondents (OR+0.965, 95%CI:0.949, 0.982). Controlling for the effect of other variables, the average likelihood of reporting good health increased by nearly 5 times. Further examination of the model (i.e. Equation (2)) revealed that this had a significant predictive power (model 2 = 259.945, p <0.001; Hosmer and Lemeshows goodness of fit 2 = 9.649, p = 0.71; Nagelkerke R2 =0.230 or 23.0%) and correctly classified 87.1% of the sample (correctly classified 98.5% of those who reported good health and 26.2% of those who indicated poor health status). The logistic regression model can be written as: Log (probability of good health/probability of not good health) = 3.131 0.645 (two health quintiles) -0.964 (Separated, Divorced or widowed) 3.195 (Ownership of Health Insurance Coverage) 0.057 (Negative Affective psychological conditions score) + 0.085 (Asset ownership score) 0.035 (Age).

208

Table 8.5: Logistic Regression of Good Health Status of Women in the Reproductive Ages
Variable Middle Quintile Two Wealthiest Quintiles Poorest quintile Log HealthCare Cost Separated, Divorced or Widowed Married Single Health Insurance Physical environment Social support Secondary schooling Tertiary schooling Primary and below Living arrangement Crowding Crime Index Landownership Negative Affective Positive Affective Asset ownership (exclude land) Age Dummy pregnant Household Head Average Income per head House tenure (rented) House tenure (owned) House tenure (squatted) Constant Coefficient -0.177 -0.645 0.000 -0.964 -0.037 -3.195 0.112 -0.046 -0.062 0.184 0.069 -0.077 0.001 -0.051 -0.057 0.007 0.085 -0.035 -0.072 0.430 0.000 -2.095 -0.036 3.131 Std Error 0.207 0.206 0.000 0.487 0.177 0.267 0.166 0.148 0.314 0.461 0.564 0.062 0.008 0.153 0.024 0.033 0.036 0.009 0.425 0.485 0.000 1.801 1.092 1.304 Odds Ratio 0.838 0.524 1.000 0.382 0.964 0.041 1.118 0.956 0.940 1.201 1.071 0.926 1.001 0.951 0.945 1.007 1.089 0.965 0.931 1.537 1.000 0.123 0.965 22.902 95.0% C.I. Lower, Upper 0.558, 1.258 0.350, 0.785** 1.000, 1.000 0.147, 0.991* 0.681, 1.364 0.024, 0.069*** 0.807, 1.549 0.715, 1.277 0.508, 1.741 0.487, 2.966 0.355, 3.234 0.820, 1.046 0.985, 1.017 0.704, 1.284 0.902, 0.990* 0.945, 1.074 1.015, 1.168* 0.949, 0.982*** 0.405, 2.141 0.594, 3.976 1.000, 1.000 0.004, 4.197 0.114, 8.198 -

2 (23) =259.945, p < 0.001; -2 Log likelihood = 1316.563 Hosmer and Lemeshow goodness of fit 2=9.649, p = 0.71 Nagelkerke R2 =0.230 Overall correct classification = 87.1% Correct classification of cases of good or beyond health status =98.5% Correct classification of cases of no dysfunctions =26.2% Reference group *p < 0.05, **p < 0.01, ***p < 0.001

209

Discussion The current paper found that of the thirteen socio-economic variables that were examined, six of them are predictors of good health status of women in the reproductive ages. These socioeconomic determinants are social standing (two wealthiest quintiles); marital status (separated, divorced, widowed); health insurance coverage; psychological condition (negative affective psychological condition); asset ownership and age of respondents. This concurs with the findings of the WHO (2005) that social determinants should be taken into consideration in the study of health status. Another study (Hambleton et al. 2005) found social, economic and biological determinants of health status of Barbadian elderlyContinuing, social determinants are

The use of self-reported health status (ie subjective wellbeing) is well established in research literature as a good measurement for health or wellbeing. Using peoples assessment of their life satisfaction and health is old, and has already been resolved. Nevertheless, it will be succinct issues here for those who are not cognizant of this discourse. Scholars have established that there is a statistical association between subjective wellbeing (self-reported wellbeing) and objective wellbeing (Diener, 2000; Lynch, 2003) and Diener went further when he found a strong correlation between the two variables (Diener, 1984). Gaspart (1998) opined about the difficulty of objective quality of life (GDP per capita) and the need to use self-reported wellbeing in the assessment of the wellbeing of people. He wrote, So its objectivism is already contaminated by post-welfarism, opening the door to a mixed approach, in which preferences matter as well as objective wellbeing (Gaspart, 1998) This speaks to the necessity of using a measure that captures more to this multidimensional construct that continues with the traditional
210

income per capita approach. Another group of scholars emphasized the importance of measuring wellbeing outside a welfarism and/or purely objectification, when they said that Although GDP per capita is usually used as a proxy for the quality of life in different countries, material gain is obviously only one of many aspects of life that enhance economic wellbeing (Becker et al, 2004) and that wellbeing depends on both the quality and the quantity of life lived by the individual. The discourse of subjective wellbeing using survey data cannot deny that it is based on the persons judgement, and must be prone to systematic and non-systematic biases (Frey & Stutzer, 2005). Diener, an early survey wrote that [the] measures seem to contain substantial amounts of valid variance (Diener, 2000). This will not be addressed in this paper as this is not the nature or its scope. Despite this limitation, a group of economists noted that happiness or reported subjective well-being is a satisfactory empirical approximation to individual utility (Frey & Stutzer, 2005) and this justifies its usage in wellbeing research. The current research used self-reported health status to examine those factors that determine good health status of rural women in the reproductive ages 15 to 49 years. Unlike a recent study conducted by Bourne and McGrowder (2009) using a randomly selected sample of 5,683 rural Jamaicans, They found that good health status was predicted by medical

expenditure; health insurance; education; house tenure; gender; psychological conditions (i.e. positive and negative affective psychological conditions); typology of household members and age of respondents and retirement income. This study concurred with age; negative affective psychological conditions; health insurance, and added some new factors such as social standing; marital status, and asset ownership. This research has revealed that there was no statistical
211

difference between the self-rated good health status of rural women who were pregnant or not pregnant.

Bourne and McGrowders work showed that 83 out of every 100 rural residents had good health status compared to this study that revealed that 85 out of every 100 rural women (ages 15 to 49 years) reported good health. This study has not only highlighted the current good health status inequality between rural Jamaicans and rural women in the reproductive ages 15 to 49 years in Jamaica, but it showed the health disparity between the typology of variables. One of the disparities between the current paper and that of Bourne and McGrowder was social standing. In the latter work this variable was not significant, while it is in the former one. The finding in this paper revealed that the odds of self-reported good current health status of those rural women in two wealthiest quintiles were 48% lower than that of the odds of rural women in the two poorest quintiles. This contradicts works that have established the correlation between poverty and health status (Murray, 2006; Marmot, 2002; Muller & Krawinkel, 2005; Bloom & Canning, 2003; Smith & Waitzman, 1994). Marmot (2002) opined that poverty influences health through malnutrition, low water and environmental quality, and the non-access to material resources further validate poor health status. This assumes that wealth accounts for better environmental quality and good health status. While wealth opens access to financial and/or other materials resources, it is an explanation of poor lifestyle choices. Wealth does not mean that people become more health conscious. Instead, it means access to liquor, cigars, hard drugs, and many excess that are of
212

themselves health hazards. The issue of poor environment is not a disparity for rural areas in Jamaica as the quality of milieu in those places is about the same. Hence, the health status difference between rural women in the reproductive years of the two wealthiest and two poorest quintiles would be owing to lifestyle practices and access to more financial resources. In this study, it can be inferred from the data that although poverty is a health hazard, it is advantageous for rural women in the reproductive years 15 to 49 years. This is supported by the morbidity data that showed the five leading causes of health conditions in women in Jamaica (heart disease, hypertension, diabetes mellitus, arthritis, and neoplasm cancer), most of those diseases are causes of lifestyle practices (Davidson et al, 2002; Jamaica Social Policy Evaluation, 2003). In an article published by CAJANUS, the prevalence rate of diabetes mellitus affecting Jamaicans was higher than in North American and many European countries (Callender, 2000, p. 67. Diabetes Mellitus was not the only challenge faced by patients; McCarthy (McCarthy, 2000) argued that between 30 to 60% of diabetics also suffered from depression, which is a psychiatric disorder. The issue of the lifestyle practices accounted for the health disparity between rural women in the reproductive years of 15 to 49 years and those in the two wealthiest quintiles compared to those in the two poorest quintiles is reinforced in the fact that there is no statistical difference between the health status of rural women who were in the two poorest quintiles and those in the middle quintile. In light of the above, the wealth disparity between the two aforementioned groups is narrowed and can aid in the explanation of the health disparity between wealthy and poor rural women in Jamaica. This research showed that hypertension and diabetes mellitus which are lifestyle causes of non-communicable diseases were higher in the wealthiest quintile
213

than the poorest quintile. An interesting finding was unwillingness of those in the poor to poorest quintile to declare their dysfunction, unlike those in the middle to upper classes. Of the sample, 4 out of every 100 rural women in the reproductive ages 15 to 49 years reported having hypertension, 2 out of every 100 had arthritis, 1 out of every 100 had diabetes mellitus and 3 out of every 100 did not specify their recurring illness. Social standing is among the variables that explain health status of rural women in the reproductive years of 15 to 49 years. Another factor is marital status. Studies have shown that a statistical correlation existed between marital status and health status. Some studies have shown that married people have a lower mortality risk in the healthy category than the nonmarried (Goldman, 1993), and this justifies why they take less life-threatening risks (Smith & Waitzman, 1994; Umberson, 1987). According to Delbs & Gaymu (2002), The widowed have a less positive attitude towards life than married people, which is not an unexpected result (Delbs & Gaymu, 2002, pp. 885-914). Using a sample of 1049 Austrians from ages 14 years and over, Prause et al. (2004) found that married individuals had greater subjective health-related quality of life index (8.3 ) than divorced persons (7.6) or singles (7.7). Smock, Manning and Gupta (1999) concurred with Prause et al that there is a direct relationship between married women and economic well-being. Drawing on longitudinal data from the National Survey of Families and Households for 19871988 (NSHH1) and a follow-up survey (NSFH2) of some 13, 008, a sample size of 2665 females from 60 years and older was used. Each study had a response rate of approximately 74 % for NSFH1 and 82% for NSFH2. The research revealed that married women had a higher economic well-being than divorced females. It was found that females who were remarried experienced an
214

equally high well-being as their married counterparts, which was higher than that experienced by single females. The current paper refutes the aforementioned finding as there was no statistical difference between current health status of married rural women in the reproductive ages of 15 to 49 years and non-married ones. However, in this study, non-married rural women in the reproductive years 15 to 49 years had a greater current health status than those divorced, separated or widowed. Furthermore, the odds of reporting good health status for divorced, separated or widowed rural women in this study was 62% less likely than the odds of reporting good health status of non-married rural women in the current work. This leads to the next variable, which is health insurance coverage. For this study, health insurance coverage was negatively correlated with good health status which concurs with Bourne and McGrowders work (2009). In the current research, the odds of good health for rural women in the reproductive ages 15 to 49 years who had health insurance coverage was 96% less than the odds of good health for rural women who do not have health insurance coverage. This indicates that health insurance coverage is not an indicator of health seeking behaviour. Instead, it can be used to evaluate poor health of rural women in the reproductive ages of 15 to 49 years. In the pursuit of healthy lifestyle, one of the measures of wellness is health seeking behaviour. Health insurance is a curative measure of illness as people hold health plan policies more if they are more likely to be ill than less likely, suggesting that people analyze their health risk and if it is highly likely to become ill, they will hold health insurance and not the vice versa. Age is the next variable which is a predictor of current good health status of rural women in this sample. It is well established in health literature that there is a negative correlation between age and health status (Abel-Smith, 1994; Grossman, 1972; Hambleton et al, 2005;
215

Bourne, 2008; Bourne & McGrowder, 2009) and this also extends to biological studies. The negative association between age and good health status is once again concurred with as the current work revealed that the odd of reporting good health status for each additional year of the rural women in the reproductive ages of 15 to 49 years is 3.5% less than the odds of a rural woman who is one year younger. Another variable that is inversely correlated with good health status was negative affective psychological conditions. Acton & Zodda (2005) aptly summarized these negative affective psychological conditions and they found that expressed emotion is detrimental to the patient's recovery; it has a high correlation with relapse to many psychiatric disorders (Acton & Zodda, 2005, pp. 373-399). Studies have revealed that up to 80% of people who committed suicide had several depressive symptoms (Rhodes et al, 2006). From a 10-year longitudinal study conducted in the United States by Beck et al (Beck et al, 1985) it is further stated that

hopelessness was a major predictor of suicidal behaviour which was equally concurred by Smyth & MacLachlan (2005). In this study negative affective psychological conditions were operationalized using loss of breadwinners, family members; jobs and general hopelessness of an individual which further explains the negative association between this variable and good health status. Continuing, the odds of reporting good health status based on increased negative affective psychological conditions is 9.8% less than the odds of lowered negative affective psychological conditions for rural women in ages 15 to 49 years. Unlike the other predictors of good health status, asset ownership was the only one that was positively correlated with current good health status for the sampled respondents. The findings revealed that the odds of reporting good health status for those who owned more assets
216

was 8.9% more than for those who owned less assets. This concurs with other studies that showed the direct correlation between asset ownership and health status (Grossman, 1972; Summers & Heston, 1995) and according to Summers & Heston (1995), The index most commonly used until now to compare countries' material well-being is their GDP POP' [production of goods and services] However, GDPPOP is an inadequate measure of countries' immediate material well-being, even apart from the general practical and conceptual problems of measuring countries' national outputs (Summers & Heston, 1995). Generally, from that perspective, the measurement of quality of life is, therefore, highly economic and excludes the psychosocial factors, and if quality of life extends beyond monetary objectification then it includes biological, nutrition, social, cultural, economic and psychological factors. The World Bank went further when it said that womens health status is influenced by a complex set of biological, social, cultural and psychological variables which are all interrelated (World Bank, 1994). An interesting finding that is embedded in this research is the quality of the health care institutions in Jamaica. The research showed that those in the poorest quintile had a greater health status than those in the wealthiest quintile, and that those in the poorest quintiles enjoyed the same good health status as those in the middle class (i.e. quintile 3). Given that 46% of the sample was in the poorest social standing and that 74% of those who were in this social standing visited public health care establishment for medical care, then a part of the explanation for the good health status of this group will be owing to the quality of primary health care and public medical health care institution in the society. Within the context that those in the wealthy and wealthiest social standings have a greater access to financial resources, they are both able to visit private health care institutions and spend substantially more on health care than those in the poor social standing. This spending does not translate into better health status, suggesting that
217

income cannot buy better health. Conclusion Poverty is synonymous with rural area and women, and inspite of this reality majority of rural women in Jamaica ages 15 to 49 years have reported good current health status. Wealth creates more access to financial and other resources and makes a difference in nutritional intake, water and food quality as well as an explanation for better environmental conditions. In this study, wealth did not mean better health but that poor women had greater health status than their wealthy counterparts. Another interesting finding was that good health is inversely correlated with the ownership of health insurance coverage, suggesting that Jamaican rural women (ages 15 to 49 years) do not buy health plans because they are healthy but owing to unhealthy risk factors. Womens health is not merely important because of academic literature; but that it is pivotal to their earning capacity, health of the children and the general household. Hence, understanding womens health is to comprehend its multiple effects on different areas of the family, the household and the nation. To summarize, good health in this study can be predicted by 6 factors (social standing, marital status, health insurance, negative affective psychological conditions, assets ownership and age of respondents) this adds more information than voluminous amount of literature on maternal mortality and/or fertility of this age cohort. In keeping with some issues raised in this paper, the researchers recommend that a lifestyle survey be conducted on this age cohort in order to provide pertinent information and direction for public health policy programmes.

218

References

Abel-Smith B (1994), An introduction to health: Policy, planning and financing. Longman Group. Acton GS, Zodda JJ (2005), Classification of psychopathology: Goals and methods in an empirical approach. Theory of Psychology,15, pp. 373-399. http://www.personalityresearch.org/acton/goals.html (accessed July 24, 2006). Beck AT, Steer, RA, Kovacs M, Garrison B (1985), Hopelessness and eventual suicide: A 10year prospective study of patients hospitalized with suicidal ideation. American Journal of Psychiatry, 142, pp. 559-563. Becker GS, Philipson TJ, Soares RR (2004), The quantity and quality of life and the evolution of world 0of%20life%22 (accessed August 22, 2006). Benzeval M, Judge K, Shouls S (2001), Understanding the relationship between income and health: How much can be gleamed from cross-sectional data? Social policy and Administration. Quoted in Benzeval M, Judge K. Income and health: the time dimension. Social Science and Medicine, 52, pp. 1371-1390. Bloom DE, Canning D (2003), The health and poverty of nations: From theory to practice. Journal of Human Development, 4, pp. 47-72. Bourne PA (2009). Good Health Status of Older and Oldest Elderly in Jamaica: Are there differences between rural and urban areas? Open Geriatric Medicine Journal, 2:18-27. Bourne PA (2007a), Determinants of Wellbeing of the Jamaican Elderly. Unpublished MSc Thesis, the University of the West Indies, Mona. Bourne PA (2007b), Using the biopsychological Model to evaluate the wellbeing of the Jamaican elderly. West Indian Medical J, 56, [Abstract], (Suppl 3), pp. 39-40.
219

inequality;

http://www.spc.uchicago.edu/prc/pdfs/becker05.pdf#search=%22preston%2C%20quality%2

Bourne PA. (2008), Medical Sociology: Modelling well-being for elderly people in Jamaica. West Indian Medical Journal, 57, pp. 596-604. Bourne, PA. & Beckford OW. (2009). Unemployment, Illness and Health Seeking Behaviour in Jamaica. Paper presented at the 34th annual conference, Caribbean Studies Association Conference 2009 June 1-5, 2009,Hilton Kingston, Jamaica.

Bourne PA, McGrowder DA (2009), Rural health in Jamaica: examining and refining the predictive factors of good health status of rural residents. Rural and Remote Health, 9, p. 1116. Available from: http://www.rrh.org.au Callender J (2000), Lifestyle management in the hypertensive diabetic. CAJANUS, 33, pp. 6770. Cochrane AL, et al (1978), Health service, input and mortality output in developed countries. Journal of Epidemiology and Community Health 32. Cohen L, Holliday M (1982), Statistics for Social Sciences. London, England: Harper and Row. Davidson W, Wright V, Lowe H (2002), The Wellness Handbook: Your Guide to Healthy Living. Kingston: Pelican Publishers. Delbs C, Gaymu J (2002), The shock of widowed on the eve of old age: Male and female experience. Demography, 3, pp. 885-914. Diener E (1984), Subjective well-being. Psychological Bulletin, 95, pp. 54275. Diener E (2000), Subjective well-being: The science of happiness and a proposal for a national index. American Psychological Association, 55, pp. 34-43. Diener E (2000), Subjective well-being: the science of happiness and a proposal for a national index. American Psychologist, 55, pp. 3443

220

Frey BS, Stutzer A (2005), Happiness Research: State and Prospects. Review of Social Economy (LXII), pp. 207-228. (Available: http://www.bsfrey.ch/articles/420_05.pdf, accessed April 6, 2009). Gaspart F (1998), Objective measures of well-being and the cooperation production problem. Social Choice and Welfare, 15, pp. 95-112. Goldman N. Marriage selection and mortality patterns: Inferences and fallacies. Demography 1993;30:189-208. Grossman M (1972), The demand for health- a theoretical and empirical investigation. New York: National Bureau of Economic Research. Hambleton IR, Clarke K, Broome Hl, Fraser HS, Brathwaite F, Hennis A (2005), Historical and current predictors of self-reported health status among elderly persons in Barbados. Rev Panam Salud Publica, 17, pp. 342-353. Harris PR, Lightsey OR Jr (2005), Constructive thinking as a mediator of the relationship between extraversion, neuroticism, and subjective wellbeing. Personality, 19, pp. 409-426. Homer D, Lemeshow S (2000), Applied Logistic Regression, 2nd edn. John Wiley & Sons Inc., New York. House JA, Lantz PM, Herd P (2005), Continuity and change in the social stratification of aging and health over the life course: Evidence from a nationally representative longitudinal study from 1986 to 2001/2002 (Americans Changing Lives Study). Journals of Gerontology: Social Sciences, 60B, (Special Issue II), pp. 15-26. Idler EL, Benyamini Y (1997), Self-reported health and mortality: a review of twenty-seven community studies. J Health Soc Behav, 38, pp. 21-37. Idler EL, Kasl S (1991), Health perceptions and survival: Do global evaluations of health status really predict mortality. Journal of Gerontology, 46, pp. S55-S65. Jamaica Social Policy Evaluation (JASPEV), (2003), Annual Progress Report on National Social Policy Goals 2003. Kingston: JASPEV, Cabinet Office Jamaica. Kelly MP, Morgan A, Bonnefoy J, Butt J, Bergman V. (2007).The social determinants of health: Developing an evidence base for political action. Final Report to World Health Organization
221

European Journal of

Commission on the Social Determinants of Health from Measurement and Evidence Knowledge April 29, 2009) Lloyd-Sherlock P (2000), Old age and poverty in developing countries: new policy challenges. World Development, 28, pp. 2157-2168. Lynch SM (2003), Cohort and Life-course patterns in the relationship between education and health: A hierarchical approach. Demography, 40, pp. 309-331. Marmot M (2002), The influence of Income on Health: Views of an Epidemiologist. Does money really matter? Or is it a marker for something else? Health Affairs, 21, pp.31-46. Matsaganis M (1992), An economic approach to international and inter-regional mortality variations with special reference to Greece. (Unpublished PhD thesis, University of Bristol). McCarthy FM (2000), Diagnosing and Treating Psychological problems in Patients with Diabetes and hypertension. CAJANUS, 33, pp. 77-83. Muller O, Krawinkel M (2005), Malnutrition and health in developing countries. Canadian Medical Association Journal, 173, pp. 279-286. Murray S. Poverty and health (2006), Canadian Medical Association Journal, 174, pp. 923-923. Pacione M (2003), Urban environmental quality of human wellbeinga social geographical perspective. Landscape and Urban Planning, 65, pp. 19-30. Planning Institute of Jamaica and the Statistical Institute of Jamaica. Survey of Living Conditions, 2003. Kingston; 2004 Planning Institute of Jamaica, Statistical Institute of Jamaica (2008), Survey of Living Conditions, 2007. Kingston Planning Institute of Jamaica, Statistical Institute of Jamaica (2002), Survey of Living Conditions, 2001. Kingston. Planning Institute of Jamaica (2008), Economic and Social Survey Jamaica, 2007. Kingston: PIOJ. Prause W, Saletu B, Tribl GG, Rieder A, Rosengerger A, Bolitschek J, Holzinger, B, Kaplhammer G, Katschning H, Kunze M, Popovic R, Graetzhofer E, Zeitlhofer J. Human Psychopharmacology Clin Exp., 2004;20:359-365.
222

Network.

Available

from

http://www.who.int/social_determinants/resources/mekn_final_report_102007.pdf (accessed

Rhodes AE, Bethel J, Bondy S (2006), Suicidality, depression, and mental health service use in Canada. Canadian Journal of Psychiatry, 51, pp. 35-41. Smith JP, & Kington R. (1997). Demographic and Economic Correlates of Health in Old Age. Demography 34:159-70. Smith KR, Waitzman NJ (1994), Double jeopardy: Interaction effects of martial and poverty status on the risk of mortality. Demography, 31, pp. 487-507. Smock P, Manning WD, Gupta S (1999), The effects of marriage and divorce on womens economic well-being. American Sociological Review, 64, pp. 794-812. Smyth C, MacLachlan M (2005), Confirmatory factor analysis of the Trinity Inventory of Precursors to Suicide (TIPS) and its relationship to hopeless and depression. Death Studies, 29, pp. 333-350. Statistical Institute Of Jamaica (2003), Jamaica Survey of Living Conditions, 2002 [Computer file]. Kingston, Jamaica: Statistical Institute Of Jamaica [producer], 2002. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors]. Summers R, Heston A (1995), Standard of Living: SLPOP An Alternative Measure of Nations' Current Material Well-Being. http://pwt.econ.upenn.edu/papers/standard_of_living.pdf (accessed January 24, 2006). Umberson D (1987), Family status and health behaviors: Social control as a dimension of social integration. Journal of Health and Social Behavior, 28, pp. 306-19. United Nations (2002), Population ageing, 2002. Department of Economic and Social Affairs, Population Division. New York: UN. United Nations Development Programme. Human development report 1997. New York: OUP; 1997. World Bank (1994), Development in Practice. A New Agenda for Womens Health and Nutrition. Washington DC: World Bank. World Bank (1996), Poverty Reduction and Human Resource Development in the Caribbean, Washington D.C. WHO. (2008). The Social Determinants
223

of

Health;

2008.

Available

at

http://www.who.int/social_determinants/en/ (accessed April 28, 2009). World Health Organization. (2005). Preventing Chronic Diseases a vital investment. Geneva: WHO. World Health Organization, (1948). Preamble to the Constitution of the World Health Organization as adopted by the International Health Conference, New York, and June 1922, 1946; signed on July 22, 1946 by the representatives of 61 States (Official Records of the World Health Organization, no. 2, p. 100) and entered into force on April 7, 1948. Constitution of the World Health Organization, 1948. In Basic Documents, 15th ed. Geneva, Switzerland: WHO.

224

9
Determinants of Quality of Life of Jamaican Women

The current paper seeks to examine the quality of life (or subjective wellbeing) of Jamaican women by building a model that will capture socio-demographic and economic determinants of their quality of life. The study reveals that the model explains 18.5% (Adjusted R-squared) of the variability of quality of life of women; with 6 variables accounting for this variance. Further examination of the sociodemographic determinants revealed that subjective social class (Beta = 0.198; 95%CI:0.380, 1.070) is the most influential factor followed by employment status (Beta = 0.167; 95%CI:0.304, 0.1.077), religiosity(Beta = 0.152; 95%CI:0.214, 0.974), income (Beta = 0.155; 95%CI:0.015, 0.101), the administration of the governance of the nation (Beta = -0.139; 95%CI:-0.893, -0.203) and lastly by interpersonal trust (Beta = 0.094; 95%CI:0.020, 0.676). In summary, the factors of quality of life of a Jamaican Woman are social class, employment, income and religiosity, with social class being the most influential of all the variables. Employment does not merely about the income, but it is about the independence, the choices, the sense of freedom, the positive psychological attributes that this freedom gives as well as the selfadvancement that it is likely to provide why this variable is of that importance in determining the quality of life of Women. The current work does not provide all the answers, but it is catalysts upon which we are able to build, modify and refute research as this provide a platform upon which this is probable in the future.

Introduction

The current paper seeks to examine the quality of life (i.e. subjective or self-reported wellbeing) of Jamaican women by building a model that will capture socio-demographic and economic determinants of quality of life of this cohort. The rationale that underpins the current work is principally driven by the lack of academic literature on the subjective wellbeing or quality of life
225

on the particular gender. Most studies on quality of life have incorporated gender as a predictive factor or a determinant of subjective wellbeing studies (or quality of life) (Bourne, 2007; Murphy and Murphy 2006; Hambleton et al. 2005; Hutchinson et al. 2004; Stutzer and Frey 2003, 2001; Easterlin 2001a, 2001b, 1995; Lyubomirsky 2001; Cummins 2000; Diener 2000, 1985; Smith and Kington 1997; Grossman 1972). One study examining a particular quality of life of an elderly man shows how medical practitioners over many years sought to address a particular issue that was eroding the wellbeing of a patient who had a certain physiological dysfunctions (Ali and colleagues 2007). Can medical practitioners and social researchers assume that the quality of life of sexes is the same, given that they are of the same species? Such a situation is simple, as the physiological composition of the sexes is different, purchase power party differs, gender culturalization is dissimilar as well as the disparity between gender opportunities. Within this context, researchers, medical practitioners and policy makers need to understand the factors that influence quality of life of each gender as they are sex specify in enhancing the specificity that is needed to planning for the sex differential. A primary rationale for this awareness is owing to the opportunity differential because of one sex in society. In 1991, the unemployment rate was 22.2 per cent for females compared to 9.4 per cent for males and in 2007, the figure fell to 6.2 per cent for males and 14.5 per cent for females (Table 9.4). The more drastic reduction in the unemployment rate for women cannot constitute any form of betterment of females over their male counterparts as in 2007 the unemployment rate for female was twice more than that of males. The statistics reveal that men enjoy a 17 per cent higher employed labour force than females; and this indicates the opportunity of greater economic resources (Table 9.4). Another good measure that can be used to evaluate betterment of the sexes is economic resources (i.e. wages or salaries). In the Economic and Social Survey of
226

Jamaica (2004, p. 21.9), the publication showed that on an average the earnings of males (mean wage = $2.4 million) was 2 times more than that of females ($ 1.7 million); and that 76 per cent of senior positions were held by males although 54 per cent of executive and managerial positions were held by females. If males are still receiving greater salaries compared females and they experience high degrees of employment, we cannot concur with Miller nor Chevannes or Gayle that they are marginalized despite the fact that they are living fewer years than females (Table 9.2). From a study, using survey data from 1988 to 1999, conducted in Argentina, Brazil and Costa Rica, the researchers found that there is no general trend of economic marginalization of males in those societies (Omar Arias, 2001), which is evident from the some of economic indicators in Jamaica. On the other hand, what about our women? The importance of women in fertility as well as the fact that they have a greater life expectancy compared to their male counterparts (Table 9.1); it is timely that a research be done on this cohort to unearth what constitute their quality of life? Scholars who have done studies on different Caribbean nations like Bourne, 2007; Eldermire 1997, 1996, 1995a, 1995b, 1994, 1987a, 1987b; Hambleton et al. 2005; Brathwaite, their works on the quality of life have been substantially on elderly people ( ages 60 years or older or 65 years and older) with no particular interest on a certain sex. Other studies on the same region have looked on the total population (Hutchinson et al. 2005). Caribbean societies have patriarchal roots and so economic resources are primarily in the hands of males; but of the quality of life of female? How are they living in Jamaica? Is there is disparity between the quality of life of the sexes? However, a survey done by Rudkin found that women have lower levels of wellbeing (i.e. economic) than men (Rudkin 1993
227

222). This finding is further sanctioned by Haveman et al (2003) whose study reveal that retired mens wellbeing was higher than that of their female counterparts, because men usually received had more material resources, and more retired benefits compared to women ages 65 years and older. Thus with men receiving more than women, and having a more durable possession than women, their material wellbeing is higher is later life. Generally, from the United Nations statistical databases, life expectancy for male is lower than of females. This is particular true for females in the old aged cohorts (United Nations 2004; Moore et al. 1997). Moore et al. (1997) added, Females life expectancies are likely to remain above that for males [Elo 2001] for the foreseeable future, among both the population as a whole and the elderly (Moore et al. 1997, 12). Among the justification for the differential between life expectancy the sexes is linked with the health consciousness of women and their approach to preventative care. Unlike women, worldwide men have a reluctance to seek health-care

compared to their female counterpart. It follows in truth that women have bought themselves additional years in their younger years, and it is a practice that they continue throughout their life time which makes the gap in age differential what it is which is approximately a 4-year difference in Jamaica. A study conducted by McDonough and Walters (2001) revealed that women had a 23 percent higher distress score than men and were more likely to report chronic diseases compared to males (30%). It was found that men believed their health was better (2% higher) than that self-reported by females. McDonough and Walters used data from a longitudinal study named Canadian National Population Health Survey (NPHS). The study was initiated in 1994, and data were collected every second year for a duration of six years. The information was taken form
228

20,000 household members who were 12 years and older. A research carried out by a group of economists (Headey and Wooden) revealed that women are slightly more likely to report higher levels of life satisfaction than men (mean=78.3, compared with 77.1 for men (Wooden and Headey 2003, 14). Based on the nature of the study, subjective wellbeing and ill being, the reported wellbeing (measure by life satisfaction) of women is higher than that for men but that males have a higher financial wellbeing than females (Headey and Wooden 2003, 16). Thus, the discourse is inconclusive and we will not add to the literature in this regard but will examine quality of life of women as this is the first of its kind in Jamaica and in the wider Caribbean literature. Theoretical Framework The overarching theoretical framework that will be adopted in this study is an econometric model that was developed by Grossman (1972), and further modified by Smith and Kington 1997. The initial model (i.e. Eqn. [1]) by Michael Grossman reads: Ht = (Ht-1, Go, Bt, MCt, ED) [1] where Ht current health in time period t, stock of health (Ht-1) in previous period, Bt smoking and excessive drinking, and good personal health behaviours (including exercise Go), MCt,- use of medical care, education of each family member (ED), and all sources of household income (including current income)- (Smith and Kington 1997, 159-160). Grossmans model further expanded upon by Smith and Kington to include socioeconomic variables (Eqn.2). Ht = H* (Ht-1, Pmc, Po, ED, Et, Rt, At, Go) ....Eqn. [2]

229

Eq. (2) expresses current health status Ht as a function of stock of health (Ht-1), price of medical care Pmc, the price of other inputs Po, education of each family member (ED), all sources of household income (Et), family background or genetic endowments (Go), retirement related income (Rt ), asset income (At,). Thus, the current paper will test this general hypothesis in seeking to establish a quality of life model for Jamaican Women (Eqn. [3]): QoLj = (Gj, PPIj, Yj, Rj , lnAj , Ci , Oj , Tj , SSj , ARj , Xj , Ej , ESj , RAj , WSj i).Eqn.[3] Method The current paper uses a sample of 723 women, with a mean age of 34.33yrs 13.4 yrs. This study is taken from a general study conducted by the Centre of Leadership and Governance, Department of Government, The University of the West Indies between July and August 2006 of some 1,338 Jamaicans. The survey uses a questionnaire of some 166 items, which probes issues relating to the orientation of democracy, leadership and governance in Jamaica. The survey was a stratified random sample of the fourteen parishes of Jamaica, using the descriptive research design. Data were collected and stored using the Statistical Packages for the Social Sciences (SPSS). Descriptive statistics were done to provide background information on the sample; tests were done for Cronbach alpha to examine the validity of the construct i.e. wellbeing and political participation. Then, multiple regressions were used to build a model for quality of life of Jamaican Women. Measures: 5 QoLj = 1/10Lij , where Li i=15 denotes each Need Item of Abraham Maslows 5-Need Hierarchy i=1 (Each is a 10-point Likert Scale: Health status; Basic Necessities; Social Needs; Self-Esteem; Self-Actualization). Reliability analysis of the 5-Need Likert Scale Item is 0.748 (or 75%). Quality of Life Index ranges from: 1Quality of Life Index10; where from 1 to 3.9 are low, 4 to 6.9 are moderate and with high being from 7 to 10. Sex. Sex is the biological makeup of males and females. This is a binary measure, where 1=male and 0=female.
230

Area of residence. This means the geographic location of ones place of abode It is a dummy variable, 1=St. Andrew, Kingston and St. Catherine, 0=Other1 Subjective Social Class. This is peoples perception of their social and economic position in life, based on social stratification. socialcl1 1=Middle class socialcl2 1=Upper class Referent group is lower class. Interpersonal Trust. The survey instrument asked the question Generally speaking would you say that most people are essentially good and can be trusted, or that most people are not essentially good and cannot be trusted. The variable was then dummied, 1 if most people essential good and can be trusted, 0 if otherwise. Trust is on a continuum, and so low trust is a proxy for distrust. Occupation is a dummy variable, 1 if in high occupation, 0 if otherwise. Those categories which are classified within this are teachers, doctors, lawyers, businessmen, managers and/or supervisors whereas in the low category the following were includes farmers, tradesmen, unskilled worker, shopkeeper, haggler, vendor, office workers and so on. Confidence in sociopolitical institutions. This is the summation of 22 likert scale questions, with each question on a scale of (4) a lot of confidence, (3) some confidence, (2) a little confidence, to (1) no confidence. The heading that precedes the question reads: I am going to read to you a list of major groups and institutions in our society. For each, tell me how much CONFIDENCE you have in that group or institution. Confidence index = summation of 22 items, with each question being weighted equally; and 0confidence index88, with a Cronbach for the 22-item scale being 0.896. The higher the scores, the more people have confidence in sociopolitical institutions within the society. Thus, the confidence index is interpreted as from 0 to 34 represents very little confidence; 35 to 61 is low confidence; 62 to 78 is moderate confidence and 79 to 88 is most confidence. Age. Age is a continuous variable, which is recorded in years. Religiosity. The frequency with which people attend religious services, which does not include attending functions such as (1) graduations, (2) weddings, (3) christenings, (4) funerals. This variable was recorded as: Religiosity1 1=High religiosity (i.e. church attendances more than once per week) Religiosity2 1=Moderate religiosity (i.e. church attendance once per week or fortnightly) Referent group is low religiosity (i.e. none to several times per year) Income. Income is an ordinary variable with twenty-categories, ranging from (1) under $5,000 to (20) $250,000 and above. Based on the nature of this variable, it will be treated as a continuous variable.
1

Others constitute St. Thomas, Portland, St. Mary, St. Ann, Trelawny, St. James, Hanover, St. Elizabeth, Westmoreland, Manchester, and Clarendon.

231

Political Participation Index. Based on Trevor Munroes work, political participation ...the extent to which citizens use their rights, such as the right to protest, the right of free speech, the right to vote, to influence or to get involved in political activity (Munroe, 2002:4; Munroe, 1999:33), w use that construct to formulate a PPI = bi, bi 0, and bi represents each response to a question on political behaviour, such as voting, involvement in protest, with 0PPI19. The Cronbach alpha for the 22-item scale, which is used to constitute this Index, is 0.828. Governance of the country, G, is defined as peoples perception of administration of the society by the elected officials. This is a dummy variable, where 1 denotes in favour of a few powerful interest groups or the affluent, 0 is otherwise Extent of the Welfare System of governance: Results: Sociodemographic Characteristics of Sampled Population The findings of the current research has a sampled population of 723 women ages 16 to 85 years with a mean age of 34.3 years 13.4 years. Most of the respondents report that they are Blacks (78%) with some indicates Browns i.e. Mixed - (14%), Caucasians (6%). Approximately 6 out of 10 women indicate that they are in the lower class. The demographic characteristics of the sample also reveal that approximately 7 out of every 10 women indicate that they are employed (i.e. full-time, part-time, temporarily, seasonally and self-employed). On an average the quality of life of the sample was high (i.e. 6.8 1.7: Range 10: 10 0). Furthermore, the findings (Table 9.2) indicate that political participation for Jamaican Women is low (i.e. 3.6 3.5: Range 17: 17 0). On the contrary, the population has moderate confidence in the various socio-political institutions in Jamaica (56.310.8: Range 79:86 7); with a sample report a high welfare system of governance of the Jamaican state. Insert Table 9.2 here Findings: Multivariate Analysis Using econometric analysis (i.e. multiple regressions) of the surveyed research data of some 723 Jamaican women we found that the final model (Eqn. [3]) explains 18.5% (Adjusted R232

squared) of some 6 variables. The model is a good fit (F statistic [15, 410] = 7.413, p value = 0.001]. (Table 9.3). QoLj = (Gj, PPIj, Yj, Rj , lnAj , Oj , Tj , SSj , ARj , Xj , Ej , ESj , RAj , WSj ,j).Eqn.[3]

where QoLj Gj person j; PPIj Yj, Rj , lnAj , Oj , Tj , SSj , ARj , Xj , Ej , ESj , RAj , WSj ,

the quality of life of person j; self-reported administration of the governance of the nation of political participation index of person j income of person j religiosity of person j logged age of person j occupation of person j interpersonal trust of person j subjective social class of person j area of residence (i.e. parish of residence) of person j gender of respondent of person j educational level of person j employment status of person j ethnicity of person j extent of welfare state of a nation as reported by person j

QoLij = (Yj, Rj ,Tj , Gj, SSj , ESj , j)...Eqn.[4] Examination of the sociodemographic determinants in Eqn. [4] revealed that subjective social class middle class with referent to lower class - (Beta = 0.198; 95%CI:0.380, 1.070) is the most influential factor followed by employment status (Beta = 0.167; 95%CI:0.304, 0.1.077), income (Beta = 0.155; 95%CI:0.015, 0.101), religiosity- high religiosity - (Beta = 0.152; 95%CI:0.214, 0.974), the administration of the governance of the nation (Beta = -0.139; 95%CI:-0.893, -0.203) and lastly by interpersonal trust (Beta = 0.094; 95%CI:0.020, 0.676) (Table 9.3). Insert Table 9.3 here

233

Further examination of the findings will now be forwarded to provide a more in-depth understanding of determinants in model (i.e. Eqn. [4]). An individual who is in the self-reported middle class with referent to lower class contributes the most to quality of life of Jamaican Women. However, those in the upper class with referent to lower class contribution are

marginally more than interpersonal trust that influence is the least. A woman who trusts other people has a greater quality of life compared to another who reported that she does not trust other people. A similar result was observed for employment status as an employed woman has greater quality of life compared to those who are unemployed, and the greater the income of a person the higher is the quality of life of that individual. Religiosity is the fourth most significant factor of quality of life of sampled population. The religiosity with which we speak is high church attendance (i.e. church attendance at least twice per week) with referent to low religiosity (i.e. from no church attendance to once per year). Those who reported that the governance of the nation (i.e. political administration) benefits mostly equally with referent to those who indicated that it favours the rich have a lower quality of life. In addition to what has been reported so far, those who cited being in moderate religiosity had a greater quality of life compared to those with had a low religiosity. Thus, a womans quality increases with greater church attendance. Limitation of the Model Although the current model used data from a cross-sectional study by way of stratified probability sampling technique, it has an adjusted R-square of less than 20%. Some statisticians argue that a cross-sectional study that is less than 30% and over is not a good predictor of the phenomenon. The current research is the first of its kind, and is more so a platform for future studies than a conclusion on the matter of quality of life of women in Jamaica. Discussion

234

The physiological composition of the sexes explains the rationale of some typologies of diseases affecting a particular sex (WHO 2005). One health psychologist, Phillip Rice, in concurring with WHO, argued that differences in death and illnesses are the result of differential risks acquired from functions, stress, life styles and preventative health practices (Rice 1998). Biomedical studies showed that there are gender specific diseases. The examples here are prostate cancer (affect only men) and cervical cancer (plague only women). Rice believed that this health difference between the sexes is due to social support. According to Rice (1998),

Rodin and Ickovics (1990) this can be explained by epidemiological trends. Lifestyle practices may justify the advantages that women enjoy compared in men concerning health status. However, a survey done by Rudkin found that women have lower levels of wellbeing (i.e. economic) than men (Rudkin 1993 222). This finding is further sanctioned by Haveman et al (2003) whose study reveal that retired mens wellbeing was higher than that of their female counterparts, because men usually received had more material resources, and more retired benefits compared to women ages 65 years and older. Thus with men receiving more than women, and having a more durable possession than women, their material wellbeing is higher is later life. The issue extends beyond those two types of chronic illnesses as Courtenay (2003) noted from research conducted by the Department of Health and Human Services (2000) and Centers for Disease Control (1997) that from the 15 leading causes of death except Alzheimers disease, the death rates are higher for men and boys in all age cohorts compared to women and girls. Embedded within this theorizing are the differences in fatal diseases that are explained by gender constitution (Seltzer and Hendricks 1989, 7), to which Courtenay (2003) explained are due to behavioural practices of the sexes and goes to explain the fact that men are dying 6 years earlier
235

than females (U.S. Preventive Services Task Force, 1996). The current research does not expand on past literature, but provides new information on factors that explain variability in quality of life of females (or women) in Jamaica. Among the fundamental characteristics of research are that adding something new to the discourse, modifying what exists and so in keeping with these epistemological traditions, we will maintain these traditions in the current work. Religion is gender bias, and this dates back to nations slavery past. In contemporary Jamaica, church attendance is substantially a woman issue; and many theologians continue to argue that there are reality benefits to have from this practice. It is well accepted that religiosity is positively associated with wellbeing; and that is goes beyond the theologians views (Krause 2006; Moody 2006: Jurkovic and Walker 2006; Ardelt 2003; Graham et al. 1978). According to Kart (1990), religious guidelines aid wellbeing in that through restrictive behavioural habits which are health risk such as smoking, drinking of alcohol, and even diet. The current paper has concurred with the literature that religiosity is positively associated with quality of life; and this is the fourth most influential predictor of quality of life of a Jamaican Woman. We go further to say that the quality of life Jamaican Woman is the highest when she has the greatest degree of church attendance followed by moderate religiosity and lastly by the lowest religiosity. Traditionally income was used to proxy wellbeing (i.e. economic wellbeing), and that Richard Easterlin (2001a, 2001b) showed that income is important to happiness, but that income does not buy unlimited happiness. In a paper titled Poverty and Health, Murray (2006) argued that there is a clear interrelation between poverty and health. She noted that financial inadequacy prevents an individual from accessing food and good nutrition, potable water, proper
236

sanitation, medicinal care, preventative care, adequate housing, knowledge of health practices and attendance at particular educational institutions among other things. The issue of resource insufficiency affects the ability and capacity of the poor from accessing the quality of goods and services comparable to the rich that are better able to add value to wellbeing. This is succinctly forwarded by Murray in her monograph that: Poverty also leads to increased dangers to health: working environments of poorer people often hold more environmental risks for illness and disability; other environmental factors, such as lack of access to clean water, disproportionately affect poor families (Murray 2006, 923) Michael Grossmans work had established the direct link between income and health (Smith and Kington 1997; Murray 2006; Sen 1999); and that incomes contribution to the quality of life of a Jamaican Woman is highly important as the current paper reveals that it (income) is the third most influential factor in determining quality of life of the sampled population. This contradicts the work of Edward Diener. Diener (1984), states that the correlation between income and subjective wellbeing was small in most countries. According to Diener (1984, 11), , there is a mixed pattern of evidence regarding the effects of income on SWB [subjective wellbeing]. The current research was subjective wellbeing (i.e. self-reported quality of life

using Abraham Maslows 5 Need Item scale), and it shows that income is the third most valued predictor of quality of life of a Jamaican woman. Arendt, using ordered logistic models, found that it cannot be rejected that the income effects are causal and this proceeded the finding that [a] robust relations exist between income and some measures of wellbeing of [the] elderly (Ardent 2005, 327). Although the current work counter the findings of Edward Dieners work (1984), what contributes the most to quality of life of a Jamaican woman? The answer is social class followed by employment status. The quality of life of a
237

Jamaican Women is primarily determined by her social class; with middle class women having the greatest quality of life and working class female experiencing the least quality of life. In this study education was not related to quality of life, which contravenes the finding of many studies (Diener 1984; Grossman, 1972; Hambleton et al. 2005; Bourne, 2007). Hambleton et al.s work, on the other hand, found that the statistical relation was a weak one. Employments contribution to the quality of life of woman is highly important because of significant of employment in socio-economic independent, opportunities, choices and freedom and power of independency in this regard. Conclusion In summary, the factors of quality of life of a Jamaican Woman are social class, employment, income and religiosity, with social class being the most influential of all the variables. Employment does not merely about the income, but it is about the independence, the choices, the sense of freedom, the positive psychological attributes that this freedom gives as well as the self-advancement that it is likely to provide why this variable is of that importance in determining the quality of life of Women. The current work does not provide all the answers, but it is a catalyst upon which we are able to build, modified and refute as these are pillows upon which research is based.

238

Reference Ali, A., Christian, D., & Chung, E. (2007). Funny Turns in an Elderly Man. West Indian Medical Journal, 56:376-379. Ardelt M. Effects of religion and purpose in life on elders subjective wellbeing and attitudes toward death. Journal of Religious Gerontology 2003;14:55-77. Arias, O. Are Men Benefiting from the New Economy? Male Economic Marginalization in Argentina, Brazil, and Costa Rica (December 18, 2001). World Bank Policy Research Working Paper No. 2740; 2001. Available at SSRN: http://ssrn.com/abstract=634452 Bourne, P. Determinants of well-being of the Jamaican Elderly. Unpublished thesis, The University of the West Indies, Mona Campus; 2007. Centers for Disease Control. Demographic differences in notifiable infectious disease morbidity United States, 1992-1994. Morbidity and Mortality Weekly Report 1997; 46:637- 641. Chevannes, B. Learning to be a man: Culture, socialization and gender identity in five Caribbean communities. Kingston, Jamaica: The Univer. of the West Indies Press; 2001. Courtenay WH. Key determinants of the health and wellbeing of men and boys. International Journal of Mens Health 2003;2:1-30. Cummins RA. Objective and subjective quality of life: an interactive model. Social Indicators Research 2000;52:5572. Department of Health and Human Services. Deaths: Final data for 1998 (DHHS Publication No. [PHS] 2000-1120) National Vital Statistics Reports, 2000;48. Hyattsville, MD: National Center for Health Statistics. Quoted in Courtenay, WH. Key determinants of the health and well-being of men and boy. International Journal of Mens Health 2003;2:1-30. Diener E. Subjective wellbeing. Psychological Bulletin 1984;95:542-575. Diener E. Subjective Well-Being: The Science of Happiness and a Proposal for a National Index. American Psychological Association 2000;55:34-43. Easterlin RA. Will raising the incomes of all increase the Happiness of all. Journal of Economic Behavior and Organization, 1995;27:35-47. Easterlin RA. Income and happiness: Towards a unified theory. Economic Journal 2001a;111: 465-484 Easterlin RA. Life cycle welfare: Evidence and conjecture. Journal of Socio-Economics 2001b; 30:31- 61. Eldemire D .The elderly A Jamaican perspective. Grell, Gerald A. C. (ed). 1987. The elderly in the Caribbean: Proceedings of continuing medical education symposium. Kingston, Jamaica: University Printery; 1987a. Eldemire D. The clinicals approach to the elderly patient. G. Grell. ed. 1987. The elderly in the Caribbean: Proceedings of continuing medical education symposium. Kingston, Jamaica: University Printery; 1987b. Eldemire D. The elderly and the family: The Jamaican experience. Bulletin of Eastern Caribbean Affairs 1994;19:31-46. Eldemire D. A situational analysis of the Jamaican elderly, 1992. Kingston: Planning Institute of Jamaica; 1995a.
239

Eldemire D. The elderly in Jamaica: A gender and development perspective. In: Robert JH, Kitts J, Arsenault LJ, eds. Gender, health, and sustainable development. Perspective from Asia and the Caribbean. Proceedings of workshops held in Singapore, 23-26 January 1995 and in Bridgetown, Barbados, 6-9 December 1994. Ottawa: International Development Research Centre; 1995b. Eldemire D. Older women: A situational analysis, Jamaica 1996. New York: United Nations Division for the Advancement of Women; 1996. Eldemire D. The Jamaican elderly: A socioeconomic perspective and policy implications. Social and Economic Studies 1997;46: 175-193. Gayle, Herbert. Adolescent Male Survivability in Jamaica. Kingston, The Jamaica Adolescent Reproductive Health Project (Youth. now); 2002. Graham TW, Kaplan BH, Cornoni-Huntley JC, James SA, Becker C, Hames CG, Heyden S. Frequency of church attendance and blood pressure elevation. Journal of Behavioral Medicine 1978;1:37-43. Grossman M. The demand for health a theoretical and empirical investigation. New York: National Bureau of Economic Research; 1972. Hambleton IR., Clarke K, Broome H L, Fraser H S, Brathwaite F, Hennis A J. Historical and current predictors of self-reported health status among elderly persons in Barbados. http://journal.paho.org/index.php?a_ID=290 (accessed March 22, 2006); 2005. Haveman R, Holden K, Wilson K, Wolfe B. Social security, age of retirement, and economic wellbeing: Inter-temporal and demographic patterns among retired-worker beneficiaries. Demography 2003;40:369-394. Hutchinson G, Simeon DT, Bain BC, Wyatt GE, Tucker MB, LeFranc E. Social and health determinants of wellbeing and life satisfaction in Jamaica. International Journal of Social Psychiatry 2004; 50;1:43-53. Jurkovic D, Walker GA. Examining masculine gender-role conflict and stress in relation to religious orientation and spiritual wellbeing in Australian men. Journal of Mens Studies 2006;14:27-46. Kart CS. The Realities of Aging: An introduction to gerontology, 3rd. Boston, United States: Allyn and Bacon; 1990. Krause N. 2006. Religious doubt and psychological wellbeing: A longitudinal investigation. Review of Religious Research 2006;47:287-302. Lyubomirsky S. Why are some people happier than others? The role of cognitive and motivational process in well-being. American Psychologist 2001;56:239-249. McDonough P, Walters V. Gender and health: reassessing patterns and explanations. Social Science and Medicine 2001;52:547-559. Miller, Errol. 1991. Men at Risk. Kingston. Kingston: Jamaica Publishing House. Miller, Errol. 1986. Marginalization of the Black Male. Kingston: Kingston Publishers. Moore EG, Rosenberg MW, McGuinness D. Growing old in Canada: Demographic and geographic perspectives. Ontario: Nelson; 1997. Murphy H, Murphy EK. Comparing quality of life using the World Health Organization Quality of Life measure (WHOQOL-100) in a clinical and non-clinical sample: Exploring the role of selfesteem, self-efficacy and social functioning. Journal of Mental Health 2006;15:289 300. Murray S. Poverty and health. Canadian Medical Association Journal 2006;174:923-923.
240

Planning Institute of Jamaica, (PIOJ). Economic and Social Survey, 1990-2007. Kingston: PIOJ; 1990-2007. Rice PL. Health psychology. Los ngeles: Brooks/Cole; 1998. Rodin J, Ickovic JR. Womens health: Review and research agenda as we approach the 21st century. American Psychologist 1990;45:1018-1034. Rudkin L. Gender differences in economic wellbeing among the elderly of Java. Demography 1993;30:209-226. Seltzer MM, Hendricks JA. On your marks: Research issues on older women. In: Hendricks JA, ed. Health and economic status of older women ed. New York: Baywood Publishing; 1989. Sen A. Development as freedom. Oxford: Oxford University Press; 1999. Smith JP, Kington R. Demographic and economic correlates of health in old age. Demography 1997;34:159-170. Stutzer A, Frey BS. Happiness and Economics. Princeton University Press; 2001. Stutzer A, Frey BS. Reported subjective well-being: A challenge for economic theory and economic policy; 2003. http://www.crema-research.ch/papers/2003-07.pdf (accessed August 31, 2006). U.S. Preventive Services Task Force. Guide to clinical preventive services 2nd ed. Baltimore: Williams & Wilkins; 1996. In: Courtenay WH. Key determinants of the health and wellbeing of men and boys. International Journal of Mens Health 2003;2:1-30 United Nations. Population, ageing and development. San Juan, Puerto Rico: UN. ECLAC; 2004. http://www.globalaging.org/agingwatch/events/regionals/eclac/popagingdevsanjuan2004.pdf (accessed October 14, 2006). Wooden M, Headey B. The effects of wealth and income on subjective well-being and ill-being. Australia: Melbourne Institute of Applied Economic and Social Research 2004. http://melbourneinstitute.com/wp/wp2004n03.pdf (accessed June 29, 2006). World Bank. World development indicators. Washington, D.C.: IRDB; 2005.

Acknowledgement The authors would like to single out the Centre for Leadership and Governance, Department of Government, The University of the West Indies, Mona, Jamaica for allowing them to utilize the dataset which facilitates this study.

241

Table 9.1: Expectation of Life at Birth by Sex, 1880-1991, Jamaicans Period Average Expected Years of Life at Birth Male Female e0 e0 1880-1882 37.02 39.80 1890-1892 36.74 38.30 1910-1912 39.04 41.41 1920-1922 35.89 38.20 1945-1947 51.25 54.58 1950-1952 55.73 58.89 1959-1961 62.65 66.63 1969-1970 66.70 70.20 1979-1981 69.03 72.37 1989-1991 69.97 72.64 1999-2001 70.94* 75.58* 2002-2004 71.26 77.07 Sources: Demographic Statistics (1972-2006); Statistical Yearbook of Jamaica, 1999 and * Economic and Social Survey, Jamaica 2005 (Quoted in Bourne, 2007, p. 150) Note e0 is life expectancy at birth

242

Table 9.2: Demographic Characteristics of Sampled Population, N=723 Number Subjective Social Class Working (lower) class Middle class Upper class Ethnicity Caucasian Blacks Browns Other Educational Level No formal Education Primary/Preparatory and All Age school Secondary Post-secondary Tertiary Employment Status Unemployed Employed Age Quality of Life Political Participation Index Extent of Welfare System of governance Confidence in sociopolitical institution index 409 259 29 46 562 104 9 8 116 246 127 195 222 494 Percent 58.7 37.2 4.2 6.4 77.9 14.4 1.2 1.2 16.8 35.5 18.4 28.1 31 69

34.33yrs 13.4 yrs.: Range 69: 85 16 yrs. 6.8 1.7: Range 10: 10 0. 3.6 3.5: Range 17: 17 0. 6.8 1.5: Range 8.7:10 1.2. 56.310.8: Range 79:86 7.

243

Table 9.3: Quality of Life of Jamaican Women by Some Explanatory Variables Unstandardized Coefficients Variable Beta P Coefficient Std. Error (Constant) 4.317 1.060 0.000 Tertiary Education 0.252 0.225 0.064 0.263 Religiosity (1=High) 0.594 0.193 0.152 0.002 Religiosity (1=Middle) 0.466 0.194 0.116 0.017 Area of Residence -0.031 0.204 -0.007 0.879 Extent of Welfare 0.034 0.053 0.028 0.522 System of Governance Dummy Occupation 0.050 0.209 0.013 0.813 (1=Lower level ) socialcl1 0.725 0.175 0.198 0.000 socialcl2 0.820 0.387 0.096 0.035 Trust 0.348 0.167 0.094 0.038 Dummy Governance(1=Benefit s most equally, -0.548 0.176 -0.139 0.002 0=Favours Rich) Income Employment status Race2 (1=black and brown) Index of Political Participation lnAge 0.058 0.691 0.235 -0.042 0.222 0.022 0.197 0.293 0.022 0.254 0.155 0.167 0.036 -0.088 0.044 0.008 0.000 0.424 0.063 0.383

CI (95%) 2.233 -0.189 0.214 0.085 -0.432 -0.071 -0.362 0.380 0.060 0.020 -0.893 0.015 0.304 -0.341 -0.086 -0.277 6.401 0.693 0.974 0.847 0.370 0.139 0.461 1.070 1.581 0.676 -0.203 0.101 1.077 0.810 0.002 0.721

R = 0.462 R2 = 0.213 Adjusted R2 = 0.185 N=425 F-test [15, 410] = 7.413, P = 0.001< 0.05 Standard error of the estimate 1.598

244

Table 9.4: Particularized Labour Force Indicators By Sex, 1990-2007 Year 1990 Male: Labour Force (000s) Employed Labour Force (000s) Unemploymen t Rate (in %) Female: Labour Force (000s) Employed Labour Force (000s) Unemploymen t Rate (in %) Compiled by Paul A. Bourne from Planning Institute of Jamaica (in Economic and Social Survey 1990 2007) 9.1 22.2 22.8 22.4 21.8 22.5 23.1 23.5 22.1 22.4 22.3 21.0 20.7 17.6 16.4 15.8 14.5 14.5 494. 0 513. 1 500. 7 389. 6 504. 8 389. 7 511. 7 397. 1 515. 8 403. 2 532. 2 412. 4 528. 2 406. 5 520. 0 397. 9 514. 2 400. 7 507. 4 393. 6 490. 3 486. 7 384. 7 506. 1 401. 6 487. 7 402. 3 531. 3 444. 3 529. 1 445. 6 557. 5 476. 9 562. 2 480. 8 15.3 9.4 9.5 10.9 9.6 10.8 10.0 10.6 10.0 10.0 10.2 10.3 10.6 9.7 7.9 7.6 7.0 6.2 564. 6 896. 3 571. 8 518. 1 570. 1 516. 0 571. 3 509. 2 574. 8 519. 9 617. 9 551. 0 614. 6 553. 3 613. 8 549. 0 614. 3 552. 9 611. 7 550. 3 615. 0 552. 4 618. 1 554. 8 618. 4 552. 8 611. 1 552. 3 663. 5 610. 9 661. 9 611. 4 695. 6 646. 8 699. 1 656. 1 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

245

10
Examining Health Status of Women in Rural, Peri-urban and Urban Areas in Jamaica
Paul A. Bourne, Denise Eldemire-Shearer, Donovan McGrowder and Tazhmoye Crawford 3

A comprehensive review of the literature revealed that less information is available in literature on health status of women, and health status of women in 3 geographical zones in Jamaica. This study examined data on the health status of women in Jamaica in order to provide some scientific explanation of those factors that account for their health status; and differences based on area of residence. Rural women had the lowest health status (OR = 0.819, 95% CI = 0.679-0.989) among all women (peri-urban OR = 1.054, 95% CI = 0.8421.320; urban OR = 1.00) and that they were the least likely to have health insurance coverage. Health insurance was the critical predictor of good health status of women in Jamaica, and this was equally the same across the 3 geographic areas; and that married women were 1.3 times more likely (OR 1.3, 95 CI = 1.036-1.501) to report good health compared to those who were never married. This study provides an understanding of womens health status in Jamaica as well as the disparity which correlates based on the different geographical regions.

Background
Latin America and the Caribbean have the second highest urbanization level in the world. For every 13 persons there are in the region, 10 of them live in cities (78.3% in 2007) [1]; and 20 of the regions largest cities are home to nearly 20% of its population. Jamaica is a predominantly
246

Afro-Caribbean society, 75% black and 13% mixed, with a class structure based on land and wealth rather than race. Although a developing country, it possesses features of a developed country. While there is much industrialization and modernization, customs, cultural and social habits of several centuries are common-place. Jamaica is the third largest English speaking Caribbean island (total area of 10,991 km2) with an estimated population of 2.7 million (2007). The country is classified into three geographical planes (Cornwall, Middlesex and Surrey) and has 14 parishes. Cornwall covers the Western belt which includes parishes such as Westmoreland, Hanover, St. James, St. Elizabeth. Middlesex constitutes the middle proportions of the island with parishes such as Clarendon and St. Catherine. Surrey comprises the Eastern region with parishes such as Kingston, St. Thomas and Portland. Another classification is cities (urban areas) which constitute 27.3% of the population, peri-urban 30.2% and rural areas, 42.5% in 2007. In 2007, Jamaicas poverty rate was 9.9%, and this was 15.3% in rural areas, 4.0% in periurban areas and 6.2% in urban areas. Furthermore, the mean annual per capita consumption for country was US $2,059.91 while it was US $2,736.60 for urban dwellers, US $2,231.04 and US $1,513.17 for rural Jamaicans. Statistics for the same period showed that the sex ratio of the population was 97 per 100 and 84 per 100 for older ages (60 years and over). This indicates that there are marginally less men than women in the population, and an even greater feminization at older ages. It was estimated that 10.9% of the population was 60 years and over which indicates an ageing population that began in the 1960s [2-4], 28.3% under 15 years, and 53.5% in the reproductive years of 15 to 49 years. Women comprised 50.7% of the population and elderly women accounted for 13.0% compared with 11.4% for elderly men. It was also found that 46.6%
247

of household heads were women; life expectancy at birth for women was 77.1 years (2002 to 2004). The unemployment rate in 2007 was 65.4% for women [3-5] with women participation rate being 55.4% compared to 72.9% for men. Fifty-three percent of women in the poorest quintile were heads of households compared to 46.9% of men. An important difference between the sexes was the mean annual per capita consumption. Statistics revealed that the mean annual consumption for male headed-household was US $2,188.03 compared to US $1,892.92 for female headed-household. It is well established that health status is determined by socio-physiological factors (age, income, education, culturalization, crime and negative psychology) and that lifestyle practices also account for good (or poor) health status [6-8]. Womens health therefore is intricately a mix of socio-physiological response or outlay and is expressed through behaviour relating to culture, religion, and legal norms [6]. Although recent attention has been directed towards exploring the ramifications of womens health in the Western Hemisphere including the Caribbean, an extensive review of the literature revealed that only a few studies have examined health determinants of women in the Caribbean, including Jamaica [7, 8]. Using secondary data from a stratified probability survey on political culture of 1,338 respondents, Bourne [7] extracted a sample of 722 women in investigating the determinants of quality of life of women in Jamaica. The study showed that the mean quality of life of Jamaican women was moderately high (6.8 out of 10; SD =1.7). Six variables (social class, employment, income, religiosity, governance of the nation and interpersonal trust) accounted for 18.5% of the variability of quality of life. Eldermire [8] investigated the general life situation of elderly Jamaican women and found that their life situations were on an average good.
248

Many economic indicators showed that women are disadvantaged in Jamaica and the wider Caribbean when compared to men [9-11]. In 2007, Statistics for Jamaica showed that the mean consumption per capita on food was US $2,378.39 for male head-household compared to US $1,898.56 for women. Studies have showed that women seek more health care then men [12-14], and that this commences in earlier childhood. Therefore, the similarities and dissimilarities based on area of residence of women was examined (via econometric models) in order to determine the composition of womens health status. Econometric models such as Bournes health determinant model [15] denotes that an individuals health is a function of cost of medical care and other factors such as: educational level, age, the environment, gender, marital status, area of residence, psychological status which include positive and negative affective status, occupancy per room, home tenure, property, and crime and victimization. Bournes work modelled health determinants of Jamaicans, and with the aforementioned issues surrounding women as were outlined above, a study on Jamaicans is not necessarily providing an understanding of womens health and significant of particular factors determining their health status or the disparity in health status of women based on the 3 geographic sub-regions in the island. This study sought to examine 1) the consumption expenditure of women in the different income quintiles (or social classes); 2) health insurance coverage, and visits to health care facilities by area of residence; 3) health status by age cohorts (ie young, other adults and elderly women); 4) diagnosed illness by age cohorts; diagnosed illness by area of residence; 5) the health status of women in Jamaica using a modification of Bournes health determinant model; 6) the health status of women in and sub-regions namely urban, periurban and rural residence; and 7) the strength of those factors which affect health status of women in the nation and the sub-regions.
249

Materials and methods


Materials and Methods The sub-sample for the current paper was all 8,541 women (ages of 15 to 100 years) extracted from a nationally representative cross-sectional survey of 25,018 Jamaicans, the Jamaica Survey of Living Status (JSLC, 2002) [16]. This survey was drawn using stratified random sampling. The design was two-stage stratified random sampling, where there was a Primary Sampling Unit (PSU) and a selection of dwelling from the primary units. The PSU is an Enumeration District (ED), which constitutes a minimum of 100 dwellings in rural areas and 150 in urban areas. An ED is an independent geographic unit that shares a common boundary. This means that the country was grouped into strata of equal size based on dwellings (EDs). Based on the PSUs, a listing of all the dwellings was made and this became the sampling frame from which a master sample of dwelling was compiled which provided the sampling frame for the labour force. Ten percent was selected for the JSLC. The survey was weighted to represent the population of Jamaica. This study used JSLC 2002 which was conducted by the Statistical Institute of Jamaica (STATIN) and Planning Institute of Jamaica (PIOJ) between June and October 2002. The researchers selected this survey because it was the second largest sample size for the survey in its history (since 1988 to 1998), and in that year, the survey had questions on crime and victimization, and the physical environment unlike previous years. A self-administered questionnaire was used to collect the data, which was stored and analyzed using SPSS for Windows 16.0. The questionnaire was modeled from the World Banks Living Standards
250

Measurement Study (LSMS) household survey. There were some modifications to the LSMS as JSLC was more focused on policy impacts. The questionnaire covered questions such as: sociodemographic, economic and wealth, crime and victimization, social welfare, health status and services, nutrition, housing, immunization of infants and physical environment. The survey was weighted in order for it to represent the population. The non-response rate for the survey was 27.7%. Descriptive statistics such as mean, standard deviation (SD), frequency and percentage were used to analyze the socio-demographic characteristics of the sample. Chi-square was used to examine association between non-metric variables; an Analysis of Variance (ANOVA) was used to evaluate the relationships between metric and non-dichotomous categorical variables. Logistic regression examined the relationship between the dependent variable and some predisposed independent (explanatory) variables because the dependent variable was a binary one (selfreported health status, with 1 if good health status was reported and 0 if poor health). Results were presented using unstandardized B-coefficients, Wald statistics, odds ratio and confidence interval (95% CI). The predictive power of the model was tested using Omnibus Test of Model and Hosmer and Lemeshow [17] was used to examine goodness of fit of the model. The correlation matrix was examined in order to ascertain whether autocorrelation (or multicollinearity) existed between variables. Based on Cohen and Holliday [18], correlation can be low (weak), from 0 to 0.39; moderate, 0.4 to 0.69, and strong, 0.7 to 1.0. This was used to exclude (or allow) a variable in the model as any variable that had at least moderate correlation was excluded from the final model. Wald statistics was used to determine the magnitude (or contribution) of

251

each statistically significant variables in comparison with the others, and the odds ratio (OR) for interpreting each significant variables. Multivariate regression framework [19, 20] was used to assess the relative importance of various demographic, socio-economic characteristics, physical environment and psychological characteristics in determining the health status of women in Jamaica. This approach allowed for the analysis of a number of variables simultaneously. Secondly, the dependent variable is a binary dichotomous one which has enabled the use of this statistic technique to be utilized in the past to do similar studies. Having identified determinants of health status from previous studies, using logistic regression techniques, final models were build for women in general as well as for each geographical sub-regions (rural, peri-urban and urban areas) using only those predictors that independently predict the outcome. The level of significance for this study is 95% (ie P < 0.05). Equation 1 is a modification of Bourne [21, 22] health determinant model which was previously used to determine the health status of the elderly in Jamaica.
Hi = (Wi, HHi, Pmci, Ci, MRi, ARi, EDi, SSi, CRi, (NAi, PAi), Mi, Fi, CHi, At, Ai, HIi, LLi, Eni, Yi, Vi,i)

(1)

The health status of person i, Hi, is a function of Wi, the two wealthiest quintiles of person i with 1 if yes or 0 for the two poorest quintiles; HHi, household head of person i, with 1 if yes or 0 if otherwise; Pmci, cost of medical care of person i, in United States (US) dollars; Ci, average consumption per person in household, in US dollars; MRi,is marital status of person i; ARi,, area of residence of person i; EDi, educational level of person i; SSi, having social support of person i with 1 if yes or 0 if no; CRi, crowding of person i, in numbers; (NAi, PAi), psychological status which is the summation of negative affective status of person i, NAi where values are in
252

continuous number; PAi, positive affective psychological status of person i, where values are in continuous numbers; Mi, number of men in household of person i; Fi, number of women in household of person i; CHi, number of children below the age of 14 years of person i; At, asset owned of person i, in continuous numbers; Ai, age of person i, in continuous numbers; HIi, of private health insurance (proxy ); LLi, living arrangement where 1 is living with family members or relative, 0 if otherwise; Eni, physical environment of person i, with 1 if affected by flood, landslides, soil erosion or 0 if not affected; Yi, average income per person in household (this variable is proxied by total expenditure); Vi, crime of person i, where values are continuous numbers, and i is the residual error. Measures Self-reported health status is self-assessed illness (cold, diarrhoea, asthma attack, hypertension, diabetes mellitus or any other illnesses) reported by respondents in the last 4-weeks of the survey period. Good health status is a dummy variable; where 1 is good health (not reporting an ailment, injury or dysfunction) and 0 is poor health (self-reported illness, injury or ailment). Household crowding is the average number of persons living in a room excluding kitchen, bathroom and verandah. Physical environment is the summation of responses reported by respondents on suffering landsides; property damage due to rains, flooding; or soil erosion in the last 4-weeks. Psychological conditions are the psychological state of an individual, and this is sub-divided into positive and negative affective psychological status. Positive affective psychological status refers to the number of responses that are hopeful and optimistic about the future and life
253

generally. Negative affective psychological status refers to the number of adverse events occurred to the respondents over the last 4-week period. Each event was equally weighted. Age is the number of years lived, which is also referred to age at last birthday. This is a continuous variable, ranging from 15 to 100 years. Age group is classified into three sub-groups. Young are ages 15 to 30 years, other adults 31 to 59 years, and elderly 60 years and over. Age is used as a continuous variable for the logistic regressions. Crime and victimization index (crime index) measures the number of cases and severity of crimes committed against a person or his/her family members but not against property. Social support (or network) denote different social networks with which the individual has or is involved (1= membership of and/or visits to civic organizations or having friends that visit ones home or with whom one is able to network, 0 = otherwise). Living arrangement denotes whether the individual is living alone or with family, friends or associates; where 1 = living with family members or relatives, and 0 = otherwise.

Results
Demographic characteristics of sampled population The sub-sample consisted of 8,541 respondents (ages 15 to 100 years), with a mean age of 40.1 years (SD 19.29 years). Of the sub-sample of respondents, 65.2% were never married, 24.7% married and 10.1% divorced, separated or widowed. The mean annual consumption per person per household was US$762.35 (SD US$917.81) (rate in 2002: 1US$ = Ja$50.97) with the
254

maximum consumption being US$136,822.08. Moreover, 36.6% of the sample was in poverty with 17.5% being below the poverty line (i.e. poorest poor) compared to 44% who were in the two wealthiest quintiles, of which 23% were in the wealthiest quintile (Table 10.1). On examination of area of residence by age group, it was found that 21% of rural women were 60+ years compared to 15.2% of peri-urban women and15.1% of urban women (Table 10.1) P < 0.001. Of the population, 17.2% reported poor health status (suffering from an illness, ailment, or injuries) in a 4-week period of the survey, with 82.8% indicated good health status. Of the 17.2% of women who reported poor health status, 6.5% visited public-private health care facilities for treatment. Of this 6.5%, 6.3% visited public health care institutions compared to 0.2% who visited private health care facilities, 66.1% of those who had reported an illness in the 4-week survey period bought the prescribed medication, with 40.9% of them took the medication in full. Some 5.6% of the sample reported that they resided alone (living arrangement), and 57.8% indicated no social support. Based on Table 10.1, there was a significant statistical correlation between good health status and area of residences P < 0.001. Rural women recorded the lowest health status among all women of the three geographic areas (Table 10.1): Rural women recorded the least good health status (75.5%) compared to 77.0% of urban women and 81.8% of semi-urban women. More crowding was in the rural sample (1.9 1.3 persons per room) compared to 1.81.3 persons per room in peri-urban and urban areas P = 0.020. A statistical difference was found between area of residences and mean number of visits made to health care facilities P = 0.023:
255

1.6 1.1 days for rural women; 1.7 1.3 days for peri-urban women and 2.0 2.7 days for urban women. A statistical correlation was found between social standing and area of residences P < 0.001: 22.3% of rural women were in the poorest 20% compared to 11.5% of peri-urban women and 9.5% of urban women. Rural women had the most of primary or below level education respondents (23.5%) compared to 18.4% of peri-urban and 14.6% urban P < 0.001. Concomitantly, mean income of rural women was US$ 2,871.86 US$2,646.39 which was 76.1% of the income of peri-urban women and 64.5% of that of urban women P < 0.001.

The general positive affective psychological condition of Jamaican women was moderate (3.5 out of 6 2.4) and negative affective psychological condition of the same sample was low (4.6 out of 15 3.4). On disaggregating both affective conditions by area of residences revealed a significant statistical difference: positive F statistic =36.205; P < 0.001 and negative F statistic = 30.774, P < 0.001. Based on Table 10.1, rural women had the highest negative affective psychological conditions 4.8 out of 15 3.2 compared to peri-urban (4.2 out of 15 3.5) and urban women (4.3 out of 15 3.8). However, there was no statistical difference between the negative affective psychological conditions of peri-urban and urban women (P = 0.655). Rural women had a lower mean score in positive psychological conditions (3.3 out of 6 2.4) than periurban women (3.9 out of 6 2.3) P < 0.001; however there was no significant statistical difference between rural and urban womens positive affective psychological conditions ( 3.4 out of 6 2.4) P = 0.990.

256

Table 10.1. Demographic characteristics of sample.


Variable Marital status Married Never married Divorced Separated Widowed Social Standing Poorest 20% Poor Middle Wealthy Wealthiest 20% Good Health status No Yes Educational level Primary and below Secondary University Social support No Yes Living arrangement With family or relative Without family (alone) Age group Young (15 30 years) Older adults (31 59 years) Elderly (60+ years) Age Mean (SD) Crowding Mean (SD) Mean Income per person (SD) Mean consumption per person (SD) Mean number of visits for health care (SD) Negative affective Mean (SD) Positive affective Mean (SD) Rural areas n (%) n = 4,962 1232 (25.7) 3032 (63.3) 25 (0.5) 51 (1.1) 453 (9.5) 1106 (22.3) 1162 (23.4) 1014 (20.4) 973 (19.6) 707 (14.2) 1184 (24.5) 3641 (75.5) 1010 (23.5) 3099 (72.0) 194 (4.5) 2724 (54.9) 2238 (45.1) 4714 (95.0) 248 (5.0) 1865 (37.6) 2055 (41.4) 1042 (21.0) 41.02 yrs (20.06) 1.9 (1.3 persons) US 2871.86 (US $2646.39) US $614.04 (US $871.47) 1.6days (1.1) 4.8 3.2 3.3 2.4 Peri-urban n (%) n = 2,283 568 (25.7) 1451 (65.7) 16 (0.7) 27 (1.2) 147 (6.7) 263 (11.5) 320 (14.0) 433 (19.0) 522 (22.9) 745 (32.6) 405 (18.2) 1820 (81.8) 355 (18.4) 1360 (70.4) 216 (11.2) 1418 (62.1) 865 (37.9) 2148 (94.1) 135 (5.9) 910 (39.9) 1025 (44.9) 348 (15.2) 38.65 yrs (18.19) 1.8 (1.3 persons) US$3773.41 (US $2752.03) US$888.24 (US $727.32) 1.7days (1.3) 4.2 3.5 3.9 2.3 Urban n (%) n = 1,296 243 (19.3) 907 (71.9) 18 (1.4) 22 (1.7) 71 (5.6) <0.001 123 (9.5) 149 (11.5) 222 (17.1) 321 (24.8) 481 (37.1) <0.001 292 (23.0) 979 (77.0) 159 (14.6) 807 (74.0) 125 (11.5) 741 (57.2) 555 (42.8) 0.005 1202 (92.7) 94 (7.3) < 0.001 501 (38.7) 599 (46.2) 196 (15.1) 39.12 yrs < 0.001 (17.91) 1.8 (1.2 0.020 persons) US$4451.23 < 0.001 (US 5181.68) US$1108.34 (US 1217.18) 2.0 (2.7) 4.3 3.8 3.4 2.4 < 0.001 0.023 < 0.001 < 0.001 < 0.001 < 0.001 P < 0.001

* Rate in 2002 was US$ 1= Ja.$50.97 The recorded p-value is for each variable by area of residence (ie rural, peri-urban and urban)
257

Upon examination of consumption and per capita income quintile (social standing), a significant statistical difference was found between consumption of women in different social standing F = US$22.32, P< 0.001 (Table 10.2). Those in the poorest quintile had a mean consumption per person per household of Jamaican US$225.38 (rate in 2002:1US$ = Ja$50.97) which was 67% less than those in quintile 2; 133% less than those in quintile 3; 237% less than quintile 4 and 659% less than those in the wealthiest quintile (quintile 5). Those in the wealthiest quintile had an average consumption per person per household of 125% more than those respondents in the wealthy quintile (quintile 4). Owing to the wide disparity in values, the best measure for average consumption per person per household is the median consumption US$554.39 (rate in 2002: 1US$ = JA$50.97).

258

Table 10.2. Average consumption per person per household by per capita income quintile. N Mean US$ Poorest 20% Poor Middle class Wealthy Wealthiest 20% 1492 1631 1669 1816 1933 225.38 376.37 525.07 759.91 1709.65 Std. Deviation Std. Error US$ 64.03 45.66 70.19 123.34 1550.86 917.81 US$ 1.66 1.13 1.72 2.89 35.27 9.93 95% CI Lower US$ 222.13 374.15 521.7 754.24 1640.47 742.88 Upper US$ 228.63 378.59 528.44 765.59 1778.83 781.82

Total 8541 762.35 F statistic = US$22.32, P < 0.001 Rate in 2002 was US$1 = Ja$50.97

259

There was no statistical difference between visits to either public or private health care facilities and area of residence of the sample (Table 10.3) P > 0.05. However, a statistical difference existed between health insurance coverage and area of residents in Jamaica (2 = 24.4, P< 0.001), with there being a weak correlation (contingency coefficient = 0.167). Of those who responded (n = 8,268), 12.5% of had private health insurance coverage. The least number of rural women had health insurance coverage (7%) compared to 16.5% of peri-urban women and 18.7% of urban women.

Table 10.3. Health insurance, self-reported good health status by area of residence (in %). Area of Residence Details Rural n = 4796 Health insurance Yes No Self-reported visits to public facilities for health care Yes No Self-reported visits to private facilities for health care Yes No 7.0 93.0 Peri-urban n = 2216 16.5 83.5 Urban n = 1263 18.7 81.3

7.1 92.9

4.4 95.6

6.1 93.9

0.3 99.7

0.0 100.0

0.0 100.0

Health insurance - P< 0.001 Self-reported visits to public health care facilities P < 0.386 Self-reported visits to private health care facilities P < 0.617

260

Table 10.4 revealed that there was a negative statistical correlation between self-reported good health status and age group (2 = 820.397, P< 0.001), with the association being a moderate one (cc = 0.301). The findings indicated that 7.5% of young respondents reported a poor health status compared to 15.6% other adults and 40.9% of elderly respondents indicating the substantial erosion of good health status of women as they age.

Table 10.4. Self-reported good health status by age group. Age group Young (n = 3,114) Good status No Yes health 234 (7.5) 2,880 (92.5) 558 (15.6) 3,021 (84.4) 631 (40.9) 913 (59.1) Other Adults (n = 3,579) Elderly (n = 1,544)

2 = 413.247, P< 0.001 Of the 1,417 respondents who reported an illness, 7.0% indicated that it was diagnosed as chronic recurring illness. A statistical correlation was found between illness being recurring and age group of respondents (2 = 413.247, P< 0.001), with relationship between the two aforementioned variables being a moderate one (cc = 0.473). Based on Table 10.5, Diabetes mellitus, hypertension and arthritis were found to be more an elderly chronic illness than the other age sub-samples. Simply put, as women age, chronic illness such as diabetes mellitus,

261

hypertension and arthritis increased, with arthritis having the greatest increase in elderly compared to middle age. Table 10.5. Diagnosed with recurring illness by age group. Diagnosed/recurring illness Cold Diarrhoea Asthma Diabetes Hypertension Arthritis Other No Age Group Young Age (n = 236) 64 (27.1) 9 (3.8) 37 (15.7) 5 (2.1) 17 (7.2) 3 (1.3) 65 (27.5) 36 (15.3) Total Middle Age (n = 562) 62 (11.1) 10 (1.8) 33 (5.9) 82 (14.6) 206 (36.7) 20 (3.6) 99 (17.6) 49 (8.7) Elderly (n = 636) 21 (3.3) 7 (1.1) 12 (1.9) 150 (23.6) 275 (43.2) 91 (14.3) 65 (10.2) 15 (2.4) (n = 1,434) 149 (10.4) 26 (1.8) 82 (5.7) 237 (16.5) 498 (34.7) 113 (7.9) 229 (16.0) 100 (7.0)

2 = 413.247, P < 0.001, cc = 0.473

When a correlation was performed between the duration of illness (How long did the last episode of illness last?) and area of residence, a relationship was found between the two variables (F = 7.513, P < 0.001). On an average, the mean duration for the illness was 11.09 days (SD 10.742 days), 95% CI: 10.51-11.67 days. Rural residents reported suffering from illness for a mean of 11.74 days (SD 10.691 days), 95% CI: 11.04-12.44 days which was not statistically different from mean number of days reported by peri-urban residents, 10.50 days (SD 10.573
262

days), 95% CI: 9.21-11.79 days, P < 0.001. The mean number of reported days in which rural and urban resident were ill [8.20 days (SD 10.879 days), 95% CI: 6.44-9.96 days, P = 0.233]. There

was no statistical difference between the mean duration of illness for peri-urban and urban residents, P = 0.091. The findings revealed that health status by area of residence had no statistical correlation, P= 0.051 (Table 10.6). Despite the no statistical difference, in excess of 30.0% in each area of residence suffered from hypertension, 16% from diabetes mellitus, 13% other and 5% arthritis. Of the 1,434 respondents who indicated poor health status, 93.0% said that these were diagnosed as recurring and acute. Table 10.6. Diagnosed/recurring illness by area of residence. Area of Residence Total Periurban Urban Rural Areas (n = (n = (n = 961) 292) 181) (n = 1,434) Diagnosed illness Cold Diarrhoea Asthma Diabetes Hypertension Arthritis Other No 98 (10.2) 12(1.2) 57(5.9) 159(16.5) 342(35.6) 82(8.5) 157(16.3) 54(5.6) 27(9.2) 10(3.4) 13(4.5) 49(16.8) 95(32.5) 17(5.8) 47(16.1) 34(11.6) 23(12.7) 4(2.2) 12(6.6) 29(16.0) 61(33.7) 15(8.3) 25(13.8) 12(6.6) 148(10.3) 26(1.8) 82(5.7) 237(16.5) 498(34.7) 114(7.9) 229(16.0) 100(7.0)

No significant association (P > 0.05)


263

Multivariate Analyses Of the 20 predisposed variables that were used in the used in Eq. [2], one was excluded because the correlation coefficient between it (consumption) and income was 0.68. Nine variables were found to be determinants of good health status of women in Jamaica (Table 10.7). The model had a statistical significant predictive power (2 = 1,249.19 P < 0.001; Hosmer and Lemeshow goodness of fit 2 = 5.606, P = 0.691). In addition, it was revealed that overall 84.9% (n = 7,251) of the data were correctly classified: 97.7% of those who indicated good health status and 37.8% of those who indicated poor health status. On examination of data (Table 10.7), it was revealed that private health insurance was the most significant factor predicting good health status of women in Jamaica (OR = 27.5, 95% CI = 21.135.8)followed by assets owned (OR = 1.1; 95% CI = 1.0-1.0); age of the respondents (OR = 0.9, 95% CI = 0.9-0.9); positive affective psychological status (OR = 1.1, 95%CI = 1.0-1.1); number of men in the household (OR = 0.9, 95% CI = 0.8-1.0); income (OR = 1.000, 95% CI = 1.0001.000); marital status married (OR = 1.2, 95% CI = = 1.1-1.6); crowding (OR = 0.9, 95%CI = 0.8-1.0); area of residence and negative affective psychological status (OR = 1.0, 95% CI = 0.91.0). All the factors explain 36.0% of the variability in health status of women in Jamaica. Income positively influences good health status of women (OR = 1.0, 95%CI 1.0-1.0) (Table 10.7). The current work has found that women who have health insurance were 27.5 times likely to report good health than those who do not have health insurance coverage. Rural women were less
264

likely to report good health status compared to urban women; and that peri-urban women were 1.1 times more likely to report good health status compared to urban women. Married women were 1.2 times more likely to report good health status with reference to those who were never married. Women who were experiencing greater positive affective psychological conditions were 1.1 times more likely to report good health status; and women who experienced greater negative affective psychological conditions were 0.03 times less likely to report good health status. The older women get, they are 0.19 times less likely to report good health status (Table 10.7).

265

Table 10.7. Logistic regression of the general good health status of Jamaican women by some explanatory variables, n = 8,541
Explanatory variables Two wealthiest quintiles Household Head Log medical Expenditure Separated, Divorced , Widowed Married Never married Rural Peri-Urban Urban Secondary level Tertiary level Primary and below level Social support Crowding Psychological conditions Positive Affective Negative Affective No. of males in household No. of females in household No. of children in household Age Asset owned Health Insurance Living Arrangement Physical Environment Average Income Crime Index Constant Nagelkerke R-square = 36.0% -2 Log likelihood= 4656.637 Hosmer and Lemeshow chi-square=5.606; P=0.691 Model: Omnibus Test - chi-square=1,249.19, P < 0.001 Overall correct classification = 84.9% Correct classification of cases of poor health status = 37.8% Correct classification of cases of good health status = 97.1% Reference group *P< 0.05. **P < 0.01, ***P < 0.001 Std. Error 0.102 0.288 0.027 0.131 0.095 0.096 0.115 0.104 0.183 0.077 0.045 Odds Ratio 1.136 0.891 1.003 1.231 1.247 1.000 0.819 1.054 1.000 1.084 1.168 1.123 0.907 95% CI 0.931 - 1.387 0.507 - 1.566 0.951 - 1.059 0.952 - 1.591 1.036 - 1.501* 0.679 - 0.989* 0.842 - 1.320* 0.883 - 1.330 0.817 - 1.671 0.965 - 1.306 0.831 - 0.991*

0.012 0.017 0.044 0.042 0.033 0.018 0.003 0.134 0.179 0.112 0.000 0.004 0.417

1.055 0.966 0.887 0.965 0.989 0.910 1.035 27.478 0.879 0.945 1.000 1.008 0.063

1.030 - 1.080*** 0.935 - 0.998* 0.814 - 0.966** 0.889 - 1.048 0.928 - 1.055 0.878 - 0.943*** 1.029 - 1.041*** 21.111 35.765*** 0.619 - 1.248 0.759 - 1.177 1.000 - 1.000** 0.999 - 1.017 -

266

Using a sub-sample of 4,962 rural residents, 20 initial predisposed explanatory variables were tested to ascertain factors and degree of significance of each factor (P < 0.05), one was omitted (consumption, because the correlation coefficient between it and income was 0.68). Of the 19 predisposed variables that were examined in the initial model, nine of them explained 38.6% of the variability in health status of rural women in Jamaica (Table 10.8). The model had a statistical significant predictive power (2 = 884.476 P < 0.001; Hosmer and Lemeshow goodness of fit 2 = 8.498, P = 0.386). Overall, 84% (n = 2,940) of the data were correctly classified: 96.8% (n = 2,593) of those who had indicated good health status and 42.3% (n = 347) of those with poor health status. Continuing, Table 10.8 revealed that health insurance was the most influential factor determining the good health status of rural women in Jamaica (OR = 25.0, 95% CI =18.0-34.9) followed by assets owned (OR = 1.0, 95% CI = 1.0-1.1); age (OR = 0.9, 95% CI = 0.8-0.9); number of men in household (OR = 0.8, 95% CI = 0.7-0.9); positive affective psychological status (OR = 1.1, 95% CI = 1.0-1.1); educational attainment secondary and post-secondary level education (OR = 1.4, 95% CI = 1.1-1.8); Social support (OR = 1.3, 95% CI = 1.1-1.6); marital status married (OR = 1.4, 95% CI = 0.8-2.3), and lastly income (OR = 1.0, 95% CI = 1.0-1.0). The current findings revealed that income plays the least role in determining good health status of rural women; women with health insurance are 25.0 times more likely to have good health status than those without health insurance coverage; married rural women are 1.4 times more likely to report good health status with reference with those who were never married; those rural women with social support were 1.3 times more likely to report good health status compared to those who did not have social support, and as rural women become older, they are 0.102 times
267

less likely to report good health status. More males in the household will reduced the good health status of rural women (OR = 0.83, 95% CI = 0.75-0.93): indicating that more males in a household will decrease rural womens good health status by 0.17 times compared to less males in the household. Table 10.8. Logistic regression of the good health status of rural-Jamaican women by some explanatory variables, n = 3,498
Explanatory variables Two wealthiest quintiles Household Head Log medical Expenditure Separated, divorced Married Never married Secondary or post-secondary Tertiary level Primary and below Social support Crowding Psychological conditions Positive Affective Negative Affective Number of males in house Number of females in house Number of children in house Age Assets owned Health Insurance Living Arrangement Physical Environment Average Income Crime Index Constant Std. Error 0.142 0.385 0.035 0.169 0.121 0.132 0.271 0.100 0.058 Odds Ratio 1.043 0.512 1.040 1.309 1.360 1.000 1.413 1.352 1.000 1.292 0.961 95% CI 0.790 - 1.378 0.241 - 1.087 0.971 - 1.114 0.940 - 1.822 1.074 - 1.724* 1.090 - 1.832** 0.795 - 2.298 1.063 - 1.571* 0.858 - 1.075

0.017 0.021 0.058 0.056 0.043 0.025 0.004 0.167 0.248 0.127 0.000 0.007

1.053 0.964 0.834 0.899 0.950 0.898 1.038 24.955 0.770 0.922 1.000 1.005

1.019 - 1.087** 0.924 - 1.005 0.745 - 0.933** 0.805 - 1.003 0.873 - 1.033 0.855 0.942*** 1.031 1.046*** 18.006 34.586*** 0.474 - 1.252 0.718 - 1.183 1.000 - 1.000** 0.991 - 1.018

0.541 0.069 Nagelkerke R-squared = 38.6% 2 Log likelihood=2,774.82 Hosmer and Lemeshow chi-square = 8.498; P=0.386; Model: Omnibus Test - chi-square=884.476, P < 0.001 Overall correct classification = 84% ; Correct classification of cases of poor health status =42.3% Correct classification of cases of good health status = 96.8% ; Reference group; *P < 0.05. **P < 0.01, ***P < 0.001

268

With regard to peri-urban areas in Jamaica, a sub-sample of 2,283 respondents were used to establish the good health status model. This model had a statistical significant predictive power (2 = 285.807 P < 0.001; Hosmer and Lemeshow goodness of fit 2 = 7.226, P= 0.512). Upon reviewing the classification table, overall, 88.2% of the data were correctly classified: 98.8% of those classified as having had good health status and 35.5% of those who had indicated poor health status (Table 10.9). Of the 19 predisposed variables that were tested in the initial model, six factors accounted for 36.6% of the variability in good health status of women in peri-urban area in Jamaica (Table 10.9). The factors that predict good health status of peri-urban Jamaican women in descending order were health insurance (OR=57.7; 95%CI: 29.8-111.7); asset ownership (OR=1.0; 95%CI: 1.0-1.0); age of respondents (OR=0.9; 95%CI: 0.9-1.0); number of men in household (OR=0.8; 95%CI: 0.6-1.0); negative affective psychological status (OR=0.9; 95%CI: 0.9-1.0); positive affective psychological status (OR=1.1; 95%CI: 1.0-1.1) and consumption (OR=1.0; 95%CI: 1.01.0). The findings revealed that income contributed the least to good health status of peri-urban residents. Another interesting finding of the current paper is peri-urban women who had health insurance coverage is 57.7 times more likely to report good health status compared another who do not have this coverage. The older peri-urban women get, they are 0.1 times less likely to record good health; more men contributes 0.2 times less to their good health; the more asset they own this increased their good health by 1.0 times more another with less assets and that the more they are positive, this direct increase their good health status and the converse is the case for those with greater scores in negative affective psychological conditions.
269

Table 10.9. Logistic regression of the good health status of peri-urban-Jamaican women by some explanatory variables, n=2,283
Explanatory variables Two wealthiest quintiles Household Head Log medical Expenditure Average Income Separated, divorced, widow Married Never married Secondary or post-secondary Tertiary level Primary and below Social support Crowding Psychological conditions Positive Affective Negative Affective Number of males in house Number of females in house Number of children in house Age Assets owned Health Insurance Living Arrangement Physical Environment Crime Index Constant Std. Error 0.220 0.722 0.058 0.000 0.291 0.203 0.235 0.367 0.169 0.095 Odds Ratio 0.949 2.554 0.953 1.000 0.853 1.143 1.000 0.704 0.622 1.000 0.849 0.874 95% CI 0.616 - 1.461 0.620 - 10.523 0.851 - 1.068 1.000 - 1.000 0.482 - 1.510 0.767 - 1.704 0.444 - 1.115 0.303 - 1.277 0.609 - 1.182 0.726 - 1.051

0.027 0.037 0.105 0.103 0.074 0.038 0.006 0.337 0.363 0.286 0.009 0.999

1.062 0.923 0.780 0.961 0.958 0.935 1.031 57.659 0.919 0.961 1.005 0.065

1.008 - 1.120* 0.859 - 0.992* 0.634 - 0.959* 0.786 - 1.175 0.829 - 1.107 0.867 - 1.008*** 1.018 - 1.044*** 29.785 - 111.619*** 0.451 - 1.870 0.549 - 1.683 0.988 - 1.023 -

Nagelkerke R-square=36.6% -2 Log likelihood = 1,071.43 Hosmer and Lemeshow chi-square=7.226; P=0.512 Model: Omnibus Test - chi-square=285.807, P < 0.001 Overall correct classification = 88.2% Correct classification of cases of poor health status =35.5% Correct classification of cases of good health status = 98.8% Reference group *P < 0.05. **P < 0.01, ***P < 0.001

270

A sub-sample of 1,296 women of urban Jamaica was used to build the good health status model. The model had a statistical significant predictive power (model chi-square = 263.08 P < 0.001; Hosmer and Lemeshow goodness of fit 2 = 8.481, P = 0.388). Upon observation of classification, overall, 83.4% of the data were correctly classified: 97.1% of those who had indicated good health status and 31.8% of those who reported poor health status. Of the 19 predisposed variables that were examine in the initial model, six of them accounted for 30.7% of the variability in good health status of urban women in Jamaica (Table 10.10). Health insurance had the most impact on good health status of urban women (OR = 22.2; 95%CI: 11.3-43.7) followed by in descending order are age of respondents (OR = 0.94; 95% CI= 0.9-1.0); two wealthiest quintiles (OR = 1.8; 95%CI: 1.1-2.9); asset ownership (OR=1.0; 95%CI: 1.0-1.0); positive affective psychological status (OR = 1.1; 95%CI: 1.0-1.1) and number of men in the household (OR = 1.2; 95%CI: 1.0-1.5). Embedded in the current findings are that urban women with health insurance coverage were 22.2 times more likely to record good health status compared to those who do not have health insurance coverage; the older urban women get, they are 0.1 times less likely to record good health status and that more men in urban household contributed 1.2 times more likely to good health status. Concomitantly, urban women in the two wealthiest quintiles were 1.8 times more likely to report good health status with reference to women in the poor-to-poorest 20%.

271

Table 10.10. Logistic regression of the good health status of urban-Jamaican women by some explanatory variables, n = 1,296
Std. Error 0.247 0.602 0.074 0.000 0.327 0.251 0.279 0.400 0.200 0.116 Odds Ratio 1.808 1.044 0.960 1.000 1.541 1.154 1.000 0.714 1.147 1.000 0.911 0.829

Explanatory variables Two wealthiest quintiles Household Head Log Medical Expenditure Average Consumption Separated, divorced, widowed Married Never married Secondary or post-secondary Tertiary level Primary and below Social Support Crowding Psychological conditions Positive Affective Negative Affective Number of males in house Number of females in house Number of children in house Age Assets owned Health Insurance Living arrangement Environment Crime Index Constant

95 % CI 1.113 - 2.935* 0.321 - 3.399 0.830 - 1.110 1.000 - 1.000 0.812 - 2.923 0.706 - 1.886 0.413 - 1.234 0.524 - 2.509 0.616 - 1.346 0.661 - 1.040

0.027 0.043 0.106 0.096 0.092 0.044 0.007 0.345 0.390 0.482 0.010 1.029

1.066 1.019 1.237 1.163 1.127 0.936 1.028 22.222 1.475 1.722 1.008 0.043

1.010 - 1.124* 0.937 - 1.108 1.005 - 1.522* 0.964 - 1.405 0.941 - 1.349 0.858 - 1.021*** 1.013 - 1.043*** 11.312 - 43.655*** 0.687 - 3.167 0.669 - 4.431 0.988 - 1.027 -

Nagelkerke R-square=41.5% -2 Log likelihood = 738.894 Hosmer and Lemeshow chi-square=8.481; P=0.388 Model: Omnibus Test - chi-square=263.08, P <0.001 Overall correct classification = 82.9% Correct classification of cases of poor health status =37.0% Correct classification of cases of good health status = 97.3% Reference group *P < 0.05. **P < 0.01, ***P < 0.001

Discussion
The findings of the current paper showed that poverty for rural women was 2.4 times more than that for urban women and 1.9 times more than that for peri-urban women. An interesting finding
272

is that on average urban women received income which was 1.6 times more than rural women and 1.2 times that of peri-urban women. Rural womens consumption expenditure was 45% less than that for urban women and 31% less than for peri-urban women. Another fundamental disparity was in education as 161 rural women for every 100 urban women had at most primary education and the ratio was 127 to 100 rural women for every peri-urban woman respectively. Those socioeconomic disparities between sub-regions in Jamaica, accounted for rural women having the lowest good health status. Overall, Jamaican women report good health status (over 80%). Those with poor health status were more likely to report having hypertension followed by diabetes mellitus, and the rates of these two chronic diseases were similar in the three geographical locations. Hypertension (43.2%) and diabetes mellitus (23.6%) was more prevalent in the elderly than in the other adult and young respondents. Interestingly, only 7.5% elderly had private health insurance coverage and the mean consumption expenditure for the poorest was 13% of that for those in the wealthiest income group, supporting the that poverty was a rural phenomenon and that this significantly retards consumption pattern of rural women in Jamaica. A critical finding of this study was that health insurance coverage accounted for the most influence on good health status of women in the 3 sub-regions; but that it had the most impact on good health for periurban women and the least for urban women. Another important finding was that income played a secondary role to factors such as health insurance, age of respondents and other psychosocial factors. Education did not explain good health status for peri-urban or urban women; and that more males contributed positively to the health status of urban women and negatively for women in the two other sub-regions. When health status of Jamaican women was deconstructed into area of residence, some major similarities were observed among them. The study revealed that the most significant factor
273

predicting health status of women in Jamaica across the three sub-regions was health insurance coverage. Embedded in this study is the fact that health insurance aids in the health care-seeking behaviour of women; but that it is more so for peri-urban women. For peri-urban women, those with health insurance coverage were approximately 60 times more likely to report good health status than those without this coverage, suggesting that lifestyle practices of these women account for their health status. This finding can be supported by the fact that women in peri-urban zones visited health care practitioners more than that of rural but less than urban women. Financial resource availability plays an important role in health care decisions. The resources regarding health care decision-making could be health insurance or monetary resources. Health insurance is important for access to health care and being uninsured significantly reduces access to health services and substantially increases health problems. Uninsured persons with poor health status are much more likely than their insured counterparts to report that they or a family member did not receive doctors care or prescription medicines [23]. Shi [24] reported that income was the most significant predictor of lack of health insurance coverage, which explains why rural women in this study had the least health insurance coverage, the lowest income and consumption and the lowest good health status. Low-income adult women tended to have lower health status and uninsured women tended to have problems accessing health care services [25], which are concurred by this study. Mead et al [26] noted that low-income women were less likely to have health insurance, while they were more likely to have health care access problems, chronic illness and lower overall health status than their richer counterparts. In Jamaica, Life of Jamaica and Blue Cross Jamaica Limited are the only total health insurance companies catering to the widest cross- section of Jamaicas population. These companies offer a
274

wide range of health insurance products to best suit the needs of clients from individuals, students, executives, associations and companies. This study revealed that an overall 11 out of every 100 of sample had health insurance coverage. In terms of geographical areas, 7 out of every 100 rural women, 17 out of every 100 peri-urban women and 19 out of every 100 urban women possessed health insurance coverage, which reinforced the aforementioned findings that income plays a critical role in health insurance coverage and health status. These results are not in agreement with findings from a study by Wong and Diaz [27], who found that almost threequarters of the urban population (73%) have coverage compared to 38% of those in rural areas; women showed a slightly higher and similar coverage (56%) than men (52%). Bennett et al. [28] postulated that rural residents were more likely to be uninsured than urban residents (17.8% versus 15.3%), and that rural respondents were more likely than urban counterparts to report having deferred health care because of cost (15.1% versus 13.1%). In studies done in rural areas, the probability of a worker being covered by an employer-sponsored insurance plan is lower than for urban workers [29, 30]; and therefore account for the health insurance disparity between the 3 sub-regions. The authors found that small firm size and low wages in rural areas are the main reasons for this difference. In this study 7.5% of women residents in all three regions reported having health insurance coverage, which is similar to 7.6% reported in a previous study [31]. Hence this justifies why rural women recorded the least number of visits to health care practitioners; because health care cost will be substantially an out of pocket expense that they would be unable to afford. Good health is a determinant of the individual and societal economic status. Unemployed women were reported to have poorer mental and physical health status than employed women [32, 33]. This causes low-income women to frequently face health care decisions. However, low275

income women often experience conflicts between their poor health status and lack of resources. Wagstaff and Doorslaer [34] reported that an individuals absolute income affected his/her mortality. These authors supported Rodgers [35] argument that the relationship between an individuals health and income is concave. This means that each additional dollar of income raises an individuals health status, but the increase gets smaller as income increases and justifies why income plays a secondary role to health insurance coverage. Another fact that this study

highlights is the increased indirect role that income plays, which is weaker than it direct role. Poverty is related to poor health, and urban poverty is a dynamic status. An individual or households position can decline or improve over short periods according to changing circumstances such as illness, unemployment, eviction or other events. The causes of urban poverty are interlinked, stemming from such factors as employment insecurity, sub-standard housing, poor health, low levels of income generation, vulnerability to market shocks, and limited education [36-39]. According to Hinrichson [40], most urban poverty does not result from a lack of jobs, but from a lack of well-paying, steady jobs. Unemployment rates are generally below 15% in most developing country cities, but wage rates are depressed in the formal sector, and many are self-employed in the informal sector. Average incomes in rural areas are often lower than in urban areas [41, 42]. In rural areas, poverty leads to health-related problems not only for single mothers but also for mothers with partners, while in urban areas this problem is usually observed in single mother-headed households. Rural Americans are more likely to be poorer [43] and less healthy than their urban counterparts, which is also the case in Jamaica. This study goes farther as it found that urban women in the two wealthy quintiles were 1.8 times more likely to report good health, and this was not the case for rural or peri-urban women. Although social class (ie wealthy class) is a predictor of good health status of urban women, once again peri-urban
276

women had the greatest good health status. This indicates that after certain sum of wealth, income adds increasing less to good health status. Income therefore will add substantially more to the good health status of poor women than it is likely to increase good health for middle and wealthy womens health status. Non-communicable diseases such as cardiovascular diseases, cancer, chronic respiratory diseases and diabetes mellitus are rapidly increasing problems for the socially disadvantaged [44]. In this study, findings of diagnosed chronic health conditions show patterns of worse health status among elderly women living in rural areas. The prevalence of hypertension and diabetes mellitus among respondents in the three regions were similar. However, reports of cancer, influenza, asthma and arthritis are low compared with hypertension and diabetes mellitus. Hypertension was higher in rural than urban and peri-urban areas. The self-report of disability and chronic status is higher for older than younger residents. Rural women tend to have higher rates of chronic status of hypertension, arthritis, spinal disorders, bursitis, hearing, and visual impairments than their urban counterparts. They also make fewer doctor visits than urban women. Furthermore, when seeking medical services, they are more likely to be ill, hospitalized than women in urban areas [45]. In this study, the duration of sickness in women residents in rural areas was longer than their counterparts in urban and peri-urban areas. In addition, health care facilities in rural areas are unfavorable compared to non-rural areas due to limited medical resources and shortage of physicians [46]. We can deduce from current paper that with rural women having less economic resources and lowered visits to health care facilities, they would be using more home remedy or non-traditional healers to treat their ill-health. Hence this would account for an aspect of premature mortality of these women.

277

In this study a higher percentages of the elderly in the rural areas reported poor health status. Bennett et al. [28] found that residents in rural area were more likely to report fair to poor health status than were residents of urban counties (19.5% versus 15.6%). Rural adults were more likely to report having diabetes mellitus than were urban adults (9.6% versus 8.4%). The authors also found that urban residents are more likely to use preventative care than their rural counterparts, but there seemed to be no differential use of doctor visits or hospitalizations [28]. According to Brenzel et al. [47] chronic diseases such as diabetes mellitus and hypertension are either undetected or medically untreated, or in the case of those who do receive treatment, the clinical management of the status is poor. In Jamaica, available hospital records show that between 1990 to 98 showed that twice as many women than men were admitted for hypertension and diabetes mellitus [48]. The predominance of women with chronic disease visiting health care facilities (82%) is in keeping with the experience of other public health areas for chronic diseases. In addition, women are more likely to report an illness; with 15% women compared to 12% men reported suffering from an illness or injury in the previous four weeks in 1991. The gender gap is widest for hypertension with twice as many women as men (12% vs. 6%) reporting having the disease [47]. In the current paper, the researchers found that diabetes mellitus for elderly women was 11.2 times more than that for young women and 1.6 times more than for middle aged adult women. Continuing, hypertension in elderly women was 6.0 times more than that in young women and 1.2 times more than in middle aged adults. Arthritis was 10.8 times more in elderly women compared to young women and 4.0 times more in elderly than in middle aged adult women. On the other hand, acute dysfunctions such as cold, diarrhoea and asthma decreases as women become older and the same was recorded for unspecified illness.

278

Womens education also affects attitudes toward health. The more highly educated are likely to better understand the importance of proper health care. Ross and Miroswky [49] reported that education significantly improved self-reported health and physical functioning. In addition, knowledge of and experiences with health care were found to affect an individuals health care behaviour more so than age. The latter was believed to be the most dominant determinant of health care behavior [50]. Majority of the women residents in this study attained secondary level education. Education is strongly associated with the level of health service utilization, the type of provider, the choice of private versus public provider, dietary and child-feeding, and sanitary practices [51]. However, studies have found that it is not just general education, but also healthspecific knowledge that is important. Barrera [52] and Caldwell [53,54] argued that educated mothers are more likely than the uneducated ones to take advantage of modern medicine and comply with recommended treatments because education changes the mothers knowledge and perception of the importance of modern medicine in the care of her children. In contrast, Rosenzweig and Schultz [55] viewed women schooling and health care services as partial substitutes for information regarding knowledge of diseases, treatment of illness and child-care practices, and hypothesized that the effect of education on child health becomes less important as access to public health care services improves. Presumably, in areas where such services are readily accessible, they are used by both educated and uneducated women, and thus the advantage conferred by schooling on health outcomes is narrowed. It is unlikely that the observed effects of maternal education on child-health outcomes simply reflects health knowledge and habits acquired in school, although they may play some role [56]. Education could thus influence a womans beliefs about disease causation and cure and the value she places on modern medicine. Mansfield et al. [57] compared the health practices of rural women with those of a large
279

metropolitan area. They found that rural women adopted more health practices overall than their urban counterparts, and younger women in both groups exhibited more awareness of health promotion. In addition, they found that there is higher utilization of doctors visits and preventative care among persons with the highest level of education and in the highest income groups. However, higher education or income seems to have no association with differential use of hospitalizations [57]. The current paper both concurs and disagrees with the aforementioned works. Education was not found to be significantly correlated with good health status of Jamaican women. However, this was not the case for rural women. An irony that lies in this study is the fact that there is a health disparity between women who have had a most primary education compared to those with secondary education, but there was none between tertiary and at most primary education for rural women. Human emotions are a mix of not only positive status but also negative factors [58]. Hence, depression, anxiety, neuroticism and pessimism are seen as measures of the negative psychological status that affect subjective wellbeing [59, 60]. Negative psychological status (loss of family members, friends etc) affect subjective wellbeing in a negative manner (guilt, fear, anger, disgust [60, 61] and that the positive factors influence self-reported wellbeing in a direct way. This was concurred in a study conducted by Fromson et al [62] and other researchers [59, 63]. In this study, negative affective psychological status was inversely affect good health status of Jamaican women, and the opposite was true for positive affective psychological conditions. On disaggregating the good health status by the 3 sub-regions, only positive affective conditions influence good health status of urban women while positive and negative affective psychological conditions determined good health status for rural and peri-urban women. Rural residents are more likely than their urban counterparts to experience negative circumstances such as
280

unemployment, lower rate of health insurance coverage, poor health status, and lowered consumption and earnings and this retards their health care seeking behaviours and further becomes challenges for their health. Hambleton et al. [20] found that an individuals psychological state influences his/her health status, which this study concurs with. People in rural areas are more likely to have characteristics that are strongly associated with depression, poor health status, chronic diseases and poverty. Probost et al. [64] found the prevalence of depression were slightly higher in residents in rural than in urban areas. Depression is subsumed in negative affective psychological condition, and so this work agrees with the literature. The current paper however found that there is no significant statistical difference between the negative psychological state of peri-urban and urban women in Jamaica as well as between positive affective psychological conditions and urban and urban women. Embedded in these findings are the higher over affective conditions of peri-urban women, and this fact accounts for peri-urban women having the greatest health status. Some limitations must be considered in interpreting these results as this study was completely based on data reported by interviewed residents, and of course, persons do not always answer factually in interview surveys. Therefore, survey participants could be subject to recall bias in their health status. Interviewers and supervisory staff were aware of this problem, and interviewer instructions included directions for probing participants on these issues. However, the strength of the study's sample design and data collection procedures compensated for these limitations. Conclusions The findings revealed that rural women had the least good health, while peri-urban women recorded the greatest self-reported good health. Concurrently, rural women were older; poorer;
281

received the lowest income per person; had the greatest percentage of primary level eduction; recorded the highest negative affective psychological conditions; were the least likely to have health insurance coverage and they recorded the lowest consumption expenditure. This study therefore provides a comprehensive understanding of health of women in Jamaica and the 3 subregions as well as the disparity in socio-demographic correlates of health based on the different geographical regions. Concomitantly, poverty continues to reduce the self-rated health status of women and while they are living 6 years longer than men, this does not mean that we neglect the reality that poverty is eroding their health status.

282

References 1. United Nations Department of Economic and Social Affairs, Population Division 5. World Urbanization Prospects: The 2007 Revision. Washington DC: UN; 2007 2. Bourne, P. Growing old in Jamaica: Population ageing and senior citizens wellbeing. Kingston: Department of Community Health and Psychiatry, Faculty of Medical Sciences, the University of the West Indies, Mona, Jamaica; 2009. 3. Bourne, P. Determinants of well-being of the Jamaican elderly. The University of the West Indies, Mona Campus; 2007. 4. Eldemire, D. A situational analysis of the Jamaica elderly, 1992. Kingston: Planning Institute of Jamaica; 1995. 5. Planning Institute of Jamaica, PIOJ, Statistical Institute of Jamaica, STATIN. Jamaica Survey of Living of Jamaica, 2007. PIOJ, STATIN; 2008. 6. Cohen M. Towards a framework for womens health. Patient Education and Counseling 1998, 33: 187-196. 7. Bourne PA. Quality of life of Jamaican women. West Indian Medical Journal 2008, 57 :( suppl 4): 49. 8. Eldemire D. Older women: A situational analysis, Jamaica 1996. New York: United Nations Division for the Advancement of Women; 1996. 9. Bailey, B. The search for gender equity and empowerment of Caribbean women: The role of education. In: Nain GT, Bailey B, editors. Gender equality in the Caribbean: Reality or Illusion. Kingston: Ian Randle Publishers, Jamaica, 2003: p. 108-144. 10. Bailey B, Rickets H. Gender vulnerabilities in Caribbean labour market. Social and Economic Studies 2003, 52: 49-81. 11. Seguino S. Why are women in the Caribbean so much more likely than men to be unemployed? Social and Economic Studies 2003, 52: 83-120. 12. Thorson. A.; Hoa, N.; Long, N. Health-seeking behaviour of individuals with a cough of more than 3 weeks. The Lancet 2004, 356: 1823-1824. 13. Planning Institute of Jamaica, [PIOJ], Statistical Institute of Jamaica, [STATIN]. Jamaica Survey of Living Conditions, 1990-2006. Kingston; PIOJ, STATIN; 1991-2007. 14. Grossman M. The demand for health - a theoretical and empirical investigation. New York: National Bureau of Economic Research; 1972. 15. Bourne PA. Health determinants: Using secondary data to model predictors of well-being of Jamaicans. West Indian Medical Journal 2008, 57: 476-481. 16. Planning Institute of Jamaica and Statistical Institute of Jamaica (PIOJ & STATIN). Jamaica Survey of Living Conditions 2002. Kingston: PIOJ, STATIN; 2003.
283

17. Homer D, Lemeshow S. Applied logistic regression, 2nd edn. John Wiley & Sons Inc., New York; 2000. 18. Cohen L, Holliday M. Statistics for Social Sciences. London, England: Harper and Row; 1982. 19. Bourne PA. Health Determinants: Using Secondary Data to Model Predictors of Wellbeing of Jamaicans. West Indian Medical Journal 2008, 57: 476-481. 20. Hambleton IR, Clarke K, Broome HL, Fraser HS, Brathwaite F, Hennis AJ. Historical and current predictors of self-reported health status among elderly persons in Barbados. Rev Pan Salud Public 2005, 17: 342-352. 21. Bourne PA. Medical sociology: Modeling well-being for elderly people in Jamaica. West Indian Medical Journal 2008, 57: 596-604. 22. Bourne P. Using the bio-psychosocial model to evaluate the wellbeing of the Jamaican elderly. West Indian Medical Journal 2007, 56(suppl 3): 39-40. 23. Families, USA. The uninsured with chronic health conditions: Access to care in the western region, 2000. Retrieved on April 7, 2002 from http://www.familiesusa.org/media/pdf/factsheet.pdf 24. Shi L. The convergence of vulnerable characteristics and health insurance in the US. Social Science and Medicine 2001, 53:519-529. 25. Wyn R, Solis B. Womens health issues across the lifespan. Womens Health Issues 2001, 11: 148-159. 26. Mead H, Witkowski K, Gault B, Hartmann H. The influence of income, education, and work status on womens well being. Womens Health Issues 2001, 11: 160-172. 27. Wong R, Diaz J. Healthcare utilization among older Mexicans: health and socioeconomic inequalities. Salud Publica de Mexico 2007, 49: S505-S514. 28. Bennett K, Olatosi B, Probost J. Health disparities: A rural-urban chart book. Rural Health Research and Policy Centre; 2008. 29. Maine Rural Health Research Center. Small and rural: The double jeopardy in expanding employer-based health insurance coverage. Research & Policy Brief; 1994. 30. Maine Rural Health Research Center. Employer-based health insurance coverage of workers: Are rural firms and workers different? Research & Policy Brief; 1998. 31. Planning Institute of Jamaica, Statistical Institute of Jamaica. Jamaica Survey of Living Conditions 2002. Kingston; 2003.

284

32. Ross C, Mirowsky J, Goldsteen K. The impact of the family on health: The decade in review. Journal of Marriage and the Family 1990, 52:1059-1078. 33. Siefert K, Heflin C, Corcoran M. Food insufficiency and the physical and mental health of low-income women. Women and Health 2001, 32: 159-177. 34. Wagstaff A, Doorslaer E. Income inequality and health: What does the literature tell us? Annual Review of Public Health 2000, 21: 543-567. 35. Rodgers G. Income and inequality as determinants of mortality: An international cross-section analysis. Population Studies 1979, 33: 343-351. 36. Becker C, Hamer A, Morrison A. Beyond urban bias in Africa, Portsmouth, NH: Heinemann, 1994. 37. Amis P. Making sense of urban poverty. Environment and Urbanization 1995, 7: 145-157. 38. Moser C. Urban social policy and poverty reduction. Environment and Development 1995, 7: 159-171. 39. Wratten E. Conceptualizing urban poverty. Environment and Urbanization 1995, 7: 11-36. 40. Hinrichsen D, Salem R, Blackburn R. Meeting the urban challenge. Population Reports. Series M, No. 16. Baltimore: The Johns Hopkins Bloomberg School of Public Health, Population Information Program, 2002. 41. Montgomery M, Stren R, Cohen B, Reed H.E. Cities transformed; Demographic change and its implications in the Developing World. The National Academy Press (North America)/Earthscan (Europe), Washington DC; 2003. 42. Montgomery M, Ezeh A. Urban health in developing countries: Insights from demographic theory and practice. In: S. Galea S, Vlahov D, eds. Handbook of Urban Health: Populations, Methods, and Practice. Springer, New York; 2005. 43. US Census Bureau. Poverty in the US, 1999. Current Population Reports, 2000. 44. WHO. The World Health Report 2000, health systems: improving performance, Geneva, WHO, 2000. 45. Bushy A. Health issues of women in rural environments: An overview. JAMA 1998, 53:5356. 46. Ricketts T. The changing nature of rural health care. Annual Review of Public Health 2000, 21: 639-657.
285

47. Brenzel L, Henry-Lee A, Le Franc E. Gender and equity in access to health care in Jamaica and Barbados. Sir Arthur Lewis Institute of Social and Economic Studies, University of the West Indies, Submitted to The Pan American Health Organization, 12th October, 2001. 48. Ministry of Health and the Jamaica Cancer Registry. Annual Report. Ministry of Health, 2000. 49. Ross C, Mirowsky J. Does medical insurance contribute to socioeconomic differentials in health? The Milbank Quarterly 2000, 78: 291-321. 50. Rosen P, Anell A, Hjortsberg C. Patient views on choice and participation in primary health care. Health Policy 2001, 55: 121-128. 51. World Health Organization. Health Finances: A basic guide. WHO, Western Pacific Region, 2006. 52. Barrera A. The role of maternal schooling and its interaction with public health programs in child health production. Journal of Development Economics 1990, 32: 69-91. 53. Caldwell JC. Education as a factor in mortality decline: an examination of Nigerian data. Population Studies 1979, 33:395-413. 54. Caldwell JC. Cultural and social factors influencing mortality in developing countries. The Annals of the American Academy of Political and Social Science 1990, 510: 44-59. 55. Rosenzweig M, Schultz P. Market opportunities, genetic endowments, and intra-family resource distribution: Child survival in rural India. The American Economic Review 1982, 72: 803-815. 56. Lindenbaum S, Chakraborty M, Elias M. The influence of maternal education on infant and child mortality in Bangladesh. In: Caldwell JC, Santow G. editors. Selected Readings in the Cultural, Social and Behavioural Determinants of Health. Canberra: Health Transition Centre, The Australian National University,1989: p.112131. 57. Mansfield, P.; Preston, D.; Crawford, C. The health behaviors of rural women: Comparisons with an urban sample. Health Values 1989, (6), 12-19. 58. Watson D, Clark L, Tellegen A. Development and validation of brief measures of positive and negative affect: The PANAS Scale. Journal of Personality and Social Psychology 1988, 54: 10631070. 59. Harris P, Lightsey O. Constructive thinking as a mediator of the relationship between extraversion, neuroticism, and subjective well-being. European Journal of Personality 2005, 19: 409-426.
286

60. Evans C, Sherer M, Nick T, Nakase-Thompson R, Yablon S. Early impaired self-awareness, depression, and subjective well-being following traumatic brain injury. Journal of Head Trauma Rehabilitation 2005, 20: 88-500. 61. Kashdan T. The assessment of subjective well-being (issues raised by the Oxford Happiness Questionnaire). Personality and Individual Differences 2004, 36: 1225-1232. 62. Fromson P. Self-discrepancies and negative affect: The moderating roles of private and public self-consciousness. Social Behavior and Personality 2006, 34: 333-350. 63. McCullough M, Bellah C, Kilpatrick S, Johnson JL. Vengefulness: Relationships with forgiveness, rumination, well-Being, and the big five. Personality and Social Psychology Bulletin 2001, 27: 601-610. 64. Probost J, Laditka S, Moore C, Harun N, Powell M, Baxley E. Rural-urban differences in depression prevalence: implications for family medicine. Public Health 2004, 95: 1695-1703.

287

11
Social determinants of self-reported health across the Life Course
The socio-psychological and economic factors produced inequalities in health and need to be considered in health development. In spite of this, extensive review of health Caribbean revealed that no study has examined health status over the life course of Jamaicans. With the value of research to public health, this study is timely and will add value to understanding the elderly, middle age and young adults in Jamaica. The aim of this study is to develop models that can be used to examine (or evaluate) social determinants of health of Jamaicans across the life course, elderly, middle age and young adults. Eleven variables emerged as statistically significant predictors of current good health Status of Jamaicans (p<0.05). The factors are retirement income (95%CI=0.49-0.96), logged medical expenditure (95% CI =0.91-0.99), marital status (Separated or widowed or divorced: 95%CI=0.31-0.46; married: 95%CI=0.50-0.67; Never married), health insurance (95%CI=0.029-0.046), area of residence (other towns:, 95%CI=1.051.46; rural area:), education (secondary: 95%CI=1.17-1.58; tertiary: 95%CI=1.47-2.82; primary or below: OR=1.00), social support (95%CI=0.75-0.96), gender (95%CI=1.281-1.706), psychological affective conditions (negative affective: 95%CI=0.939-0.98; positive affective: 95%CI:1.05-1.11), number of males in household (95%CI:1.07-1.24), number of children in household (95%CI=1.12-1.27) and previous health status. There are disparities in the social determinants of health across the life course, which emerged from the current findings. The findings are far reaching and can be used to aid policy formulation and how social determinants of health are viewed in the future.

INTRODUCTION Health is a multidimensional construct which goes beyond dysfunctions (illnesses, ailment or injuries) [1-14]. Although World Health Organization (WHO) began this broaden conceptual framework in the late 1940s [1], Engel [3] was the first to develop the biopsychosocial model that
288

can be used to examine and treat health of mentally ill patient. Engels biopsychosocial model was both in keeping with WHOs perspective of health and again a conceptual model of health. Both WHO and Engels works were considered by some scholar as too broad and as such difficult to measure [15]; although this perspective has some merit, scholars have ventured into using different proxy to evaluate the ideal conceptual definition forwarded by WHO for some time now. Psychologists have argued that the use of diseases to proxy health is unidirectional (or negative) [2], and that the inclusion of social, economic and psychological conditions in health is broader and more in keeping with the WHOs definition of health than diseases. Diener was the first psychologist to forward the use of happiness to proxy health (or wellbeing) of an individual [16, 17]. Instead of debating along the traditional cosmology health, Diener took the discussion into subjective wellbeing. He opined that happiness is a good proxy for subjective wellbeing of a person, and embedded therein is wider scope for health than diseases. Unlike classical economists who developed Gross Domestic Product per capita (GDP) to examine standard of living (or objective wellbeing) of people as well this being an indicator of health status along with other indicators such as life expectancy, Diener and others believe that people are the best judges of their state. This is no longer a debate, as some economists have used happiness as a proxy of health and wellbeing [18-20]; and they argued that it is a good measurement tool of the concept. Theoretical Framework Whether the proxy of health (or wellbeing) is happiness, self-reported health status, selfrated health conditions, life satisfaction or ill-being, it was not until in the 1970s that econometric analyses were employed to the study of health. Grossman [9] used econometric to capture factors that simultaneously determine health stock of a population. Grossmans work transformed the
289

conceptual framework outlined by WHO and Engel to a theoretical framework for the study of health. Using data for the world, Grossman established an econometric model that captures determinants of health. The model read (Model 1): Ht = (Ht-1, Go, Bt, MCt, ED) .. Model (1) where Ht current health in time period t, stock of health (Ht-1) in previous period, Bt

smoking and excessive drinking, and good personal health behaviours (including exercise Go), MCt,- use of medical care, education of each family member (ED), and all sources of household income (including current income). Grossmans model was good at the time; however, one of the drawbacks to this model was the fact that some crucible factors were omitted by the aforementioned model. Based on that limitation, using literature, Smith and Kington [10] refined, expanded and modified Grossmans work as it omitted important variables such as price of other inputs and family background or genetic endowment which are crucible to health status. They refined Grossmans work to include socioeconomic variables as well as some other factors [Model (2)]. Ht = H* (Ht-1, Pmc, Po, ED, Et, Rt, At, Go) .. Model (2) Model (2) expresses current health status Ht as a function of stock of health (Ht-1), price of medical care Pmc, the price of other inputs Po, education of each family member (ED), all sources of household income (Et), family background or genetic endowments (Go), retirement related income (Rt ), asset income (At). It is Grossmans work that accounts for economists like Veenhovens [20] and Easterlins [19] works that used econometric analysis to model factors that determine subjective wellbeing.
290

Like Veenhoven [20], Easterlin [19] and Smith and Kington [10], Hambleton et al. [6] used the same theoretical framework developed by Grossman to examine determinants of health of elderly (ages 65+ years) in Barbados. Hambleton et al.s work refined the work of Grossman and added some different factors such as geriatric depression index; past and current nutrition; crowding; number of children living outside of household; and living alone. Unlike Grossmans study, he found that current disease conditions accounted for 67.2% of the explained variation in health status of elderly Barbadians, with life style risks factors accounting for 14.2%, and social factors 18.6%. One of the additions to Grossmans work based on Hambleton et al.s study was actual proportion of each factor on health status and life style risk factors. A study published in 2004, using life satisfaction and psychological wellbeing to proxy wellbeing of 2,580 Jamaicans, Hutchinson et al. [21] employed the principles in econometric analysis to examine social and health factors of Jamaicans. Other studies conducted by Bourne on different groups and sub-groups of the Jamaican population have equally used the principles of econometric analysis to determine factors that explain health, quality of life or wellbeing [5, 8, 22, 23]. Despite the contribution of Hutchinson et als and Bournes works to the understanding of wellbeing, there is a gap in the literature on a theoretical framework explains good health status of the life course of Jamaicans. The current paper will model predictors of good health status of Jamaicans as well as good health status of young adults, middle age adults and elderly in order to provide a better understanding of the factors that influence each cohort.

291

METHODS Participants and questionnaire The current research used a nationally cross-sectional survey of 25,018 respondents from the 14 parishes in Jamaica. The survey used stratified random probability sampling technique to draw the 25,018 respondents. The non-response rate for the survey was 29.7% with 20.5% who did not respond to particular questions, 9.0% did not participated in the survey and another 0.2% was rejected due to data cleaning. The study used secondary cross-sectional data from the Jamaica Survey of Living Conditions (JSLC). The JSLC was commissioned by the Planning Institute of Jamaica (PIOJ) and the Statistical Institute of Jamaica (STATIN). These two organizations are responsible for planning, data collection and policy guideline for Jamaica. The JSLC is a self-administered questionnaire where respondents are asked to recall detailed information on particular activities. The questionnaire covers demographic variables, health, immunization of children 0 to 59 months, education, daily expenses, non-food consumption expenditure, housing conditions, inventory of durable goods, and social assistance. Interviewers are trained to collect the data from household members. The survey is conducted between April and July annually. Model The multivariate model used in this study is a modification of those of Grossman and Smith & Kington which captures the multi-dimensional concept of health, and health status. The present study further refine the two aforementioned works and in the process adds some new factors such as psychological conditions, crowding, house tenure, number of people per household and a deconstruction of the numbers by particular characteristics i.e. males, females and children (ages
292

14 years). Another fundamental difference of the current research and those of Grossman, and Smith and Kington is that it is area specific as it is focused on Jamaican residents. The proposed model that this research seeks to evaluate is displayed below [Model (3)]:
Ht = f(Ht-1,Pmc, EDi, Rt, At, Qt, HHt, Ci, Eni, MSi, HIi, HTi, SSi, LLi,Xi, CRi, Di, Oi, (NPi,PPi), Mi,Ni, FSi, Ai, Wi, i).. Model (3)

The current health status of a Jamaica, Ht, is a function of 23 explanation variables, where Ht is current health status of person i, if good or above (i.e. no reported health conditions four week leading up to the survey period), 0 if poor (i.e. reported at least one health condition); Ht-1 is
stock of

health for previous period; lnPmc is logged cost of medical care of person i; EDi is

educational level of person i, 1 if secondary, 1 if tertiary and the reference group is primary and below; Rt is retirement income of person i, 1 if receiving private and/or government pension, 0 if otherwise; HIi is health insurance coverage of person i, 1 if have a health insurance policy, 0 if otherwise; HTi is house tenure of person i, 1 if rent, 0 if squatted; Xi is gender of person i, 1 if female, 0 if male; CRi is crowding in the household of person i; (NPi,PPi) NPi is the summation of all negative affective psychological conditions and PPi is the summation of all positive affective psychological conditions; Mi is number of male in household of person i and Fi is number of female in household of person i; Ai is the age of the person i and Ni is number of children in household of person i; LLi is living arrangement where 1= living with family

members or relative, and 0=otherwise and social standing (or social class), Wi. Statistical analysis Statistical analyses were performed using Statistical Packages for the Social Sciences (SPSS) for Windows, Version 16.0 (SPSS Inc; Chicago, IL, USA). A single hypothesis was tested, which
293

was health status of rural resident is a function of demographic, social, psychological and economic variables. The enter method in logistic regression was used to test the hypothesis in order to determine those factors that influence health status of rural residents if the dependent variable is a binary one; and linear multiple regression in the event the dependent variable was a normally distributed metric variable . The final model was established based on those variables that are statistically significant (ie. p < 0.05) ie 95% confidence interval (CI), and all other variables were removed from the final model (p>0.05). Continuing, categorical variables were coded using the dummy coding scheme. The predictive power of the model was tested using Omnibus Test of Model and Hosmer and Lemeshow [24] was used to examine goodness of fit of the model. The correlation matrix was examined in order to ascertain whether autocorrelation (or multi-collinearity) existed between variables. Cohen and Holliday [25] stated that correlation can be low/weak (0 to 0.39); moderate (0.4-0.69), or strong (0.7-1.0). This was used in this study to exclude (or allow) a variable in the model. Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final construction of the model. To derive accurate tests of statistical significance, we used SUDDAN statistical software (Research Triangle Institute, Research Triangle Park, NC), and this was adjusted for the surveys complex sampling design. Finally, Wald statistics was used to determine the magnitude (or contribution) of each statistically significant variables in comparison with the others, and the odds ratio (OR) for the interpreting each significant variables.

294

Results: Modelling Current Good Health Status of Jamaicans, Elderly, Middle Age and Young adults Predictors of current Good Health Status of Jamaicans. Using logistic regression analyses, eleven variables emerged as statistically significant predictors of current good health status of Jamaicans (p<0.05, see Model 4). The factors are retirement income, logged medical expenditure, marital status, health insurance, area of residence, education, social support, gender, psychological affective conditions, number of males in household, number of children in household and previous health status (Table 11.1). Ht = f(Ht-1, Rt, Pmc, EDi, MSi, HIi, SSi,ARi, Xi, (NPi,PPi), Mi,Ni, i)... ..... Model (4) The model [ie Model (4)] had statistically significant predictive power (2 (27) =1860.639, p < 0.001; Hosmer and Lemeshow goodness of fit 2=4.703, p = 0.789) and overall correctly classified 85.7% of the sample (correct classified 98.3% of cases of good health status and correctly classified 33.9% of cases of dysfunctions). There was a moderately strong statistical correlation between age, marital status, education, retirement income, per capita income quintiles, property ownership, and so these were omitted from the initial model (ie model 3). Based on that fact, three age groups were classified (young adults ages 15 to 29 years; middle age adults ages 30 to 59 years; and elderly ages 60+ years) and the initial model was once again tested. There were some modifications of the initial model in keeping with the age group. For young adults the initial model was amended by excluding retirement income, property ownership, divorced, separated or widowed, number of children in household, and house tenure. The exclusion was based on the fact that more than 15% of cases missing in some categories and a high correlation between variables.
295

Predictors of current Good Health Status of elderly Jamaicans. From the logistic regression analyses that were used on the data, eight variables were found to be statistically significant in predicting good health Status of elderly Jamaicans (P < 0.5) (see Model 5). These factors were education, marital status, health insurance, area of residence, gender, psychological conditions, number of males in household, number of children in household and previous health status (see Table 11.2). Ht = f(Ht-1, EDi, MSi, HIi, ,ARi, Xi, (PPi), Mi,Ni, i)... ..... Model (5) The model had statistically significant predictive power (model 2 (27) =595.026, P < 0.001; Hosmer and Lemeshow goodness of fit 2=5.736, p = 0.677) and overall correctly classified 75.5% of the sample (correctly classified 94.6% of cases of good or beyond health status and correct classified 44.7% of cases of dysfunctions).

Predictors of current Good Health Status of middle age Jamaicans. Using logistic regression, six variables emerged as statistical significant predictors of current good health status of middle age Jamaican (p < 0.05) (Model 6). These factors are logged medical expenditure, physical

environment, health insurance, gender of respondents, psychological condition, number of children in household and previous health status (see Table 11.3) Ht = f(Ht-1, Pmc, Eni, HIi, Xi, (NPi),Ni, i).................. ..... Model (6) Based on Table 11.3, the model had statistically significant predictive power (model 2 (27) =547.543, p < 0.001; Hosmer and Lemeshow goodness of fit 2=4.318, p = 0.827) and overall correctly classified 87.2% of the sample (correctly classified 98.3% of cases of good or beyond health status and correct classified 28.2% of cases of dysfunctions).

296

Predictors of current Good Health Status of young adult in Jamaica. Using logistic regression, two variables emerged as statistically significant predictors of current good health status of young adults in Jamaica (p<0.05) (Model 7). These are health insurance coverage, psychological condition, social class and previous health status (Table 11.4). Ht = f(Ht-1, Wi, HIi, (NPi), i)........................... .....Model (7) From Table 11.3, the model had statistically significant predictive power (model 2 (19) =453.733, p < 0.001; Hosmer and Lemeshow goodness of fit 2=5.185, p = 0.738) and overall correctly classified 92.6% of the sample (correctly classified 99.0% of cases of good or beyond health status and correct classified 28.2% of cases of dysfunctions). Limitations to the Models Good Health Status of Jamaicans [ie Model (4)], elderly [ie Model (5)], middle age adults [ie Model (6)], and young adults [ie Model (7) are derivatives of Model (3). Good Health Status[ie Model (4) Model (7)] cannot be distinguished and tested over different time periods, person differential, and these are important components of good health.

Ht = f(Ht-1, Rt, Pmc, EDi, MSi, HIi, SSi,ARi, Xi, (NPi,PPi), Mi,Ni, i)........ Model (4) Ht = f(Ht-1, EDi, MSi, HIi, ,ARi, Xi, (PPi), Mi,Ni, i)........ Model (5) Ht = f(Ht-1, Pmc, Eni, HIi, Xi, (NPi),Ni, i)......................................... Model (6) Ht = f(Ht-1, Wi, HIi, (NPi), i)...............................................................Model (7) Ht = f(Ht-1,Pmc, EDi, Rt, At, Qt, HHt, Ci, Eni, MSi, HIi, HTi, SSi, LLi,Xi, CRi, Di, Oi, (NPi,PPi), Mi,Ni, FSi, Ai, Wi,i).. Model (3)

297

The current work is a major departure from Grossmans theoretical model as he assumed that factors affecting good health Status over the life course are the same, this study disagreed with this fundamental assumption. This study revealed that predictors of good health status are not necessarily the same across the life course, and differently from that of the general populace. Despite those critical findings, healthy time gained can increase good health status directly and indirectly but this cannot be examined by using a single cross-sectional study. Health does not remain constant over any specified period, and to assume that this is captured in age is to assume that good or bad health change over year (s). Health stock changes over short time intervals, and so must be incorporated within any health model. People are different even across the same ethnicity, nationality, next of kin and socialization. This was not accounted for in the Grossmans or the current work, as this is one of the assumptions. Neither Grossmans study nor the current research recognized the importance of differences in individuals owing to culture, socialization and genetic composition. Each individuals is different even if that persons valuation for good health Status is the same as someone else who share similar characteristics. Hence, a variable P representing the individual should be introduced to this model in a parameter (p). Secondly, the individuals good (or bad) health is different throughout the course of the year and so time is an important factor. Thus, the researcher is proposing the inclusion of a time dependent parameter in the model. Therefore, the general proposition for further studies is that the function should incorporate (p, t) a parameter depending on the individual and time. An unresolved assumption of this work which continues from Grossmans model is that people choose health stock so that desired health is equal to actual health. The current data cannot
298

test this difference in the aforementioned health status and so the researcher recommends that future study to account for this disparity so we can identify factors of actual health and difference between the two models. Discussions This study has modelled current good status of Jamaicans. Defining health into two categories (ie good not reported an acute or illness; or poor reported illness or ailment), this study has found that using logistic regression health status can be modeled for Jamaicans. The findings revealed that the probability of predicting good health status of Jamaicans was 0.789, using eleven factors; and that approximately 86% of the data was correctly classified in this study. Continuing, in Model (4) approximately 98% of those who had reported good health status were correctly classified, suggesting that using logistic regression to examine good health status of the Jamaican population with the eleven factors that emerged is both a good predictive model and a good evaluate or current good health status of the Jamaican population. This is not the first study to examine current good health status or quality of life in the Caribbean or even Jamaica [6, 2123, 26], but that none of those works have established a general and sub-models of good health over the life course. In Hambleton et als work, the scholars identified the factors (ie historical, current, life style, diseases) and how much of health they explain (R2=38.2%). However, they did not examine the goodness of fit of the model or the correctness of fit of the data. Bournes works [12,13] were similar to that of Hambleton et als study, as his study identified more factors (psychological conditions; physical environment, number of children or males or females in household and social support) and had a greater explanatory power (adjusted r square = 0.459) but again the goodness
299

of fit and correctness of fit of the data were omitted. Again this was the case in Hutchinson et al.s research. Like previous studies in the Caribbean that have examined health status [6, 21-23, 26], those conducted by the WHO and other scholars [27-32] did not explore whether social determinants of health vary across the life course. Because this was not done, we have assumed that the social determinants are the same across the life. However, a study by Bourne and Eldemire-Shearer [33] introduced into the health literature that social determinants differ across social strata for men. Such a work brought into focus that there are disparities in the social determinants of health across particular social characteristic and so researchers should not arbitrarily assume that they are the same across the life course. While Bourne and EldemireShearers work [33] was only among men across different social strata in Jamaica (poor and wealthy), the current paper shows that there are also differences in social and psychological determinants of health across the life course. The current paper has concluded that the factors identified to determine good health status for elderly, had the lowest goodness of fit (approximately 68%) while having the greatest explanatory power (R2= 35%). The findings also revealed low explanatory powers for young adults (R2=22.6%) and middle age adults (R2=23%), with latter having a greater goodness of fit for the data as this is owing to having more variables to determine good health. Such a finding highlights that we know more about the social determinants for the elderly than across other age cohorts (middle-aged and young adults). And that using survey data for a population to ascertain the social determinants of health is more about those for the elderly than across the life course of a population.
300

Another important finding is of the eleven factors that emerge to explain good health status of Jamaicans, when age cohorts were examine it was found that young adults had the least number of predictors (ie health insurance, social class and negative affective psychological conditions). This suggests that young adults social background and health insurance are important factors that determine their good health status and less of other determinants that affect the elderly and middle age adults. It should be noted that young adult is the only age cohort with which social standing is a determinant of good health. Even though the good health status model that emerged from this study is good, the low explanatory power indicates that young adults are unique and further study is needed on this group in order to better understand those factors that account for their good health. Furthermore, this work revealed that as people age, the social determinants of health of the population are more in keeping with those of the elderly than at younger ages. Hence, the social determinants identified by Grossman [9], Smith and Kington [10] and purported by Abel-Smith [11] as well as the WHO [27] and affiliated researchers [28-32] are more for the elderly population than the population across the life course. Conclusions There are disparities in the social determinants of health across the life course, which emerged from the current findings. The findings are far reaching and can be used to aid policy formulation and how we examine social determinants of health. Another issue which must be researched is whether there are disparities in social determinants of health based on the conceptualization and measurement of health status (using self-reported health, and health conditions). Disclosures The author reports no conflict of interest with this work.
301

Disclaimer
The researcher would like to note that while this study used secondary data from the Jamaica Survey of Living Conditions (JSLC), none of the errors in this paper should be ascribed to the Planning Institute of Jamaica (PIOJ) and/or the Statistical Institute of Jamaica (STATIN), but to the researcher.

Acknowledgement
The author thanks the Data Bank in Sir Arthur Lewis Institute of Social and Economic Studies, the University of the West Indies, Mona, Jamaica for making the dataset (2002 JSLC) available for use in this study, and the National Family Planning Board for commissioning the survey.

302

REFERENCES 1. World Health Organization, WHO. Preamble to the Constitution of the World Health Organization as adopted by the International Health Conference, New York, and June 19-22, 1946; signed on July 22, 1946 by the representatives of 61 States (Official Records of the World Health Organization, no. 2, p. 100) and entered into force on April 7, 1948. Constitution of the World Health Organization, 1948. In Basic Documents, 15th ed. Geneva, Switzerland: WHO, 1948. 2.Brannon L, Feist J. Health psychology. An introduction to behavior and health 6th ed. Los Angeles: Thomson Wadsworth; 2007. 3. Engel G. A unified concept of health and disease. Perspectives in Biology and Medicine. 1960; 3:459-485. 4. Engel GL. The clinical application of the biopsychosocial model. Am J of Psychiatry. 1980; 137:535-544. 5. Bourne P. Using the biopsychosocial model to evaluate the wellbeing of the Jamaican elderly. West Indian Med J. 2007b; 56: (suppl 3), 39-40. 6. Hambleton IR, Clarke K, Broome HL, Fraser HS, Brathwaite F, Hennis AJ. 2005. Historical and current predictors of self-reported health status among elderly persons in Barbados. Rev Pan Salud Public. 2005; 17:342-352. 7. Longest BB, Jr. Health Policymaking in the United States, 3rd. Chicago, IL: Health Administration Press; 2002. 8. Bourne PA. Quality of Life of Jamaican women. West Indian Med J. 2008; 57: (suppl 4), 49. 9. Grossman M. The demand for health- a theoretical and empirical investigation. New York: National Bureau of Economic Research; 1972. 10. Smith JP, Kington R. Demographic and economic correlates of health in old age. Demography. 1997; 34:159-170. 11. Abel-Smith B. An introduction to health: Policy, Planning and Financing. Harlow: Pearson Education; 1994. 12. Bourne PA. Health Determinants: Using Secondary Data to Model Predictors of Well-being of Jamaicans. West Indian Med J. 2008; 57:476-481. 13. Bourne PA. Medical Sociology: Modelling Well-being for elderly People in Jamaica. West Indian Med J. 2008; 57:596-604. 14. Bourne PA. Health measurement. Health 2010;2(5):465-476.
303

15. Bok S. Rethinking the WHO definition of health. Working Paper Series, 14; 2004. Retrieved from http://www.golbalhealth.harvard.edu/hcpds/wpweb/Bokwp14073.pdf (accessed May 26, 2007). 16. Diener E. Subjective wellbeing. Psychological Bulletin. 1984; 95:542-575. 17. Diener E. Subjective wellbeing. The science of happiness and proposal for a national index. Am Psychol Ass. 2000; 55:34-43. 18. Crisp R. Wellbeing. The Stanford Encyclopedia of Philosophy (winter edition) E N Zalta ed; 2005. Retrieved from http://plato.stanford.edu/archives/win2005/entries/wellbeing (accessed August 22, 2006). 19. Easterlin RA. Building a better theory of well-being. Prepared for presentation at the conference Paradoxes of Happiness in Economics, University of Milano-Biococca, March 21-23, 2003. Retrieved from http://www-rcf.use.edu/ easterl/papers/BetterTheory.pdf (accessed August 26, 2006). 20. Veenhoven R. Happiness in nations, subjective appreciation of in 56 nations 1946-1992. Rotterdam, Netherlands: Erasmus University; 1993. 21. Hutchinson G, Simeon DT, Bain BC, Wyatt GE, Tucker MB, LeFranc E. Social and health determinants of wellbeing and life satisfaction in Jamaica. International J of Social Psychiatry. 2004; 50:43-53. 22. Bourne PA. Modelling social determinants of self-evaluated health of poor old people in a middle-income developing nation. J Biomedical Sci and Engineering, 2010;3: 700-710. 23. Bourne PA. Social and environmental correlates of self-evaluated health of poor aged Jamaicans. HealthMed journal 2010;4(2):284-296. 24. Homer D, Lemeshow S. Applied Logistic Regression, 2nd edn. John Wiley & Sons Inc., New York, 2000. 25. Cohen L, Holliday M. Statistics for Social Sciences. London, England: Harper and Row, 1982. 26. Asnani MR, Reid ME, Ali SB, Lipps G, Williams-Green P. Quality of life in patients with sickle cell disease in Jamaica: rural-urban differences. Rural and Remote Health. 2008; 8: 1-9. 27. WHO. The Social Determinants of Health; 2008. http://www.who.int/social_determinants/en/ (accessed April 28, 2009).
304

Available

at

28. Kelly M, Morgan A, Bonnefog J, Beth J, Bergmer V. The Social Determinants of Health: developing Evidence Base for Political Action, WHO Final Report to the Commission; 2007. 29. Wilkinson RG, Marmot M. Social Determinants of Health. The Solid Facts, 2nd ed. Copenhagen: World Health Organization; 2003. 30. Solar O, Irwin A. A Conceptual Framework for Analysis and Action on the Social Determinants of Health. Discussion paper for the Commission on Social Determinants of Health DRAFT April 2007. Available from http://www.who.int/social_determinants/resources/csdh_framework_action_05_07.pdf (Accessed April 29, 2009). 31. Graham H. Social Determinants and their Unequal Distribution Clarifying Policy Understanding The MilBank Quarterly 2004;82 (1), 101-124. 32. Pettigrew M, Whitehead M, McIntyre SJ, Graham H, Egan M. Evidence for Public Health Policy on Inequalities: 1: The Reality According To Policymakers. Journal of Epidemiology and Community Health 2004;5, 811 816. 33. Bourne PA, Eldemire-Shearer D. Differences in social determinants of health between men in the poor and the wealthy social strata in a Caribbean nation. North Am J of Med Sci 2010;2(6):267-275.

305

Table 11.1: Good Health Status of Jamaicans by Some Explanatory Variables


Wald statistic Variable Middle Quintile Two Wealthiest Quintiles Poorest-to-poor Quintiles* Retirement Income Household Head Logged Medical Expenditure Average Income Average Consumption Environment Separated or Divorced or Widowed Married Never married* Health Insurance Other Towns Urban Area Rural Area* House Tenure - Rent House Tenure - Owned House Tenure- Squatted* Secondary Education Tertiary Education Primary and below* Social Support Living Arrangement Crowding Land ownership Gender Negative Affective Positive Affective Number of males in household Number of females in household Number of children in household Constant -0.17 -0.06 -0.01 -0.07 0.39 -0.04 0.07 0.14 0.06 0.17 1.89 0.07 0.13 0.04 0.07 0.07 0.01 0.01 0.04 0.04 0.03 0.65 6.33 0.20 0.08 0.90 28.67 14.96 26.26 13.36 2.36 29.16 8.31 0.012 0.659 0.772 0.342 0.000 0.000 0.000 0.000 0.124 0.000 0.004 0.85 0.95 0.99 0.93 1.48 0.96 1.08 1.15 1.06 1.19 6.59 0.75 0.73 0.91 0.81 1.28 0.94 1.05 1.07 0.98 1.12 0.96 1.22 1.07 1.08 1.71 0.98 1.11 1.24 1.14 1.27 0.31 0.71 0.08 0.17 15.81 18.09 0.000 0.000 1.36 2.03 1.17 1.45 1.58 2.82 -1.08 -0.42 0.88 0.55 1.48 0.58 0.224 0.447 0.34 0.66 0.06 0.23 1.93 1.93 -3.31 0.21 -0.01 0.12 0.08 0.13 776.64 6.64 0.00 0.000 0.010 0.952 0.04 1.24 0.99 0.03 1.05 0.78 0.05 1.46 1.27 Coefficient -0.03 -0.11 -0.38 0.17 -0.05 0.00 0.00 0.01 -0.97 -0.55 Std Error. 0.10 0.10 0.17 0.29 0.02 0.00 0.00 0.07 0.10 0.08 0.09 1.26 4.88 0.37 5.10 1.56 0.16 0.02 87.36 53.05 P 0.764 0.261 0.027 0.543 0.024 0.212 0.689 0.891 0.000 0.000 CI (95%) Odds Ratio 0.97 0.90 0.68 1.19 0.95 1.00 1.00 1.01 0.38 0.58 Lower 0.81 0.74 0.49 0.68 0.91 1.00 1.00 0.88 0.31 0.50 Upper 1.17 1.09 0.96 2.08 0.99 1.00 1.00 1.16 0.46 0.67

2 (27) =1860.639, p < 0.001; n = 8,274 -2 Log likelihood = 6331.085 Hosmer and Lemeshow goodness of fit 2=4.703, p = 0.789. Nagelkerke R2 =0.320 Overall correct classification = 85.7% (N=7,089) Correct classification of cases of good or beyond health status =98.3% (N=6,539) Correct classification of cases of dysfunctions =33.9% (N=550); *Reference group

306

Table 11.2: Good Health Status of Elderly Jamaicans by Some Explanatory Variables
Coefficient Middle Quintile Two Wealthiest Quintiles Poorest-to-poor quintiles Retirement Income Household Head Logged Medical Expenditure Average Income Environment Separated or Divorced or Widowed Married Never married* Health Insurance Other Towns Urban Rural areas* House tenure - rented House tenure - owned House tenure squatted* Secondary Education Tertiary Education Primary or below* Social support Living arrangement Crowding Landownership Gender Negative Affective Positive Affective Number of male Number of females Number of children Constant -0.08 0.26 -0.05 0.17 0.47 -0.03 0.07 0.18 0.05 0.22 -1.32 0.11 0.18 0.09 0.13 0.12 0.02 0.02 0.07 0.07 0.06 1.44 0.47 2.11 0.29 1.72 14.67 1.97 9.26 6.75 0.49 12.09 0.83 0.495 0.146 0.593 0.190 0.000 0.160 0.002 0.009 0.485 0.001 0.362 0.93 1.30 0.95 1.19 1.60 0.97 1.07 1.19 1.05 1.24 0.27 0.75 0.91 0.80 0.92 1.26 0.94 1.03 1.04 0.91 1.10 1.15 1.84 1.14 1.54 2.04 1.01 1.12 1.36 1.21 1.40 -0.46 0.81 0.11 0.35 16.06 5.45 0.000 0.020 0.63 2.26 0.51 1.14 0.79 4.47 -20.37 1.22 40192.9 1.24 0.00 0.96 1.000 0.327 0.00 3.38 0.00 0.30 -3.35 0.33 0.40 0.22 0.14 0.21 241.88 5.32 3.48 0.000 0.021 0.062 0.04 1.39 1.49 0.02 1.05 0.98 0.05 1.83 2.27 -0.10 0.12 -0.22 0.89 -0.06 0.00 -0.16 -0.49 -0.33 Std Error 0.15 0.17 0.22 0.65 0.04 0.00 0.12 0.15 0.15 Wald statistic 0.47 0.47 1.00 1.86 2.16 0.93 1.80 11.00 4.82 P 0.495 0.491 0.317 0.172 0.142 0.335 0.180 0.001 0.028 Odds Ratio 0.90 1.12 0.81 2.44 0.95 1.00 0.86 0.61 0.72 CI (95%) Lower 0.67 0.81 0.53 0.68 0.88 1.00 0.68 0.46 0.54 Upper 1.22 1.56 1.23 8.76 1.02 1.00 1.08 0.82 0.97

38.60

2 (27) =595.026, p < 0.001; n = 2,002 -2 Log likelihood = 2,104.66 Hosmer and Lemeshow goodness of fit 2=5.736, p = 0.677. Nagelkerke R2 =0.347 Overall correct classification = 75.5% (N=1.492) Correct classification of cases of good or beyond health status =94.6% (N=1,131) Correct classification of cases of dysfunctions =44.7% (N=361); *Reference group

307

Table 11.3: Good Health Status of Middle Age Jamaicans by Some Explanatory Variables
Coefficient Middle Quintile Two Wealthiest Quintiles Poorest-to-poor Quintiles* Retirement Income Household Head Logged Medical Expenditure Average Income Environment Separated or Divorced or Widowed Married Never married* Health Insurance Other Towns Urban Rural areas* House tenure - rented House tenure - owned House tenure squatted* Secondary education Tertiary education Primary or below* Social support Living Arrangement Crowding Landownership Gender Negative Affective Positive Affective Number of males in house Number of female in house Number of children in house Constant 0.03 -0.29 -0.57 0.50 -0.09 0.00 0.31 Std Error 0.15 0.15 0.36 0.45 0.04 0.00 0.12 Wald statistic 0.04 3.67 2.44 1.24 6.44 0.53 7.41 P 0.834 0.055 0.119 0.265 0.011 0.465 0.006 Odds Ratio 1.03 0.75 0.57 1.66 0.91 1.00 1.37 CI (95%) Lower 0.76 0.56 0.28 0.68 0.85 1.00 1.09 Upper 1.40 1.01 1.16 4.01 0.98 1.00 1.71

-0.20 -0.18 -3.04 0.11 -0.01 17.94 -1.33 0.19 0.34 -0.08 -0.19 -0.05 -0.13 0.51 -0.08 0.05 0.03 0.08 0.10 3.29

0.23 0.11 0.17 0.12 0.19 20029.78 1.12 0.13 0.23 0.10 0.21 0.06 0.11 0.11 0.02 0.02 0.06 0.06 0.04 1.25

0.77 2.68 320.76 0.75 0.00 0.00 1.43 2.11 2.23 0.57 0.87 0.65 1.47 21.41 24.66 4.51 0.23 2.09 5.47 6.89

0.380 0.102 0.000 0.387 0.963 0.999 0.232 0.146 0.135 0.450 0.351 0.419 0.226 0.000 0.000 0.034 0.630 0.149 0.019 0.009

0.82 0.84 0.05 1.11 0.99

0.53 0.68 0.03 0.87 0.68 0.00 0.03 0.94 0.90 0.76 0.55 0.85 0.71 1.34 0.90 1.00 0.92 0.97 1.02

1.28 1.04 0.07 1.42 1.44

0.26 1.20 1.41 0.93 0.83 0.95 0.88 1.66 0.92 1.05 1.03 1.08 1.11 26.77

2.35 1.55 2.21 1.13 1.24 1.07 1.08 2.06 0.95 1.10 1.14 1.21 1.21

2 (27) =547.543, p < 0.001; n = 3,799 -2 Log likelihood = 2,776.972 Hosmer and Lemeshow goodness of fit 2=4.318, p = 0.827. Nagelkerke R2 =0.230 Overall correct classification = 87.2% (N=3,313) Correct classification of cases of good or beyond health status =98.3% (N=3,143) Correct classification of cases of dysfunctions =28.2% (N=170); *Reference group

308

Table 11.4: Good Health Status of Young Adults Jamaicans by Some Explanatory Variables
Coefficient Middle Quintile Two Wealthiest Quintiles Poorest-to-poor quintiles* Household Head Logged Medical Expenditure Average Income Environment Health Insurance Other Towns Urban Rural area* Secondary education Tertiary education Primary and below* Social support Crowding Gender Negative Affective Positive Affective Number of males in house Number of females in house Married Never married* Constant -0.06 -0.39 -0.14 0.04 0.19 -0.04 0.07 0.13 0.06 0.08 2.75 0.41 0.47 0.13 0.06 0.15 0.02 0.03 0.07 0.06 0.22 0.67 0.02 0.70 1.22 0.65 1.60 4.22 6.81 3.67 0.87 0.13 16.62 0.886 0.405 0.269 0.420 0.206 0.040 0.009 0.055 0.351 0.717 0.000 0.94 0.68 0.87 1.05 1.20 0.96 1.07 1.13 1.06 1.09 15.57 0.43 0.27 0.68 0.94 0.90 0.93 1.02 1.00 0.94 0.70 2.09 1.69 1.12 1.16 1.60 1.00 1.13 1.29 1.20 1.68 -0.06 -0.59 -0.25 0.01 0.00 -0.03 -3.73 0.23 -0.05 Std Error 0.19 0.18 0.39 0.04 0.00 0.13 0.21 0.15 0.18 Wald statistic 0.10 11.10 0.41 0.09 3.29 0.04 321.51 2.42 0.07 P 0.747 0.001 0.520 0.760 0.070 0.840 0.000 0.120 0.788 Odds Ratio 0.94 0.55 0.78 1.01 1.00 0.97 0.02 1.26 0.95 CI (95%) Lower 0.65 0.39 0.36 0.93 1.00 0.75 0.02 0.94 0.68 Upper 1.37 0.78 1.68 1.10 1.00 1.26 0.04 1.69 1.34

2 (19) =453.733, p < 0.001; n = 4,174 -2 Log likelihood = 2,091.88 Hosmer and Lemeshow goodness of fit 2=5.185, p = 0.738. Nagelkerke R2 =0.226 Overall correct classification = 92.6% (N=3,864) Correct classification of cases of good or beyond health status =99.0% (N=3,757) Correct classification of cases of dysfunctions =28.2% (N=107); *Reference group

309

12
Modeling social determinants of self-rated health status of hypertensive in a middleincome developing nation
Paul A. Bourne & Christopher A.D. Charles

A piecemeal approach has been taken in studies on hypertension, but there is a void in the literature on (1) the socio-demographic profile of those with the disease in a Latin American and Caribbean nation, (2) healthcare seeking behaviour, (3) healthcare utilization, and (4) modelling social determinants of self-rated health status. The aim of this paper is to elucidate information on hypertension and the socio-demographic profile of those with the disease in a Latin American and Caribbean nation as well as to model self-rated health status of the hypertensive. Twentyseven in every 100 hypertensive persons had at least good self-rated health status. The current paper found that 2.5 times more females than males were affected by hypertension; and the hypertensives were more likely to: be married, be elderly, utilise private health care facilities, record moderate health status, be in the lower socioeconomic strata, and be rural dwellers. Most had sought medical care during the last 4-week period. Rural hypertensives recorded the greatest very poor health status, and two variables emerged as statistically significant factors of the self-rated health status of hypertensives in Jamaica. The findings provide policy makers with evidence that can be used to enhance policy formulation and intervention programmes.

Introduction
In 2007, statistics revealed that there were 2,682,120 Jamaicans (end of year population) [1], of whom 22.4% had hypertension [2]. A study conducted in 2007/2008 on Jamaicans between 15
310

and 74 years of age found that 25% of population had hypertension as well as obesity [3]. This denotes that between 1 in 5 and 1 in 4 Jamaicans are living with at least one chronic illness [2, 3]. In the 1950s, tuberculosis, heart diseases, nephritis, syphilis, pneumonia and influenza were the leading causes of mortality in the Caribbean, and in the 1980s, a shift occurred which saw cardiovascular disease, heart disease, malignant neoplasm, hypertension and diabetes being the leading causes of death. Another shift was observed in the 1990s when malignant neoplasm, cardiovascular disease, diabetes mellitus, ischaemic heart disease, other heart diseases and hypertension were among the 10 ten leading causes of death. In 2007, hypertension stood as the third leading cause of mortality in females and the 6th cause for males. Hypertension is not only a silent killer; it is an epidemic and needs to be examined as such in the developing world. Globally, chronic diseases account for 60% of deaths, and this is as high as 80% in lowto-middle income nations [4]. Jamaica like the rest of the developing world is experiencing an epidemic in cardiovascular diseases, as they are the leading cause of mortality [5], but despite this reality, obesity is the studied epidemic in the Americas, and not the face behind hypertension [6]. While 11 to 21% of Latinos in the Americas are obese, obesity accounts for between 20 to 33 1/3% of the populations in Chile, Jamaica, Mexico, Peru and Venezuela [3, 5]. Hypertension, on the other hand, increases exponentially in middle to late ages and accounts for more deaths in the world as well as in developing countries, than obesity. Diabetes, cardiovascular disease, cancers, and hypertension are among the main causes of death in the world except in South Asia and sub-Saharan Africa. The sedentary lifestyle of urban dwellers explains much of the chronic illness in the world, and come 2030 with 80% of the globes population residing in cities compared to over 50% in 2008, more people will be
311

expected to die from chronic diseases. Urban zones continue to attract many people and some of them, being poor, will not be able to change their lifestyles (cigarette consumption, sugar, diet, saturated fat and environmental factors) like the wealthy. While urban settings appeal to too many people, the better financial pull factors that appear to people do not mean that they will have less chronic illness. In fact, it is well established that there is a direct relationship between poverty and chronic illness [7- 9], which suggests that those in the lower socioeconomic strata in the developing world will in the future be vulnerable to more illnesses, and in particular chronic diseases, despite urban-rural migration. In 1998, Forrester et al. [10], using hypertension as an indicator of the emergence of chronic cardiovascular diseases, found that early blood pressure problems were virtually nonexistent in rural Africans, and were modest in Caribbean people. They noted, however, that in recent times hypertension in Nigeria, Jamaica and the US has seen remarkably steep gradients. In Jamaica [2, 3], as in Nigeria, hypertension is an important cardiovascular risk factor which affects between 20-25% of the population [11]. Clearly, hypertension in Jamaica as well as some nations in Africa is a silent epidemic [12], and while researchers have recognized this as the case in the latter state, those in the former are still to admit this reality. Studies on hypertension have shown differences between areas of residence [13, 14], stressors [15], diet [16], Western lifestyle [10], sex [17], measurement and treatment [18], and educational level [19, 20], income [20] and advanced aging [21-23]. Since blood pressure was measured for the first time in 1733 by Stephen Hales, many piecemeal studies have been conducted on the matter. An extensive research of the literature unearthed no study on selfreported hypertension that evaluates who hypertensives are, as well as modelling their self-rated
312

health status. In 2001, Swab et al. [24] stated that 3 in every 10 Jamaicans (ages 30+ years) had hypertension, and in 2007 1 out of every 4 Jamaicans had the disease. The face of hypertension is no longer middle-to-late ages in Jamaica, as the current paper found that 2.9% are young adults (15-30 years). Chronic diseases are the next tsunami facing developing countries. The swelling increases in those conditions, and in particular the high prevalence of hypertension which is a predisposing factor for cardiovascular diseases [25, 26]; highlight the importance of a comprehensive study of the face of the hypertensive person. This is no longer a silent epidemic, as mortality figures indicate that a red alert needs to be sounded for hypertension among the other chronic ailments in developing countries. If the Rule of Halves (half of those detected are treated or controlled) holds true [27-29], hypertension requires an immediate assessment of the sociodemographic characteristics and health status of its patients. Thus, the aim of this paper is to elucidate information on hypertension and the socio-demographic profile of those with the disease in a Latin American and Caribbean nation as well as to model self-rated health status of hypertensive.

Methods and materials


Sample The current paper used the 2007 Jamaica Survey of Living Condition (JSLC) dataset to carry out the analyses. The 2007 JSLC was conducted in May and August of that year. The current paper extracted a sub-sample of 206 respondents who indicated being diagnosed with hypertension from a larger nationally cross-sectional survey of 6,782 Jamaicans. The JSLC was conducted by the Planning Institute of Jamaica (PIOJ) and the Statistical Institute of Jamaica (STATIN).
313

The PIOJ and STATIN are non-profit organizations focusing on data collection and policy assessment, and they aid in the evaluation of governments social programmes including census taking, among other issues. Funded by the central government, the organizations deliver evidence-based information. Since 1989, the organizations have been collecting data on Jamaicans in order to evaluate social programmes instituted by the government. The data is collected by way of an administered questionnaire, and published in a document entitled the Jamaica Survey of Living Conditions (JSLC). The JSLC is a modification of the World Banks Living Standards Measurement Study (LSMS) household survey [30]. The survey was drawn using stratified random sampling. This design was a two-stage stratified random sampling design where there was a Primary Sampling Unit (PSU) and a selection of dwellings from the primary units. The PSU is an Enumeration District (ED), which constitutes a minimum of 100 residences in rural areas and 150 in urban areas. An ED is an independent geographical unit that shares a common boundary. This means that the country was grouped into strata of equal size based on dwellings (EDs). Based on the PSUs, a listing of all the dwellings was made, and this became the sampling frame from which a Master Sample of dwellings was compiled, which in turn provided the sampling frame for the labour force [30, 31]. The sample was weighted to reflect the population of the nation. Measurement Age is a continuous variable which is the number of years alive since birth (using last birthday). Age group is a non-binary measure: children (ages less than 15 years); young adults (ages 15 to 30 years); other-aged adults (ages 31 to 59 years); young elderly (ages 60 to 74 years); old elderly (ages 75 to 84 years) and oldest elderly (ages 85 years and older).
314

Self-reported illness (or self-reported dysfunction): The question was asked: Is this a diagnosed recurring illness? The answering options were: Yes, Cold; Yes, Diarrhoea; Yes, Asthma; Yes, Diabetes; Yes, Hypertension; Yes, Arthritis; Yes, Other; and No. Self-reported health status: How is your health in general? And the options were very good; good; fair; poor and very poor. For this study the construct was categorized into 3 groups (i) good; (ii) fair, and (iii) poor. A binary variable was later created from this variable (1 = good and fair, 0 = otherwise) [32-34]. Social class: This variable was measured based on income quintile: The upper classes were those in the wealthy quintiles (quintiles 4 and 5); the middle class was quintile 3, and those in lower classes quintiles 1 and 2. Income is measured using total expenditure. Analytic model Using econometric analyses (multiple logistic regressions), Bourne and McGrowder [27] modeled social determinants of health of rural Jamaicans. The chosen method allows for the testing of many possible variables which account for health status, which was measured as a binary variable. The literature has shown that health status can be dichotomized into good-tovery good health status and poor-to-moderate health status [32-34]. Clearly, based on the findings in the literature, care should be taken in where moderate health status is placed as Bourne [34] opined that moderate health status is best fitted into good-to-very good health status. Thus, for this study the dichotomization of health status was moderate-to-very good and very poor-to-poor. Furthermore, the selected variables which used in this model building were based
315

on the established evidence on social determinants of health. Some modifications were made to Bourne and McGrowders model as not all the variables which emerged in that model were applicable to the current work. In this model building, the variables were entered in block from which the significant ones emerged as factors which account for moderate-to-very good health status of hypertensive in Jamaica. Statistical analysis We used the SPSS computer statistical package, Version 16.0 (SPSS Inc; Chicago, IL, USA), and STATA. Cross tabulations were performed in order to examine demographics, health, and particular variables, and where 33.3% of the cells are less than 5 data vales, Fisher exact test was used instead of Chi-square. Multiple logistic regressions were used to analyze possible explanatory variables (health care-seeking behaviour in the last 4weeks, health insurance coverage, medical expenditure, marital status, income, area of residence, sex, household head and age) of self-rated health status. The results were presented using coefficients, Wald statistics, and Odds ratio, with a confidence interval of 95% (CI 95%). The predictive power of the model was tested using the Omnibus Test of Model, and Hosmer & Lemeshow [36] was used to examine goodness of fit of the model. In order to develop accurate tests of statistical significance, the researchers used SUDAAN statistical software (Research Triangle Institute, Research Park, NC; 1989), adjusted for the surveys complex sampling design [37]. A p-value < 0.05 was selected to indicate statistical significance. The final model was based on those variables that were statistically significant (p < 0.05). Categorical variables were coded using the dummy coding scheme.

316

Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final model construction. The final decision on whether to retain was based on the variables contribution to the predictive power of the model and its goodness of fit [38].

Results
Table 12.1 presents information on the sociodemographic characteristics of the sample, illness, health care utilisation, and health care-seeking behaviour. The sample was 206 respondents (mean age = 62.5 years 16.8 years). Twenty-eight percent of respondents had health insurance coverage (private, 8.3%). The majority of the respondents indicated fair self-rated health status (44.2%) compared to 26.7% who said at least good (very good, 4.9%) and 29.1% who mentioned at least poor (very poor, 3.9%). Most of the sample purchased the prescribed medication (70.2%), and 3.9% had been involved in an accident in the last 4weeks. The preferred health care utilisation of the sample was private health facilities (including hospitals, 55.2%). Predominantly, the face of hypertension in Jamaica was elderly (60+ years, 60.2%). The average number of visits to medical practitioners(s) in the last 4 weeks were 1.3 (SD = 0.7); and the mean length of illness of the sample was 24.8 days (SD = 85.3 days). The mean cost of private medical expenditure (USD 15.54 36.95) was 3.7 times more than that for public medical expenses (US $1.00 = Ja. $80.47, in 2007). Table 12.2 examines sociodemographic characteristics and health care utilisation by selfrated health status. A significant statistical association emerged between area of residence and self-rated health status (2 = 24.69, P = 0.002, contingency coefficient = 0.33).

317

Table 12.3 presents information on sociodemographic characteristics and health care utilisation by population income quintile of sample. No significant statistical association existed between self-reported illness and self-rated health status (2 = 2.98, P = 0.562); health care-seeking behaviour and population income quintile (2 = 5.49, P = 0.241) as well as between health care-seeking behaviour and sex (2 = 0.072, P = 0.788). Table 12.1 presents information on the health care-seeking behaviour of people in different marital statuses and sex of respondents. Married people had sought the most medical care (42.1%) in the last 4weeks, compared to never married people (36.4%) and other social partnerships. Married men were 2.2 times more likely to have visited a health care practitioner in the last 4 weeks compared to never-married men. Multivariate analyses Using logistic regression analyses, one variable emerged as a statistically significant factor of the self-rated health status of hypertensive Jamaicans (Table 12.3): area of residence (urban: OR = 4.15, 95% CI =1.44 11.97; other towns: OR = 3.47, 95% CI = 1.23 9.78). The model had statistically significant associative power (Model 2 = 32.6, P = 0.003; Hosmer and Lemeshow goodness of fit 2 = 9.6 (8), P = 0.8), and it correctly classified 75.1% of the sample (correctly classified 93.4% of those who self-rated their health as moderate-to-very good and 31.6% of those who self-rated their health as poor-to-very poor).

Discussion
Diabetes mellitus, cardiovascular diseases and neoplasm are among the leading causes of mortality in the world, and more so in developing countries. While infectious diseases, low
318

nutrient intake, and accidents continue to claim lives, chronic conditions are rising faster and will account for more deaths in the future. Despite this reality hypertension, which is an important cardiovascular risk factor, does not have a clear face, or factors which explain the self-rated health status of this group. The current paper found that 2.5 times more females than males are affected by hypertension; and the hypertensives are more likely to be married, elderly, to utilise private health care facilities, to record moderate health status and to be in the lower socioeconomic strata or rural dwellers. Most had sought medical care in the last 4 weeks, rural hypertensives recorded the greatest very poor health status, and two variables emerged as statistically significant factors of the self-rated health status of hypertensives in Jamaica. More Jamaicans have hypertension than any other type of chronic condition, yet more extensive and comprehensive studies have been conducted on diabetes, heart disease, neoplasms and arthritis. Traditionally, chronic diseases were viewed as middle-to-late life ailments, but there is a growing decrease in the age of contracting those conditions. In this paper, the findings concur with the literature that hypertension is a middle-to-later life ailment [20-23], as 97 out of every 100 hypertensive persons were ages 31+ years and 60 out of every 100, 60+ years old. What is evident is that 3 out of every 100 hypertensives are 15-30 years old, which supports the changing image of hypertension, and how we research this fact. Studies have used 30+ years old to examine chronic illness [24], which means that public health planning, relying on research, will be under-planning for a critical cohort in the population. Public health planners use information from within and outside of their geopolitical boundaries to enhance decision-making. While outside information affords a pertinent source of data in understanding a phenomenon, this may not provide the correct knowledge about a
319

localized group with different socioeconomic, biological and environmental conditions. Urbanization is well established in the literature as having a key role to play in human health conditions such as hypertension, diabetes mellitus and other chronic ailments. While urbanisation affects peoples lifestyle in relation to the food they eat, where they work, the surrounding environmental conditions and concern as to what they are exposed to, and their sedentary lifestyle, with almost 50% of Jamaicans residing in cities, 6 out of every 10 hypertensive person in this nation dwells in urban zones. Clearly, low nutritional intake and poverty account for more hypertensive people than the bad elements of urbanization. In Jamaica, statistics reveal that 71% of poverty is in rural areas [2]. Poverty means the incapacitation of financial resources, material deprivation, nutritional deficiency and environmental degradation, which are associated with low health and higher morbidity and mortality. Those realities form the core of the rationale for developing nations having more deaths owing to chronic illness than the developed world. A study by Van et al. [7] found that chronically ill people in the Netherlands were more likely to be poor, suggesting that material deprivation is directly associated with particular health conditions. This research concurs with Van et al.s work, and went further to find that poverty is associated with area of residence, area of residence is related to illness, and by extension hypertension is higher among rural respondents. Smith and Kington [39] postulated that money is able to buy health, from which it can be extrapolated that poverty is associated with low health, increased morbidity and mortality. While their argument is not entirely true, as health is not exchangeable (cannot be bought), money provides access to better nutrition, lifestyle, choice of health care services, good sanitation and
320

physical milieu, which otherwise is difficult for the poor to obtain without governmental or other interventions. In this paper 40 out of every 100 hypertensive persons were poor compared to 37 out of every 100 in the wealthy social strata, which somewhat supports Smith and Kingtons postulation. So when it is said that chronic illness is becoming the next tsunami in developing countries, the swelling increases in chronic illness, and in particular hypertension, are more evident among those in the lower socioeconomic group in those societies. The push-pull factors associated with migration in developing countries are accounted for by poverty, among other psychosocial conditions. Poverty hinders opportunity, life expectancy, quality of life, economic progress, and brings nutritional deficiencies, and material deprivation, which are the very reasons that pull rural residents to urban areas. In this research, urban dwellers were 4.1 times more likely to record moderate-to-very good self-rated health status than their rural counterparts; and those who live in semi-urban areas were 3.5 times more likely to have greater moderate-to-excellent self-rated health status. Material deprivation in rural areas in Jamaica is accounting for more morbidity and low health status, and clearly this will be a push factor for urban-rural migration, despite the negatives of urban living. In this study no significant statistical relationship existed between health care-seeking behaviour and population income quintile (social standing). This may appear paradoxical, as financial deprivation should affect peoples ability to afford health care, and rightfully so, but since 2005, the Jamaican government has instituted free health care in all public hospitals except the University Hospital of the West Indies, which means that money will influence the choice of care and not health care demand. This therefore accounts for the greater percentage of hypertensives having sought medical care in the last 4 weeks (68%) compared to the population
321

(66%) [2]. Despite the removal of access fees from public health care institutions, there is a preference for private health care utilisation. The preference for private health care utilisation among hypertensives is embedded in long queues, low privacy, social treatment of patients, and milieu the environment of public health care facilities - that push people into private health care demand. The reality still exists that public health care is the choice of 44 out of every 100 hypertensive Jamaicans, suggesting that public health will be required to plan for this group. While the onset of hypertension commences at 15 years in Jamaica, the non-children public health care system needs to cater to this cohort, as their choices, lifestyle, demands and tolerance for disrespectful behaviour are not the same as elderly or middle-aged adults. A public health concern must be the ratio of males to females with hypertension in Jamaica. Swaby et al. [24] opined that there is a preponderance of females with chronic illness and treatment for chronic illness, as compared to males, but this study found that the disparity was as much as 2.5 females to 1 male (using hypertension). There was no statistical association between the health care-seeking behaviour of male (67.2%) and female hypertensives (69.2%) in Jamaica, which refutes Swaby et al.s [24] earlier, findings. Furthermore, the preponderance of females to males with hypertension accounts for why this health condition is the third leading cause of mortality in the former, and the sixth leading cause for the latter group. Hypertension is brought on by various stressors in lifestyle practices, and with the influx of females into the labour force, top managerial positions, higher education and single parents, they are now exposing themselves to the risk factors associated with those social roles that were once dominated by males. Statistics reveal that the unemployment rate for females (14.3%) is 2.6
322

times more than that for males [40], indicating that unemployment, as well as other types of social deprivation, are associated with greater hypertension among females. A study by Atallah et al. [41] found that hypertension was greater among unemployed Caribbean people than those who were employed, which also emerged in the current research. The unemployed females are vulnerable to the dictates of males, and during this period there are the social challenges of child rearing for mothers, the psychological stressors of unemployment, the psychological situation of a dictatorial male, the material deprivation, dietary deficiency, and these influence the higher blood pressure count seen in them, compared to males. The 21st Century has brought with it urbanization, lifestyle and role changes, and risk factors related to chronic diseases for many Caribbean peoples, as well as the economic burden of chronic illnesses such as diabetes mellitus and hypertension. For some time now Caribbean governments have instituted data collection units to examine epidemiological data [42] on prevalence, gender-specific population and age-specific mortality, but for the purpose of effective public health policy planning more information is needed on the face behind hypertension. The current work opens a comprehensive discussion and analysis of the hypertensives in Jamaica, and while economic development is associated with economic growth, increased employment of females in the labour force means lower male dependency, and while money reduces material deprivation, the side effect is increased hypertension among this group. Interestingly, in this study there is a greater prevalence of hypertension among married than non-married Jamaicans, but no difference in the self-rated health status between the groups. According to Smith and Waitzman [43] many observers have theorized that married individuals have access to more informal social support than do non-married individuals, which explains
323

the social reality of a higher quality of life for married couples than non-married individuals [44]. Furthermore, studies have shown that married people have a lower mortality risk in the healthy category than the non-married [45], and this justifies why they take less life-threatening risks [46]. Clearly, the benefits of marriage as put forward by other scholars do not provide protection from hypertension among this cohort. In fact they recorded a greater prevalence of hypertension than other marital states. Married people are more likely to seek medical care than non-married people, and this accounts for the greater prevalence of hypertension among them. Although males do not like to seek medical care, those who are married seek more care on the request of their wives which accounted substantially for more of them visiting a medical practitioner in the last 4-week period, compared to those who were never married. Smith and Waitzman [43] opined that wives were found to dissuade their husbands from particular risky behaviours such as the use of alcohol and drugs, and would ensure that they maintain a strict medical regimen coupled with proper eating habits. With more married people utilising health care services, this means that more nonmarried Jamaicans would be ill but have not yet been diagnosed. If the Rule of Halves (half of those detected are treated or controlled) holds true [27-29], the greater prevalence of hypertension among married people is as a result of the greater half seeking more medical care than non-married people. This speaks to a public health problem, as the treatment and prevalence of hypertension is undoubtedly greater than the percentage currently planned for in the nation. There is a need to have more people seeking medical care, but this must be done in a holistic way, as outlined earlier from the findings of this paper. The hypertension epidemic is clearly highlighted as an important public health problem, but in order to effectively combat this
324

reality,

poverty,

opportunity,

social

exclusion,

unemployment,

malnutrition,

disease

management, early testing and lifestyle practices must be coalesced by health planners. A study as early as in the 1980s had stated that hypertension was the most prevalent chronic illness in the West Indies [47] and in 2000 Barcelo [48] called it a silent killer, but researchers have continued to examine its aetiology, management, programmes and even a study conducted in 2007/08 [3], like its predecessors, used the standard age-specific, gender and education-specific conditions. The social explanations are rarely examined, and when done the traditional variables (age, gender, and educational level) are examined by scholars, instead of the more demographic variables such as marital status, area of residence, social class and health care utilisation, as well as self-rated health status. This study is more comprehensive than other works and provides research experts with social justification for the face behind hypertension in Jamaica. It should be used to help public health practitioners, policy makers and governments to understand the complexity of effectively implementing programmes to address the management of hypertension, as well as other chronic illnesses. Poverty is the underlying challenge to greater health in the population, despite the gains of economic development, growth, removal of health care user fees, and social programmes.

Conclusion
In summary, the current evidence shows that hypertension has changed compared to the traditional late life disease to middle-to-late years, and that it mostly affect females, rural residents, married respondents and marginally inflect the poor more than those in the wealthy social strata. And that the social determinants of self-rated health status are fundamentally different from those identified in the literature on the population, or other sub-populations.
325

Conflict of interest
The authors have no conflict of interest to report.

Disclaimer
The researcher would like to note that while this study used secondary data from the Jamaica Survey of Living Conditions, none of the errors in this paper should be ascribed to the Planning Institute of Jamaica or the Statistical Institute of Jamaica, but to the researchers.

326

References
1. Statistical Institute of Jamaica (STATIN). Demographic statistics 2007. Kingston: STATIN; 2008. 2. Planning Institute of Jamaica, (PIOJ), Statistical Institute of Jamaica, (STATIN). Jamaica Survey of Living Conditions, 1989-2007. Kingston: PIOJ, STATIN; 1989-2008. 3. Wilks R, Younger N, Tulloch-Reid M, McFarlane S, Francis D: Jamaica health and lifestyle survey 2007-8. Kingston: Tropical Medicine Research Institute (TMRI), University of the West Indies, Mona; 2008. 4. World Health Organization. Preventing Chronic Diseases a vital investment. Geneva: WHO; 2005. 5. Yach D, Hawkes C, Gould CL, Hofman KJ. The global burden of chronic diseases: Overcoming impediments to prevention and control. JAMA 2004; 29:2616-2622. 6. Jacoby E. The obesity epidemic in the Americas: making healthy choices the easiest choices. Rev Panam Salud Public 2004; 15:278-84. 7. Van Agt HME, Stronks K, Mackenbach JP. Chronic illness and poverty in the Netherlands. Eur J of Public Health 2000; 10:197-200. 8. Fox J ed. Health inequalities in European Countries. Aldershot: Gower Publishing Company Limited; 1989. 9. Illsley R, Svenson PG, ed. Health inequalities in Europe. Soc Sci Med 1990; 31(special issue):223-420. 10. Forrester T, Cooper RS, Weatherall D. Emergence of Western diseases in the tropical world: the experience with chronic cardiovascular diseases. British Med Bulletin 1998; 54:463-473. 11. Sani MU. Commentary: Cardiovascular disease in Sub-Saharan Africa: An emerging problem. Ethn Dis 2007; 17:574-575. 12. Mufunda J, Nyarango P, Kosia A, Obgamariam A, Mebrahtu G, Usman A, Ghebrat J, Gebresillosie S, Goitom S, Kifle A, Tesfay A, Bebremichael A. Noncommunicable diseases in Africa: a silent hypertension epidemic Eritrea. J of Human Hypertension 2005; 19:255-256. 13. Kaufman JS et al. Determinants of hypertension in West Africa: contribution of anthropometric and dietary factors to urban-rural and socioeconomic gradients. Am J Epidemiol 1996; 143: 12031218. 14. Mufunda J, Chifamba J, Somova L & Sparks HV. Hypertension and cardiovascular disease in migrating population. Migration within Africa. High Blood Pressure 1995; 4: 4649. 15. Strogatz DS et al. Social support, stress and blood pressure in black adults. Epidemiology 1997; 8: 482497. 16. Mufunda J, Chifamba J, Chitate N & Vengesa PM. Salt sensitivity is not associated with hyperinsulinemia in a sample of rural black Zimbabweans. S Afr Med J 1998; 88: 361364. 17. Padmanabhan S, Menni C, Lee, Wai K; Laing S, Brambilla P, Sega R, Perego R, Grassi G, Cesana G, Delles C, Mancia G, Dominiczak AF. The effects of sex and method of blood pressure measurement on genetic associations with blood pressure in the PAMELA study. J of Hypertension 2010; 28:465-477.
327

18. van Onzenoort HAW, Verberk WJ, Kroon AA, Kessels AGH, Nelemans PJ, van der Kuy, P-HM, Neef C, de Leeuw PW. Effect of self-measurement of blood pressure on adherence to treatment in patients with mild-to-moderate hypertension. J of Hypertension 2010; 28:622-627. 19. Williams RB. How does lower education get inside the body to raise blood pressure? What can we do to prevent this? Hypertension 2010; 55:617-618. 20. Mendez MA, Cooper R, Wilks R, Luke A, Forrester T. Income, education, and blood pressure in adults in Jamaica, a middle-income developing country. Int J of Epidemiol 2003; 32:400-408. 21. Ragoobirsingh D, McGrowder D, Morrison E, Lewis-Fuller E, Fray J, Johnson P. The Jamaican Hypertension Prevalence Study. Journal of National Medical Association 2002; 94 (7): 561-565. 22. Eldemire D. A situational analysis of the Jamaican elderly, 1992. Kingston: Planning Institute of Jamaica; 1995. 23. Wilks R. Hypertension: Burden and risk factors. In: Morgan O, ed. Health issues in the Caribbean. Kingston: Ian Randle Publishers; 2005: pp. 98-114. 24. Swaby P, Wilson E, Swaby S, Sue-Ho R, Pierre R. Chronic diseases management in the Jamaican setting: HOPE worldwide Jamaicas experience. PNG Med J 2001; 44:171-175. 25. Wilber JA. The problem of undetected and untreated hypertension in the community. Bull NY Acad Med 1973; 49: 510-520. 26. Cooper RS, Walf-Maier K, Luke A, Adeyemo A, Banegas JR, Forrester T, Giampaoli S, Joffres M, Kastarinen M, Primatesta P, Stegmayr B, Thamm M. An international comparative study of blood pressure in populations of European vs. African descent. BMC Medicine 2005; 3:2. 27. Boume PA, McGrowder DA. Health status of patients with self-reported chronic diseases in Jamaica. North Am J Med Sci 2009; 1: 356-364. 28. Smith WCS, Lee AJ, Crombie IK, Tunstall-Pedoe H. Control of blood pressure in Scotland: the Rule of Halves. BMJ 1990; 300: 981-983. 29. Cruickshank JK, Mbanya JC, Wilks R, BAlkau B, Forrester T, Anderson SG, Mennen L, Forhan A, Riste L, McFarlane-Anderson N. Hypertension in four African-origin populations: current Rule of Halves, quality of blood pressure control and attributable risk of cardiovascular disease. J of Hypertension 2001; 19:41-46. 30. World Bank, Development Research Group, Poverty and Human Resources. Jamaica Survey of Living Conditions, 1988-2000. Basic information. Washington: The World Bank; 2002. (September 2, 2009, at http://siteresources.worldbank.org/INTLSMS/Resources/33589861181743055198/3877319-1190214215722/binfo2000.pdf). 31. Statistical Institute Of Jamaica, Jamaica Survey of Living Conditions. [Computer file]. Kingston, Jamaica: Statistical Institute Of Jamaica [producer], 2007. Kingston, Jamaica: Planning Institute of Jamaica and Derek Gordon Databank, University of the West Indies [distributors]; 2008. 32. Finnas F, Nyqvist F, Saarela J. Some methodological remarks on self-rated health. The Open Public Health Journal 2008; 1: 32-39. 33. Idler EL, Benjamin Y. Self-rated health and mortality: A Review of Twenty-seven Community Studies. Journal of Health and Social Behavior 1997; 38: 21-37.
328

34. Bourne PA. Dichotomising poor self-reported health status: Using secondary crosssectional survey data for Jamaica. North Am J Med Sci 2009; 1: 295-302. 35. Bourne PA, McGrowder DA. Rural health in Jamaica: Examining and refining the predictive factors of good health status of rural residents. Journal of Rural and Remote Health 9 (2), 2009:1116. 36. Homer D, Lemeshow S. Applied logistic regression, 2nd edn. John Wiley & Sons Inc., New York; 2000. 37. LaVange LM, Stearn SC, Lafata JE, Koch GG, Shah BV. Innovative strategies using SUDAAN for analysis of health surveys with complex samples. Stat Methods Med Res 1996; 5:311-329. 38. Polit DF. Data analysis and statistics for nursing research. Stamford: Appleton & Lange Publisher; 1996. 39. Smith JP, Kington R. Demographic and economic correlates of health in old age. Demography 1997; 34: 159-170. 40. Planning Institute of Jamaica (PIOJ). Economic and Social Survey Jamaica 2007. Kingston: PIOJ; 2008. 41. Atallah A, Inamo J, Larabi L, Chatellier G, Rozet J-E, Machuron C, de Gaudernaris R, Lang T. Reducing the burden of arterial hypertension: what can be expected from an improved access to health care? Results from a study in 2420 unemployed subjects in the Caribbean. J of Human Hypertension 2007; 21:316-322. 42. Cunninghma-Myrie C, Reid M, Forrester TE. A comparative study of the quality and availability of health information used to facilitate cost burden analysis of diabetes and hypertension in the Caribbean. West India Med J 2008; 57:383-392. 43. Smith KR, Waitzman NJ. Double jeopardy: Interaction effects of martial and poverty status on the risk of mortality. Demography 1994; 31:487-507. 44. Lillard LA, Panis CWA. Marital status and mortality: The role of health. Demography 1996; 33:313-327. 45. Goldman N. Marriage selection and mortality patterns: Inferences and fallacies. Demography 1993; 30:189-208. 46. Umberson D. Family status and health behaviors: Social control as a dimension of social integration. Journal of Health and Social Behavior 1987; 28:306-19. 47. Grell GAC. Hypertension in the West Indies. Postgraduate Med J 1983; 59:616-621. 48. Barcelo A. Diabetes and hypertension in the Americas. West Indian Med J 2000; 49:262265.

329

Table 12.1. Health seeking behaviour (in %) by marital status and sex

330

Table 12.1. Sociodemographic characteristics of study population, n = 206 Characteristic Sex Male Female Marital status Married Never married Divorced Separated Widowed Partner in household Yes No Did not respond Social assistance (PATH) Yes No Area of residence Urban Semi-urban Rural Population income quintile Poorest 20% Poor Middle Second wealthy Wealthiest 20% Age cohort Young adults Other aged adults Young-old Old-old Oldest-old Illness (self-reported) Yes No Health care seeking behaviour Yes No Health care utilization Public hospital Private hospital Public health centre Private health centre

n 58 148 91 69 3 5 37 93 12 105 41 165 47 41 118 47 35 48 38 38 6 76 61 49 14 205 1 140 64 35 7 34 78

% 28.2 71.8 44.4 33.7 1.5 2.4 18.0 45.1 5.8 51.0 19.9 80.1 22.8 19.9 57.3 22.8 17.0 23.3 18.4 18.4 2.9 36.9 29.6 23.8 6.8 99.5 0.5 68.6 31.4 22.7 4.5 22.1 50.7

331

Table 12.2. Sociodemographic characteristics and health care utilization by self-rated health status Self-reported health status Very good Good Fair Characteristic n (%) n (%) n (%) Area of residence* Urban 1 (10.0) 13 (28.9) 26 (28.6) Semi-urban 5 (50.0) 6 (13.3) 24 (26.4) Rural 4 (40.0) 26 (57.8) 41 (45.1) Population income quintile Poorest 20% 1 (10.0) 9 (20.0) 22 (24.2) Second poor 1 (10.0) 8 (17.8) 16 (17.6) Middle 2 (20.0) 11 (24.4) 17 (18.7) Second wealthy 2 (20.0) 7 (15.6) 19 (20.9) Wealthiest 20% 4 (40.0) 10 (22.2) 17 (18.7) Health care seeking behaviour No 3 (30.0) 21 (46.7) 22 (24.7) Yes 7 (70.0) 24 (53.3) 67 (75.3) Age cohort Young adults 1 (10.0) 3 (6.7) 2 (2.2) Other aged adults 6 (60.0) 20 (44.4) 34 (37.4) Young-old 2 (20.0) 14 (31.1) 26 (28.6) Old-old 0 (0.0) 6 (13.3) 22 (24.2) Oldest-old 1 (10.0) 2 (4.4) 7 (7.7) Sex Male 3 (30.0) 8 (17.8) 26 (28.6) Female 7 (70.0) 37 (82.2) 65 (71.4) Marital status Married 3 (33.3) 16 (35.6) 43 (47.3) Never married 6 (66.7) 21 (46.7) 28 (30.8) Divorced 0 (0.0) 1 (2.2) 1 (1.1) Separated 0 (0.0) 1 (2.2) 1 (1.1) Widowed 0 (0.0) 6 (13.3) 18 (19.8)
*P < 0.05

Poor n (%) 7 (13.5) 5 (9.6) 40 (76.9) 13 (25.0) 8 (15.4) 16 (30.8) 9 (17.3) 6 (11.5) 16 (30.8) 36 (69.2) 0 (0.0) 16 (30.8) 17 (32.7) 16 (30.8) 3 (5.8) 17 (32.7) 35 (67.3) 25 (48.1) 14 (26.9) 1 (1.9) 2 (3.8) 10 (19.2)

Very poor n (%) 0 (0.0) 1 (12.5) 7 (87.5) 2 (25.0) 2 (25.0) 2 (25.0) 1 (12.5) 1 (12.5) 2 (25.0) 6 (75.0) 0 (0.0) 0 (0.0) 2 (25.0) 5 (62.5) 1 (12.5) 4 (50.0) 4 (50.0) 4 (50.0) 0 (0.0) 0 (0.0) 1 (12.5) 1 (12.5)

332

Table 12.3. Sociodemographic characteristics and health care utilization by Population Income Quintile Population Income Quintile Poorest 20% Second poor Middle Second wealthy Characteristic n (%) n (%) n (%) n (%) Area of residence Urban 3 (6.4) 8 (22.8) 5 (10.4) 10 (26.3) Semi-urban 7 (14.9) 3 (8.6) 16 (22.9) 12 (31.6) Rural 37 (78.7) 24 (68.6) 32 (66.7) 16 (42.1) Health care seeking behaviour No 26 (56.5) 27 (77.1) 32 (66.7) 26 (70.3) Yes 20 (43.5) 8 (22.9) 16 (33.3) 11 (29.7) Age cohort Young adults 1 (2.2) 0 (0.0) 1 (2.1) 1 (2.7) Other aged adults 14 (29.8) 12 (34.3) 15 (31.3) 17 (44.7) Young-old 16 (34.0) 10 (28.6) 13 (27.1) 9 (23.7) Old-old 12 (25.5) 8 (22.8) 15 (31.2) 11 (28.9) Oldest-old 4 (8.5) 5 (14.3) 4 (8.3) 0 (0.0) Sex Male 12 (25.5) 8 (22.9) 10 (20.8) 14 (36.8) Female 35 (74.5) 27 (77.1) 38 (79.2) 24 (63.2) Marital status Married 19 (40.3) 17 (50.0) 20 (41.7) 19 (50.0) Never married 18 (38.3) 10 (29.4) 13 (27.1) 12 (31.6) Divorced 1 (2.1) 0 (0.0) 0 (0.0) 1 (2.6) Separated 1 (2.1) 0(0.0) 2 (4.1) 2 (5.3) Widowed 8 (17.0) 7 (20.6) 13 (27.1) 4 (10.5) Self-reported illness Yes 47 (100.0) 35 (100.0) 47 (97.9) 38 (100.0) No 0 (0.0) 0 (0.0) 1 (2.1) 0 (0.0) Health Insurance* no coverage 38 (80.9) 23 (65.7) 38 (79.2) 27 (71.0) private 1 (2.1) 3 (8.6) 0 (0.0) 6 (15.8) public 8 (17.0) 9 (25.7) 10 (20.8) 5 (13.2) *P < 0.05 333

Wealthiest 20% n (%) 21 (55.3) 8 (21.0) 9 (23.7) 29 (76.3) 9 (23.7) 3 (7.9) 18 (47.4) 13 (34.2) 3 (7.9) 1 (2.6) 14 (36.8) 24 (63.2) 16 (42.1) 16 (42.1) 1 (2.6) 0 (0.0) 5 (13.2) 38 (100.0) 0 (0.0) 22 (57.9) 7 (18.4) 9 (23.7)

Table 12.4. Logistic regression: Variables of self-rated health status Variable Health seeking behaviour Health insurance (1=Yes) Logged medical expenses Never married (reference) Married Divorced, separated or widowed Lower class (reference) Middle class Upper class Logged income Rural area (reference) Urban area Other town Sex (1= male) Household head Age
Coefficient Std. error Wald statistic Odds ratio CI (95%)

-0.57 0.04 -0.36 -0.48 -0.75 -0.09 0.03 0.06 1.42 1.24 -0.31 -0.26 -0.02

0.41 0.41 0.19 0.44 0.55 0.49 0.61 0.54 0.54 0.53 0.42 0.41 0.01

1.99 0.01 3.41 1.19 1.87 0.03 0.00 0.01 6.92 5.51 0.53 0.38 1.21

0.57 1.04 0.70 1.00 0.62 0.48 1.00 0.92 1.03 1.07

0.26 - 1.25 0.47 - 2.31 0.48 - 1.02 0.26 - 1.47 0.16 - 1.38 0.35 - 2.39 0.31 - 3.41 0.37 - 3.07

1.00 4.15** 1.44 - 11.97 3.47* 1.23 - 9.78 0.74 0.76 0.99 0.32 - 1.68 0.35 - 1.74 0.96 - 1.01

Model chi-square = 32.6, P = 0.003 Hosmer and Lemeshow goodness of fit 2 = 9.6 (8), P = 0.8 -2Log Likelihood = 201.7 Nagelkerke R2 = 0.22 Overall correct classification = 75.1% Correct classification of cases of self-rated moderate-to-very good health status = 93.4% Correct classification of cases of self-rated poor-to-very poor health status = 31.6% *P < 0.05, **P < 0.01, ***P < 0.001

334

13
Factor Differentials in contraceptive use and demographic profile among females who had their first coital activity at most 16 years versus those at 16+ years old in a developing nation
Previous studies have examined age at first sexual intercourse and factors which determine contraceptive use, but none have explored factors which determined method of contraception use between females whose first coital activity began at 16+ years and those who started < 16 years old. This research aims to bridge the gap in the literature by elucidating information on the differentials in factors of contraceptive use between females whose first coital activity was < 16 years and 16+ years old as well as sociodemographic and reproductive health characteristics of these respondents. More females whose first coitus was < 16 were currently in a sexual union (83%) compared with 79% of those who began at 16+ years old. Factor differentials on contraceptive use emerged between the two cohorts. These were social class (upper class: OR = 0.72, 9%% CI = 0.55 0.94) for those who begin < 16 years old but not for those 16+ and area of residence (Rural area: OR = 1.26, 95% CI = 1.07 1.47) for the latter but not the former. The current results are far reaching and can be used to guide new public health intervention programmes.

Introduction
For decades, the developing countries like the developed nations have been experiencing lowered age at first coital activity, which commences during the adolescence years. Young people (ie. adolescents) continue to be engaged in sexual activities outside of marriage and even
335

the statutes. The continuity of early sexual debut means that there are some health and social matters that will face the society because of early sexual relationships. It is well documented that early sexual initiation is associated with increased HIV, human papillomavirus (HPV), cervical cancers, teenage pregnancy, unwanted pregnancies, abortion (safe and unsafe), and lowered levels of education and financial opportunities [1-6]. While the developing nations have been plagued by the HIV/AIDS epidemic and lowered age at sexual debut, the developed world is more so experiencing lowered age at first sexual debut than the prevalence and incidence of HIV/AIDS epidemic faced by the developing societies. A previous study established that the lowering of the age of first coital activity has been so for the past 3 decades in developed nations, and particularly in New Zealand [7]. Furthermore, Dickson et al.s work [7]; using a longitudinal study of a cohort born in Dunedin in 1972-3, found that there were young people who were engaged in sexual activities before 13 years old. This concurs with a five community ethnographic study carried out by Chevannes in the Caribbean [8], which found that sex among adolescents starts as early as 14 years. The aforementioned early sexual debut in the Caribbean and New Zealand is also obtained in the United States [9], and a group of researchers found that almost 12 out of every 25 individuals aged 15-19 years in the United States reported having had sexual intercourse at least once [10]. In United States, the median age at first sexual debut was 17 years, which is higher than that in Jamaica (15.0 years) [11, 12]. Like United States, New Zealand and Jamaica, some African nations (such as Uganda, Kenya, Ghana, Tanzania, Zambia and Zimbabwe) had a median age which is statistical the same, suggesting that premarital sexual behaviour is similar in many developing and particular developed societies. A previous study conducted by Wilks et al [13], using a national probability same survey of 2,848 Jamaicans aged 15-74 years, found that
336

22 out of every 25 people aged 15-24 years have had sexual intercourse - 21 out of every 25 males aged 15-24 years and 19 out of every 25 females of the same age [13]. The sexual expression and practices of young Jamaicans (aged 15-24 years) is embedded in the fact that 11 out of every 25 have sex at least once per week - 11 out of every 25 males and 10 out of every 25 females [13]. Statistics also showed that 2.6% of Jamaicans aged 15-24 years had a STI in the last 12 months compared with 2.4% of Jamaicans aged 15-74 years old. Comparatively between the United States and Jamaica, less Americans aged 14-22 years were sexually active compared to Jamaicans aged 15-24 years [9, 13]. However, there were similarities between Jamaica and the United States as the age at sexual debut for males and females was relatively close [9, 13], suggesting congruency in sexual expressions. Using dataset for the 2002 Reproductive Health Survey in Jamaica [12], the mean age at first coitus was 14.7 years (SD = 3.1, median age at first intercourse = 15.0, range = 13 16 years) [14], and the median age of first coitus among females aged 16-49 years was 16.0 years in 2001, this fell from 17.3 years in 1997 [12]. The rationales for using < 16 years and 16+ are (1) the age of individual sexual consent is 16 years, and (2) the median age of first coitus among females aged 15-49 years was 16 years. Inspite of public health campaigns to address (1) the lowering of age of sexual intercourse, (2) HIV/AIDS among the population, particularly among adolescents and young adults, (3) sexual promiscuity, (4) inconsistent condom usage, (5) unwanted pregnancies and (6) better sexual practices in the world, particularly in Jamaica, the society has seen the continuous erosion of values because the aforementioned matters continue unabated and there seems to be no end in sight. Many developed nations such as New Zealand and the United States is
337

experiencing the early age of sexual debut epidemic like Jamaica. Apart of the justification of this public health challenge is that lifestyle practices, cultural values and expectation as well as orientations which are changing in the 21st century. Although females in world have been living longer than males (life expectancy or healthy life expectancy), which is the case in Jamaica, statistics revealed that the incidence of STIs among female for 2007/2008 in Jamaica were greater for them than their male counterparts [13]. This is within context of increased public health education campaigns on sexual responsibility and the rise of HIV/AIDS in the nation. Embedded in the incidence of STIs are the cultural values, lifestyle, norms, beliefs and sexual practices of females, which will not easily change because external agents such as health educators and professionals say that they are to do this. The literature on age at first sexual intercourse is extensive but recent and factors that determine contraceptive use of female [2-7, 15, 16], but no research existed that examined differentials in factors of contraceptive use between females whose first coital activity was < 16 years and 16+ years old. Bourne et al. [16] eight factors were statistical associated with

contraceptive use among females aged 15-49 years. The factors were age (OR = 0.95, 95%CI = 0.98 0.99); social class (upper class, OR = 0.83, 95%CI = 0.73 0.95); area of residence (rural, OR = 1.16, 95%CI = 1.02 1.32); currently pregnant (OR = 0.01, 95%CI = 0.00 0.02); had sex in last 30 days (OR = 2.29, 95%CI = 1.95 2.70); number of sexual partners (OR = 1.85, 95%CI = 1.57 2.17); age began using method of contraception (OR = 0.99, 95%CI = 0.98 1.00), and crowding (OR = 1.4, 95%CI = 1.21 1.60). If research provides an understanding of issues in our physical and social milieu, then, a study on the aforementioned is critical and timely as it would provide insights into their behaviour, thereby allowing health practitioners and educator to
338

better understand how to address the increasing HIV/AIDS virus and other public health problems such as unwanted pregnancies and unsafe abortions. With previous studies having demonstrated that early sexual activities are associated with increased HIV/AIDS infections, cervical cancers and other health problems [1-6, 15], understanding early sexual activity (before the statutory age 16 years in Jamaica) and post the statutory age will provide invaluable insights into practices and measure that can be formulated to address the lifestyle of these individuals. This current paper, recognizing limitations of previous research on the aforementioned issue within the context of the increased HIV/AIDS virus, unwanted pregnancy, abortions and high fertility [17-19] coupled with the continuous lowering of age of sexual debut over the decades, can add value to public health by studying factor differentials in contraceptive use between females whose first coital activity was < 16 years and those 16+ years old as well as their demographic profile. Such a research is timely and will guide policy formulation and intervention programmes. The rationales for the study are primarily based on (1) females vulnerability in contracting HIV/AIDS and other STI, (2) females being less economic independent than their male counterparts, (3) the vetoing power of males over females reproductive health choices in developing nations, (4) income inequalities between the genders, and (5) the issue of survivability. This research aims to elucidate information on the differentials in factors of contraceptive use between females whose first coital activity was < 16 years and 16+ years old and to provide a socio-demographic and reproductive health profile of these individuals.

Methods
Sample (participants) and procedures
339

A descriptive cross-sectional study was carried out by the National Family Planning Board (Reproductive Health Survey or RHS). There are two sets of inclusion criteria, which are females and ages. The eligibility criterion for age was 15 to 49 years at last birthday. In 2002, RHS collected data on Jamaican men ages 15-24 years as well as women 15-49 years old. The current paper extracted only females aged 15-49 years from 2002 Reproductive Health Survey (RHS) dataset to carry out this research. The female sample for the 2002 RHS was 7,168 women of the reproductive ages, with a response rate of 77.6%. Of those who responded (n=5, 565), 32.5% had first coitus before 16 years old compared with 67.5% who began at 16+ years old. Thus, the entire female sample for the 2002 RHS that responded to the survey was used for this study. Stratified random sampling was used to design the sampling frame from which the sample was drawn. Using the 2001 Census sector (or sampling frame), a three-stage sampling design was used. Stage 1 was the use of a selection frame of 659 enumeration areas (or enumeration districts, EDs). This was calculated based on probability proportion to size. Jamaica is classified into four health regions, which constitute particular parishes (there are 14 parishes). Region 1 is composed of Kingston, St. Andrew, St. Thomas and St. Catherine; Region 2 comprises Portland, St. Mary and St. Ann; Region 3 is made up of Trelawny, St. James, Hanover and Westmoreland, with Region 4 being St. Elizabeth, Manchester and Clarendon. The 2001 Census showed that Region 1 comprised 46.5% of Jamaica compared to Region 2, at 14.1%; Region 3 at 17.6% and Region 4 at 21.8% [12]. In stage 2, the households were clustered into primary sampling units (PSUs), and each PSU constituted an ED, which in turn was comprised of 80 households. The previous sampling frame was in need of updating, and so this was performed between January and May 2002. The
340

previous sampling frame was in need of updating, and so this was carried out between January 2002 and May 2002. The new sampling frame formed the basis upon which the sampling size was computed for the interviewers to use. Again, the sample was selected based on probability proportion to size of the four regions, and interviewers were given particular ED(s) which they exhausted in a clockwise manner. Stage 3 was the final selection of one eligible female from each sampled household and this was done by the interviewer on visiting the household [12]. The Statistical Institute of Jamaica (STATIN) provided the interviewers and supervisors, who were trained by McFarlane Consultancy, to carry out the survey. The instrument administered was a 35-page questionnaire. The data collection began on Saturday, October 26, 2002 and was completed on May 9, 2003. Prior to the date of the final data collection, pre-testing of the instrument was conducted between March 16 and 20, 2002. Modifications were made to the pre-tested instrument (questionnaire), after which the final exercise was carried out. Validity and reliability of the data were conducted by many statisticians, statistical agency, and university scholars before the data was used as the data are for national policy planning [12]. After which it was released to the University of the West Indies, Mona, Data Bank for use by scholars. The data was weighted in order to represent the population of female aged 15 to 49 years in the nation [12]. Measures Age at first sexual debut (or initiation or intercourse) was measured based on a respondents answer to the question At what age did you have your first intercourse? Crowding is the total number of persons in a dwelling (excluding kitchen, bathroom and verandah). Age is the number
341

of years a person is alive up to his/her last birthday (in years). Contraceptive method comes from the question Are you and your partner currently using a method of contraception? , and if the answer is yes Which method of contraception do you use? Age at which began using contraception was taken from How old were you when you first used contraception? Area of residence is measured from In which area do you reside? The options were rural, semi-urban and urban (1 = rural, 0 = otherwise; 1 = semi-urban, 0 = otherwise, and urban is the reference group). Currently having sex is measured from Have you had sexual intercourse in the last 30 days? (1=yes, 0 = otherwise). Education is measured from the question How many years did you attend school? Marital status is measured from the following question Are you legally married now?, Are you living with a common-law partner now? (that is, are you living as man and wife now with a partner to whom you are not legally married?), Do you have a visiting partner, that is, a more or less steady partner with whom you have sexual relations?, and Are you currently single? Age at menarche is measured from How old were you when your first period started (first started menstruation)? Gynaecological examination is taken from Have you ever had a gynaecological examination? (1 = yes, 0 = no). Pregnancy was assessed by Are you pregnant now? (1=yes, 0 = otherwise or no). Religiosity was evaluated from the question With what frequency do you attend religious services? The options range from at least once per week to only on special occasions (such as weddings, funerals, christenings et cetera) (1=frequent attendance from response of at least once per week, 0 = otherwise). Subjective social class is measured from In which class do you belong? The options are lower, middle or upper social hierarchy (1 = middle class, 0 = otherwise; 1 = upper class, 0 = otherwise; reference group is lower class). Forced to have sexual relations was assessed from the question Were you forced to have sex at your first intercourse? and the options were yes, no, dont know and refused to
342

answer (1= yes, 0 = otherwise). Age at first sexual debut, age at menarche, age at first contraceptive use, and years of schooling were used as continuous variables. Early sexual debut is having sexual intercourse before the statutory legal age to do so (in Jamaica, this is 16 years old). Statistical analyses Data were entered, stored and retrieved using SPSS for Window, Version 16.0 SPSS Inc; Chicago, IL, USA). Descriptive statistics were performed on particular sociodemographic characteristics of the sample (frequency, mean, standard deviation (SD), and range). All metric variables were tested for normality (age at first sexual debut, crowding, age, and years of schooling). Where skewness was found to be less than 0.5, the variable was used in its current form and a value more than 0.5 was normalized by natural log. Independent sample t-test was used to examine differences in age at sexual debut between those who frequently attend churches and those who infrequently visit churches and F-statistic was employed for age of sexual debut and subjective social class (Table 13.4). Chi-square analyses were used to examine two nonmetric variables (Table 13.4). Pearson Product Moment correlation was used to evaluate statistical association between age of first sexual intercourse and number of sexual partners for the sample. Stepwise logistic regression analyses were used to fit the one outcome measure (contraceptive use) by different sociodemographic as well as reproductive health variables. Thus, only explanatory variables (i.e. statistically significant variables) are shown in Table 13.5. Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final model construction [19]. To derive accurate tests of statistical significance, we used SUDDAN statistical software (Research Triangle Institute,
343

Research Triangle Park, NC), and this adjusted for the surveys complex sampling design. A pvalue < 0.05 (two-tailed) was used to establish statistical significance.

Results
Demographic characteristic of sample Table 13.1 presents information on the demographic characteristic of the studied population by age at first coital activity (< 16 years or 16+ years old). Of the studied respondents, 7.3% had their first sexual intercourse at most 13 years old, 16.7% at most 14 years old, 32.5% at most 15 years old, 51.4% by at most 16 years, 92.6% by at most 20 years old and 99% by at most 26 years old. Twenty one percentages of the respondents had no sexual partner, 75.6% had one sexual partner compared with 3.4% who had 2+ sexual partners. Table 13.2 highlights particular reproductive health characteristic of studied population by age at first coital activity (< 16 years or 16+ years old). Table 13.3 displays information on methods of contraception Method of contraception and when began using by age at first coital activity (i.e. < 16 or 16+ years old). Table 13.4 forwards information on particular demographic variables by subjective social class of respondents controlled for by age at first coital activity (i.e. < 16 or 16+ years old). On examination of age at first sexual intercourse and number of sexual partners for the past month and the former 3 months, a significant statistical correlation was found between (1) age at first sexual intercourse and number of sexual partners in the last 4 weeks (rxy = - 0.034, P = 0.011), and (2) age at first sexual intercourse and number of sexual partner in the last 12 weeks (rxy = - 0.037, P = 0.006).
344

A significant statistical difference was found among the subjective social classes and age at first sexual intercourse (F = 187.4, P<0.0001). Females in the lower socioeconomic stratum began having sex at 16.0 years (SD = 2.3) compared with 16.5 years (SD = 2.4) for those in the middle class and 17.8 years (SD = 3.2) for those in the wealthy socioeconomic stratum. However, no statistical difference emerged among the subjective social classes and number of sexual partners (F = 2.23, P = 0.107). On average, crowding was 1.9 persons (SD = 0.30) among females who were in the lower socioeconomic stratum compared with 1.8 persons (SD = 0.43) for those in the middle stratum and 1.3 persons for those in the wealthy socioeconomic stratum F-statistic = 252.03, P<0.0001. Females who frequently attend church begins having sex at 17.4 years (SD = 3.5) compared with 16.4 years for those infrequent female church attendees (t-test = - 12.56, P<0.0001). Multivariate analyses Table 13.5 shows explanatory factors which account for contraceptive use among females in Jamaica aged 15-49 years based on age at first sexual activity that the individual is classified in (i.e. < 16 or 16+ years old).

Discussion
A previous study had that Experiences at sexual debut may be linked to reproductive health later in life [21, p. 1] and that the age of first sexual debut is associated with future reproductive health outcomes [1-6]. The current works concurs with the literature, and provide detailed information on the differences on demographic profile and factor differentials in contraceptive
345

use between the two cohorts (females aged 15-49 years who began having sexual intercourse < 16 years and those who started at 16+ years). This research found that females whose first sexual intercourse happened before 16 years old were less likely to use a condom with a steady partner, do Pap smear and gynaecological examination as well as utilize the pill as a method of contraception, but they were more likely to be in the lower socioeconomic stratum, live in rural areas, have a lower educational level, first sexual intercourse was forced, use injection as a method of contraception, shared sanitary convenience, currently in a sexual relationship, sexual partnerships in last 3 months and unemployed. Factor differentials on contraceptive use emerged between the two cohorts. These were social class (upper class: OR = 0.72, 9%% CI = 0.55 0.94) for those who begin < 16 years old but not for those 16+ and area of residence (Rural area: OR = 1.26, 95% CI = 1.07 1.47) for the latter but not the former. Embedded in those findings is the fact that females who are in the upper socioeconomic stratum that commenced sexual intercourse before 16 years are engaged in riskier sexual practices than those in the lower class. In Jamaica, statistics revealed that females are poorer and less employed compared with males [22, 23]. This reality means that there is high economic dependence of females on males for financial survivability, making young females within the lower socioeconomic stratum having different reproductive health outcome than those in the wealthy socioeconomic strata because of their socio-economic marginalized situation. Many of these females commenced sex at an early age because of economic vulnerability, and so they are likely to be engaged high-risk behaviours [21]. On the other hand, in order to provide for themselves many females who are within the lower socioeconomic stratum become involved with older men who expose them to the same risk
346

of pregnancy, STIs, and HPV. With females in the lower socioeconomic stratum having more people in a dwelling area compared with those in the other socioeconomic strata, they will turn outside the household for financial assistance and oftentimes this is provided in visiting sexual unions in which the males are older. In such unions, because females are in a socioeconomic vulnerable position and by extension poorer and marginalized, males are able to dictate many things including reproductive health choices. Females, therefore, in those income class will bear children as an economic flows and/or some will have unsafe abortions, but those in the upper class are able to carry out safe abortions compared with those in the lower class because of access to financial resources, and where they consider their lives. Thus, the aforementioned arguments justify female who began sexual intercourse at most 15 years who are more likely to be in the lower class, dwell in rural areas, unemployed, have multiple sexual partners and less educated were more likely to be engaged in sexual relationships, and forced into sexual activities. Their economic vulnerabilities account for the rationale of using fewer condoms as a method of contraception because this is vetoed by the male. Money is important to women, but the risky sexual behaviour of upper class females whose first sexual activity begins before 16 years old is not for the money as those in the lower socioeconomic strata. The high risk sexual behaviour among upper class females whose first sexual intercourse was before 16 years, suggests that many of them would have abortions, STIs and even HPV because of their lifestyle practices. The work also showed a negative correlation between number of sexual partners and age at first coitus, indicating that younger females are more promiscuous and that this changes with age at they move into stable sexual unions. Simply put the adolescence years are about fun, frolic, sexual freedom, sexual expression, inconsistent

347

condom usage and sexual carelessness, which seems to continue even in the adult years among wealthy females. Even though money is important to particular reproductive health outcomes (such as safe abortions), early sexual intercourse comes with less likeliness of a method of contraception, which is because of ignorance. It was revealed from the findings that those females who commenced sexual intercourse at older ages were more likely to use a particular method of contraception (pill) than condoms that expose them to STIs, HPV, HIV/AIDS and pregnancies, which is in keeping with the literature from other nations [2, 21, 24,25]. Embedded in this finding is the influence of knowledge of contraceptive with age, and not money. While money is associated with employment and other socioeconomic benefits, it is not responsible for lower method of contraception among Jamaicans females. Rural poverty in Jamaica is about twice urban poverty, with more people residing in rural areas and a sex ratio that is greater for females than males [22, 26], if money matters, then rural females who begins having sexual intercourse at 16+ years would not be 1.3 times more likely to use a method of contraception compared with those in urban areas. Or, those in those whose families are in the wealthy strata would be more likely to use a method of contraception compared with those in the lower socioeconomic stratum, but the reverse is true in Jamaica. Embedded in these findings are inexperience and the euphoria surrounding first sexual activity as well as the age of the initiating partner that account for lower contraceptive use based on age at first sexual coital activity than money. According to Gomez et al. [21], Sixty-five percent of women reported sexual initiation with a partner younger or less than 5 years older, 28% with a partner 5 to 10 years older, and 7% with a partner 10 or more years older, and in Jamaica a
348

study revealed that many young women began their sexually initiation with men at least 5 years older than them [12]. Embodied here is an understanding of the lifestyle of adolescents in regarding to sex, and how older men can expose them to sexually transmitted infections. The media continues to glamorize sex and sexuality, which are capturing the attention and practices of young people. The young females are culturized in sex, and this they see to explore as they become cognizant of sex during the adolescent years when there is growth and development of the body. Even with age, knowledge, exposure and high accessibility to method of contraception and low cost of contraceptives, inconsistent condom use and condom use is low among Jamaican women aged 15-49 years. The current work revealed that 42.5% of those who began having sex before 16 years old currently use a condom consistently with their steady partner and the figure was 2.5% more among those who started at 16+ years old. This finding provides evidence of the difficulty to change lifestyle practices as although the majority of people in Jamaica have been exposed to public health education and intervention programmes [12], this has not significantly change their sexual behaviour as the age of sexual initiation continues to fall as well as an increase prevalence of HIV/AIDS among the populace. Abel-Smith is correct, therefore, when he claimed that people are prisoners of their lifestyle [27], suggesting that values, customs, norms and early socialization are difficulty to change, but that it is still possible over time. Apart of the Caribbean culture is that a woman is not a woman without bearing children, like the man [8, 28]. Such an orientation and culture, implies and dictates a diet of sex, inconsistent contraceptive use and risky sexual practices.

349

School is an agent of socialization, in which people are provided the tools of socioeconomic survivability, has become a place of indirectly promoting sex through sexual education and peers of different socioeconomic situations and background. The current findings revealed that 43% those whose first sexual activity started before 16 years old began using a method of contraception during school compared with 7% who started at 16+ years. With there being an inverse association between age and contraceptive use [4-7, 10, 16], it can be deduced that high contraceptive use is associated with sexual activities. Like Gomez [21], this study recognizing the importance of age and gender-based power differentials between the sexes regarding sex note that delaying sexual debut must understand those differences as well as the educational system. Dickson [7] opined that adolescent sexual behaviour is influenced by social factors. It can be deduced from Dicksons work that educational system is able to change sexual practices and particular reproductive health outcome. From the current research, the educational system has modified the use of contraception, but not increasing the age at sexual debut. During school, children are not only exposed to health and reproductive health education and subjects trainings, they are interfacing with other children of different socialization, lifestyle, values and orientations. With the glamorization of sex in the media, on cable television, many children are exposed to a diet of sex, and some will seek to practice this while attending school. This is reinforcing sex, sexuality and orientation of sex that is even covertly reinforced with reproductive health education in schools. Based on Bourne et al.s work [29] that Health education and health promotion are driven based on understanding lifestyle practices of a population [29], the current findings
350

provide some critical information that can be used for a new thrust into public health intervention programmes in the future that can be used to modify current practices. As formal educational is not able to change the sexual practices and/or reproductive health behaviour of females because more than 55% of the sample have tertiary level education (or have attained this level) compared to only 9.6% who have at most primary level education. The social and cultural values, orientation, beliefs, and expectations of the society are such that formal education is not modifying the lifestyle practices that public health specialists and behaviouralists would want to change. Clearly, a public health problem that emerged from the current paper is that 1.5 times more females who had sex before 16 years were sexually assaulted compared to those who began at 16 years and older. Outside of the obvious that many early sexual encounters among females at most 16 years is as a result of rape, the perpetrators are normally friends, family members and/or acquaintances who carry out these acts against the physical vulnerable adolescents and children [30, 31]. Such abase leave an indelible psychological scar for the adolescent and Lowe et al. [32] posited that this leaves immense psychological trauma which are sometimes are suicidal. Another psychological matter which is a consequence of sexual assault of is aggression on the path of the victim [33], suggesting that the sexual appetite of Jamaican males is exposing female adolescent and children to future psychological traumas as well as reproductive health problems. This matter becomes even more complex when the adolescent is found to be pregnant, family is poor, lowly educated, unemployed and religious. One researcher found positive statistical correlations between poverty and not seeking medical care (R = 0.576), and poverty
351

and unemployment (R = 0.48) [34], indicating that economic vulnerable adolescents and their families are likely to see the young female doing unsafe abortion, carrying the pregnancy to term and going into depression and/or other psychological traumas because of socioeconomic deprivation. No or little access to money means less choices including abortion for females who become pregnant as a result of rape and the economic power of the perpetrator is also able to change the outcome of criminal conviction. Thus public health practitioners need to recognize money and power as influencing reproductive health, and how these may retard self autonomy of the females, particularly those young females who are from low socioeconomic background. The socio-economic consequences of poverty, low educational attainment, self-esteem and social isolation can, therefore, influence public health intervention programmes [36], making it difficult for public health practitioners to be effective in meeting their objectives without addressing those inadequacies and the social structure in the society. Religiosity is associated with better sexual practice as it increased the age of first sexual intercourse, which concurs with the literature [20, 37, 38]. The church which is a part of the social structure is delaying sexual intercourse among Jamaican females aged 15-49 years by one year, which speaks to the embedded sex culture and the difficulty in changing this practice without structural and cultural changes, over time. Again this reinforces the fact that delaying early sexual behaviour is also a future good as people will continue bad practices if they start early in life. Research evidence demonstrates that the religiosity network in which the adolescent involved as well as the friends religious positively lowers age at first coital activity [39]. With the number of churches in Jamaica, particularly in the lower socioeconomic areas, it is paradoxical that age at first sexual intercourse continues to fall. Some of those issues can be explained by the economic deprivation in inner-city communities and the culture values, beliefs
352

and customs within the society as well as the sub-cultures and countercultures on sex and sexuality. Clearly, the culture in inner-city communities coupled with crowding are fostering early sexual intercourse because those in the lower socioeconomic stratum commenced sexual intercourse on average at 16 years compared with 16.5 years for those in the middle class and 17.8 years among those in the wealthy stratum. It can be deduced and extrapolated from those figures that men are using the economic vulnerability of young females against them, and this is resulting in those females becoming engaged in transactional sex. They are exchanging sex for good, commodities and other support things for sex from older men. Although the same is not the case for females in the wealthy socioeconomic stratum, those who starting having sex before 16 years old are currently engaged in risky sexual behaviour. This speaks to the early lifestyle practices, values which were garnered during that period and its bearing on current practices. Thus, old habits are difficult to change. This is the difficulty that public health practice need to tackle, those who began having sexual intercourse at most 15 years old as they are high sex risk takers even in the adults years. One study demonstrates this aptly as the researchers found that children are significantly more likely to become sexually active before age 14 if their mother had sex at an early age and if she has worked extensively [40] Previous studies have demonstrated that many of the cases of sexual assault and rapes are perpetrated by acquaintances. With the crowding being an issue in inner-city communities (or lower socioeconomic areas), a number of the sexual initiations occur as a result of this fact. The adolescents are sometimes gullibly encourages to become involvement in sexual activities with family members, household members and friends. With the crowding in inner-city communities
353

means that many of the rapes are perpetrated by non-household members by acquaintances in the area. The next issue is the associations of the adolescents, and whether those networks are among religious members or non-religious individuals. Hardy and Raffaelli [38] provide an explanation for the previously mentioned situation. They opined that religiosity delay the transition of adolescents venturing into sexual activity, suggesting that religion is a social control. It follows, therefore, that adolescents who are friends of non-religious individual would not have this level of control and will initiate sexual intercourse early. The peer group influences the reproductive health outcome of people, particularly children and/or adolescents as well as adults [41] and increases early sexual practices which in this case justify future sexual behaviour of adults. It is this explanation why public health practitioners need to address social institutions in thwarting a campaign that will foster better sexual practices of adults as early as childhood and during their adolescence years. The traditional approach to health behaviour modification was to give people knowledge about a particular issue, practice or happenings within their sociophysical milieu and instruct them into a new path [42]. According to one group of researchers, in 2009, Knowledge about the prevalence of sexual risk behaviour (SRB) in adolescence is needed to prevent unwanted health consequences [43], and this justifies the continuation of poor sexual practices in the future. Such an argument implies that lifestyle behaviour is easily changeable, which is the fartherest from the reality. This is captured in the current work which showed that educational attainment is not associated with usage of contraceptives. On the contrary, those in the wealthy income stratum had the greatest prevalence of tertiary level education, yet those who started having sexual intercourse before 16 years were less likely to use a method of contraception. Thus, education cannot easily change peoples behaviour and so it is about knowledge on a
354

particular issue. This is capture in the Wilks et al.s work [13] which found that in 2002 78.3% of Jamaicans aged 15-74 years used a condom with their main partner and this fell to 43.1% in 2008 although the percentage of Jamaicans with secondary-to-tertiary level education had increased, with 11.3% having had tertiary level training. They also found that more people were engaged in visiting and/or single unions compared with married and common law, and the more people had 2+ sexual partner in 2008 (24.4%) compared to 2000 (23.0%). The current work that showed that adult women who began having sex at 16+ years were more likely to use a method of contraception than those who started before 16 years, this suggests that risky sexual behaviour which commenced early in life is likely to continue into adulthood. Again, people are prisoners to their culture, social structure, values, beliefs, and socialization. Cohen, Scribner and Farley [44] developed a model for behaviour change using structural modeling which addresses physical structures, social structures, cultural and media messages. Like Cohen et al. [44], Bourne et al. opined that health promotion for Jamaicans must include social, economic, and lifestyle choices [29]. In the previous works, the authors recognizing the complexity of humans have coalesced a multidimensional apparatus to address behaviour change and not simply imparting knowledge or by formal education. Although a group of scholars found that the womens level of education and that of her spouse and age determine contraceptive use, this concurs and disagrees with those findings [45, 46]. For the current work, age is a factor in contraceptive use, which is supported by the literature [16, 45], but the same cannot be said about education. Education is not changing sexual practice as it relates to contraceptive use among Jamaica females, despite its provision in imparting knowledge and behaviour medications. People are not barrels in which they are fed a
355

diet of information from external sources such as health educators to want them to carry out a particular action or cease one because the social and environmental factors influence behaviour, particularly contraceptive use [47]. Hogan et al. [47], provided some clarifications to the social and other factors which are associated with contraceptive use, when they postulated that, Social and environmental variables were found to affect contraceptive preparedness at 1st intercourse only, and not subsequent initiation of contraceptive practice [47]. Outside of this clarification, it is evident that the culture, physical milieu, values, and beliefs impact on people behaviour and this include education, but that this is not the case among female Jamaicans aged 15-49 years old whether sexual initiation was < 16 years or 16+ years old. There is cultural conflict among female Jamaicans, the health care system and the health care educators because the symbols of the culture and ways of life are not supported by the health care educators, particularly related with sexual practices, sex and reproductive health matters. Embedded in the current findings is the value of the social environment in which these females live and grow, which fashion their cultural development, identification and belief system. Those are the reasons why Morally unacceptable policies designed to pressure or compel people to limit their fertility have been shown to be unnecessary and thus have been abandoned, except in China [48] as well as being ineffective in behaviour medication, and any such similar public health intervention programmes that used force, moral suasion or dictatorial stance.

Conclusion
Early sexual initiation is influencing future health and reproductive health outcomes among Jamaican women aged 15-49 years old. Those outcomes include more coital activity, involvement in sexual unions, and less contraceptive use. Despite reproductive health education
356

programmes in Jamaica, the culture is clearly retarding good reproductive health practices and sexual lifestyle. In Jamaica, although fertility is lower and educational advancement is greater in urban than in rural areas, rural females whose first coitus began at 16+ years were more likely to use a method contraception compared with their urban counterparts. Clearly, there is a lifestyle change occurring among females in rural areas which needs examination, and equally so is the risky sexual practices of affluent females who started having sex before 16 years old. With the global economic downturn, sexual autonomy of female Jamaicans will be further reduced, particularly those in the lower socioeconomic stratum, unemployed, uneducated, and young because males will now have greater vetoing powers over sex, sexuality and reproductive health matters. Public health practitioners have not begun to address those realities in the communities and human rights of women will be thwarting because money is important in survivability. Sexual rights of women cannot be supported by merely ascribing it to them or penning social constructions in this regards in must be supported by economic independency. While legislation and policies that promote sexual autonomy are good, the reality is money is power, and with the economic downturn in the Jamaican economy there will be greater promiscuity as women seek more assistance in sexual relationships, which is embedded in Wilks et al.s work which showed an increase in visiting unions and number of sexual concurrent partners between 2000 and 2008. Because money is associated with better education, physical milieu, social opportunities, good nutrition and sexual autonomy; to asked the question If women are so keen to avoid pregnancy, why do they not use a method of contraception? [49] is to deny people of their social environment and the role of money in it. There will be in social justice in society that does
357

not understand the factors which are associated with sexuality, rights and sexual justice; and the role of money in influencing health and reproductive health matters. It means that apart of the sexual lifestyle of females is justified by the economic situation in the communities [50, 51], nation and the world. Such social and financial environments means that public health must begin to address the new reality as all the gains that have been accomplished in past decades will be erodes because of the increased economic vulnerability of peoples and economic marginalization of the poor, particularly among young, uneducated, and unemployed females. In summary, delaying age at first sexual intercourse influences contraceptive use, by increase methods of contraception. It also fosters good sexual practices in the future. Clearly, the reproductive health problems in Jamaica are structurally driven which care embedded in the cultural values that make it difficult for public health practitioners to address without including those issues in health education, communication and intervention programmes. Because people are sexual being, sex will always be a part of their social existence and an issue that cannot be left unaddressed by public health policies makers within the current findings and the global economic downturn. There is a need for structural changes in developing as well as developed nations to address many reproductive health matters. The factors of method of contraception are not the same across the age cohort at which a female began having sexual intercourse, and they are also some different to those of women in the reproductive ages 15-49 years old. The findings which emerged from the current results are far reaching and can be used to guide new public health intervention programmes. Disclosures The authors report no conflict of interest with this work.
358

Disclaimer
The researchers would like to note that while this study used secondary data from the Reproductive Health Survey, none of the errors in this paper should be ascribed to the National Family Planning Board, but to the researchers.

Acknowledgement
The authors thank the Data Bank in Sir Arthur Lewis Institute of Social and Economic Studies, the University of the West Indies, Mona, Jamaica for making the dataset (2002 Reproductive Health Survey, RHS) available for use in this study, and the National Family Planning Board for commissioning the survey.

359

References
1. World Health Organization (WHO). Comprehensive cervical cancer control: A guide to essential practice. Geneva: WHO; 2006. 2. Pettifor AE, van der Stratan A, Dunbar MS, Shiboski SC, Padian NS. Early age of first sex: A risk factors for HIV infection among women in Zimbabwe. Epidemiology and Social 2004; 18(10):1435-1442. 3. Andersson-Ellstrom A, Forssman L, Milsom I. Age of sexual debut related to life-style and reproductive health factors in a group of Swedish teenage girls. Acta Obstet Gynecol Scand 1996; 75:4849. 4. Coker AL, Richter DL, Valois RF, et al. Correlates and consequences of early initiation of sexual intercourse. J Sch Health 1994; 64:3727. 5. Zaba b, Pisani E, Slaymaker E, et al. Age at first sex: Understanding recent trends in African demographic surveys. Sex Transm Infect 2004; 80 (Suppl II):ii28-ii35. 6. Kaestle CE, Halpern CT, Miller WC, Ford CA. Young age at first sexual intercourse and sexually transmitted infections in adolescents and young adults. Am J of Epidemiology 2005; 161(8):774-780. 7. Dickson N, Paul C, Herbison P, Silva P. First sexual intercourse: Age, coercion, and later regrets reported by a birth cohort. BMJ 1998; 316:29-33. 8. Chevannes B. Learning to be a man: Culture, socialization and gender identity in five Caribbean communities. Kingston, Jamaica: The Univer. Of the West Indies Press; 2001. 9. Santelli JS, Brener ND, Lowry R, Bhatt A, Zabin LS. Multiple sexual partners among U.S. adolescents and young adults. Family Planning Perspectives 1998; 30:271-275. 10. Abma JC et al., Teenagers in the United States: sexual activity, contraceptive use, and childbearing, 2002, Vital and Health Statistics, 2004, Series 23, No. 24. 11. The Alan Guttmacher Institute (AGI), In Their Own Right: Addressing the Sexual and Reproductive Health Needs of American Men, New York: AGI, 2002. 12. Jamaica National Family Planning Board (NFPB). Reproductive Health Survey, 2002. Kingston: NFPB; 2005. 13. Wilks R, Younger N, Tulloch-Reid M, McFarlane S, Francis D. Jamaica health and lifestyle survey 2007-8. Kingston: Tropical Medicine Research Institute, University of the West Indies, Mona; 2008. 14. National Family Planning Board, 2002 [Computer file]. Kingston, Jamaica: National Family Planning Board [producer], 2002. Kingston, Jamaica: Derek Gordon Databank, University of the West Indies [distributors], 2003. 15. Ma Q, Ono-Kihara M, Cong L, et al. Early initiation of sexual activity: a risk factor for sexually transmitted disease, HIV infection, and unwanted pregnancy among university students in China. BMC Public Health 2009; 9: 111. 16. Bourne PA, Charles CAD, Crawford TV, Kerr-Campbell MD, Francis CG, South-Bourne N. Current use of contraceptive method among women in a middle-income developing country. Open Access J of Contraception 2010; 1:39-49. 17. Pan American Health Organization (PAHO). Health in the Americas, 2007, volume II Countries. Washington, D.C.; 2007: pp.448-464. 18. Bain B. HIV/AIDS the rude awakening/stemming the tide. In: Morgan O. ed. Health issues in the Caribbean. Kingston: Ian Randle Publisher; 2005: pp. 62-76.
360

19. World Health Organization (WHO). World Health Statistics, 2009. Geneva: WHO; 2009. 20. Bourne PA, Charles CAD. Contraception usage among young adult men in a developing country. Open Access J of Contraception 2010; 1:51-59. 21. Gomez AM, Speizer IS, Reynolds H, Murray N, Beauvais H. Age differences at sexual debut and subsequent reproductive health: Is there a link? Reproductive Health 2008; 5:8. 22. Planning Institute of Jamaica (PIOJ), Statistical Institute of Jamaica. Jamaica Survey of Living Conditions, 1988-2007. Kingston: PIOJ, STATIN; 1989-2008. 23. Planning Institute of Jamaica (PIOJ). Economic and social survey, Jamaica, 1970-2008. Kingston; PIOJ; 1971-2009. 24. Mardh PA, Creatsas G, Guaschino S, Hellberg D, Henry-Suchet J: Correlation between an early sexual debut, and reproductive health and behavioral factors: a multinational European study. Eur J Contracept Reprod Health Care 2000, 5:177-182. 25. Mnyika KS, Klepp KI, Kvale G, Ole-Kingori N: Determinants of high-risk sexual behaviour and condom use among adults in the Arusha region, Tanzania. Int J STD AIDS 1997, 8:176-183. 26. Statistical Institute of Jamaica (STATIN). Demographic statistics, 1989-2007. Kingston; STATIN; 1990-2008. 27. Abel-Smith B. An introduction to health: Policy, planning and financing. London: Pearson Education; 1994. 28. Barrow C. Caribbean Gender Ideologies: Introduction and Overview. In: Barrow C. ed. Caribbean Portraits: essays on Gender Ideologies and Identities. Kingston, Jamaica: Ian Randle Publishers; 1989: pp.xi-xxxviii 29. Bourne PA, McGrowder DA, Holder-Nevins D. Public Health Behaviour-Change Intervention Model for Jamaicans: Charting the Way Forward in Public Health. Asian Journal of Medical Sciences 2010; 2(2):56-61. 30. United Nations Childrens Fund (UNICEF). Child Protection. 2008. Retrieved on August 26, 2010 from: http://www.unicef.org/jamaica/violence.html. 31. Walker et al. Nutritional and health determinants of school failure and dropout in adolescent girls in Kingston, Jamaica. International Centre for Research on Women: Washington DC; 1994. 32. Lowe GA, Gibson RC, Christie CD. HIV infection, sexual abuse and social support in Jamaican adolescents referred to a psychiatric service. West Indian Med J 2008; 57(3):307-311. 33. Le Franc E, Samms-Vaughan M, Hambleton I, Fox K, Brown D. Interpersonal violence in three Caribbean countries: Barbados, Jamaica, and Trinidad and Tobago. Rev Panama Salud Public 2008; 24(6):409-421. 34. Bourne PA. Impact of poverty, not seeking medical care, unemployment, inflation, selfreported illness, health insurance on mortality in Jamaica. North American Journal of Medical Sciences 2009; 1(3):99-109. 35. Berer M. Power, money and autonomy in National Policies and Programmes. Reproductive Health Matters 2003; 12(23):6-13. 36. Buston K, Williamson L, Hart G. Young women under 16 years with experience of sexual intercourse: who becomes pregnant? J Epidemiol Community Health 2007; 61(3):221-225.
361

37. Sexual Initiation. American Sexual Behaviour. http://www.newstrategist.com/productdetails/Sex.SamplePgs.pdf (Accessed on April 1, 2010). 38. Hardy SA, Raffaelli M. Adolescent religiosity and sexuality: an investigation of reciprocal influences. Journal of Adolescence 2003; 26:731-739. 39. Adamczyk A, Felson J. Friends religiosity and first sex. Social Science Research 2006; 34(4):924-947. 40. Mott FL, Fondell MM, Hu PN, Kowaleski-Jones L, Menaghan EG. The determinants of first sex by age 14 in a high adolescent population. Family Planning Perspectives 1996; 28(1):13-18. 41. Sieving RE, Eisenberg ME, Pettingell S, Skay C. Friends influence on adolescents first sexual intercourse. Perspectives on Sexual and Reproductive Health 2006; 38(1):13-39. 42. Glanz K, Rimer BK, Lewis FM. Health Behavior and Health Education: Theory, Research, and Practice, 3rd. CA: John Wiley and Sons; 2002. 43. Kalina O, Geckova AM, Jarcuska P, Orosova O, van Dijk JP, Reijneveld SA. Psychological and behavioural factors associated with sexual risk behaviour among Slovak students. BMC Public Health 2009, 9:15. 44. Cohen D, Scribner R, and Farley T. A structural model of health behavior: A pragmatic approach to explain and influence health behaviors at the population level. Preventive Medicine, 2000; 30, 146-154. 45. Tehrani FR, Farahani FKA and Hashemi MS. Factors influencing contraceptive use in Tehran. Family Practice 2001; 18(2): 204208. 46. Ekabua JE, Ebabua KJ, Odusolu P, Iklaki U, Agan TU, Etokidem AJ. Factors associated with contraceptive use and initiation of coital activity after childbirth. Open Access J Contraception 2010; 1:85-91. 47. Hogan DP, Astone NM, Kitagawa EM. Social and environmental factors influencing contraceptive use among black adolescents. Fam Plann Prospect 1985; 17(4):165-169. 48. Bongaarts J, Sinding SW. A response to critics of family planning programs. Int Perspec on Sexual and Reproductive Health 2009 35(1):39-44. 49. Ravindran TKS, Balasubramanian P. Yes to abortion but No to sexual rights: The paradoxical reality of married women in rural Tamil Nadu, India. Reproductive Health Matters 2004; 12(23):88-99. 50. Bourne PA, Rhule J. Good Health Status of Rural Women in the Reproductive Ages. International Journal of Collaborative Research on Internal Medicine & Public Health, 2009;1(5):132-155. 51. Bourne PA. Health status and Medical Care-Seeking Behaviour of the poorest 20% in Jamaica. International Journal of Collaborative Research on Internal Medicine & Public Health, 2009;1(6&7):167-185.

362

Table 13.1: Demographic characteristics of studied population Age at first coital activity 2, Pvalue < 16 years old 16+ years old Characteristic n = 1811 n = 3754 n (%) n (%) Area of residence 19.48, < 0.0001 Urban 265 (14.6) 668 (17.8) Semiurban 470 (26.0) 1156 (30.8) Rural 1076 (59.4) 1930 (51.4) Educational level 195.95, < 0.0001 Primary and below 225 (12.4) 306 (8.1) Secondary 832 (45.9) 1096 (29.2) Tertiary 739 (40.8) 2352 (62.7) Shared sanitary convenience 40.36, < 0.0001 No 1380 (76.9) 3101 (83.0) Yes 414 (23.1) 636 (17.0) Social class 182.61, < 0.0001 Lower 603 (33.3) 771 (20.5) Middle 839 (46.3) 1560 (41.6) Upper 369 (20.4) 1423 (37.9) Employed 71.05, < 0.0001 No 1158 (63.9) 1938 (51.6) Yes 653 (36.1) 1816 (48.4) Frequent church attendance 47.40, < 0.0001 No 1289(71.2) 2799 (62.0) Yes 522 (28.8) 1714 (38.0) Partner main source of financial 0.001, 0.979 support No 89 (42.6) 93 (42.3) Yes 120 (57.4) 127 (57.7) Age at first coital activity mean (SD) 14.1 yrs (1.1) 29.8 yrs (26.3) t=-40.01, <0.0001 Current age of respondents, mean (SD) 30. 5 years (9.2 yrs) 33.1 yrs (8.4) t=10.27, <0.0001 Crowding, mean (SD) 1.8 persons (0.42) 1.7 persons (0.5) t=9.02,<0.0001 SD denotes standard deviation

363

Table 13.2: Particular reproductive health characteristic of studied population Age at first coital activity < 16 years old 16+ years old Characteristic n = 1811 n = 3754 n (%) n (%) Want to be pregnant No 228 (12.6) 486 (12.9) Yes 83 (4.6) 172 (4.6) Missing 1500 (82.8) 3096 (82.6) Had sex (in last 30 days) No 573 (31.6) 1318 (35.1) Yes 1238 (68.4) 2436 (64.9) Forced to have sex (ever) No 1321 (73.1) 3072 (82.0) Yes 485 (26.9) 675 (18.0) Forced to have sex (first time had coital activity) No 1491 (82.9) 3367 (90.5) Yes 309 (17.1) 356 (9.5) Currently pregnant No 1725 (95.3) 3592 (95.7) Yes 85 (4.7) 161 (4.3) In sexual union No 310 (17.1) 789 (21.0) Yes 1501 (82.9) 2965 (79.0) Currently used method of contraception No 601 (34.3) 1302 (36.0) Yes 1151 (65.7) 2312 (64.0) Frequency of condom usage With steady partner Always 221 (42.5) 463 (45.0) Most times 259 (49.8) 493 (47.9) Seldom 29 (5.6) 66 (6.4) Never 2 (0.4) 1 (0.1) Never had a steady partner 9 (1.7) 7 (0.7) Missing 1291 (71.3) 2724 (72.6) With non-steady partner Always 93 (18.1) 111 (10.8) Most times 41 (8.0) 60 (5.9) Seldom 0 (0.0) 2 (0.2) Never 30 (5.8) 59 (5.8) Never a non-steady partner 351 (68.2) 7923 (77.3) Missing 1296 (71.6) 2730 (72.7) Number of sexual partners in last month mean (SD) 0.7 person (0.7) 0.7 person (0.8) Number of sexual partners in last 3 months - mean (SD) 1.1 person (1.4) 0.9 person (1.2)

2, Pvalue

0.005, 0.943

9.71, 0.002 64.19, <0.0001 82.18, < 0.0001 0.481, 0.488 16.22, < 0.0001 2.98, 0.084

8.58, 0.073

22.23, < 0.0001

t=1.78, 0.076 t=3.02, 0.003

364

Table 13.3: Method of contraception, when began using, gynaecological and Pap Smear examination by age at first coital activity (ie. < 16 or 16+ years old) Age at first coital activity < 16 years 16+ years old old n = 1811 n = 3754 n (%) n (%) 216 (11.9) 5 (0.3) 251 (13.9) 274 (15.1) 1 (0.1) 20 (1.1) 35 (1.9) 5 (0.3) 447 (24.7) 1 (0.1) 0 (0.0) 388 (10.3) 6 (0.2) 380 (10.1) 706 (42.9) 2 (0.1) 42 (1.1) 100 (2.6) 19 (0.5) 969 (25.8) 0 (0.0) 5( 2, Pvalue

Characteristic

Contraceptive method used (or using) Female sterilization (tubal ligation) Implant (Norplant) Injection Pill Morning after pill (ECP) IUD/coil Withdrawal Rhythm, calendar Condom Foaming tablets/cream/jelly Other
Were you in or out of school, when you began using method of contraception

25.22, 0.009

In Out Both Gynaecological examination No Yes Pap Smear No Yes

33 (41.8) 32 (40.5) 14 (0.8) 420 (61.9) 258 (38.1) 1474 (81.4) 337 (18.6)

4 (6.8) 46 (78.0) 9 (15.3) 4.57, 0.033 1272 (57.3) 947 (42.7) 22.73, < 0.0001 3423 (75.8) 1090 (24.2)

365

Table 13.4: Particular demographic variables by subjective social class of respondents controlled for by age at first coital activity Subjective social class Subjective social class 2, Pvalue Lower Middle Upper Lower Middle Upper Characteristic n = 603 n = 839 n = 369 % % % % % % Area of residence 71.72, 0.0001 Urban 9.3 15.6 21.1 8.6 19.0 21.4 Semiurban 20.9 25.1 36.0 18.6 28.7 39.6 Rural 69.8 59.2 42.8 72.6 52.3 38.9 Educational level 78.72, 0.0001 Primary or below 20.0 11.5 9.5 15.0 8.0 4.6 Secondary 52.7 45.4 36.0 39.3 32.9 19.7 Tertiary 29.2 43.1 54.5 45.7 59.1 75.8 Partner main source of financial 0.559, 0.756 support No 39.5 44.6 45.2 42.6 42.2 42.0 Yes 60.5 55.4 54.8 57.4 57.8 58.0 Employed 20.39, 0.0001 No 69.8 63.4 55.6 70.0 55.4 37.5 Yes 30.2 36.6 44.4 30.0 44.6 62.5 In sexual union 1.81, 0.405 No 18.7 16.6 15.7 21.4 20.4 21.5 Yes 81.3 83.4 84.3 78.6 79.6 78.5 Currently pregnant 1.18, 0.555 No 94.7 95.3 96.2 94.4 96.3 95.8 Yes 5.3 4.7 3.8 5.6 3.7 4.2 Forced to have sex (in life) 1.20, 0.549 No 72.6 72.6 75.4 77.9 82.5 83.7 Yes 27.4 27.4 24.6 22.1 17.5 16.3 1 Crowding mean (SD) 1.9 1.8 1.4 < 0.0001 1.9 1.8 1.4 1 2 F-statistic = 209.22, P<0.0001; F-statistic = 537.28, P<0.0001 366

2, Pvalue 234.20, 0.0001

248.36, 0.0001

0.006, 0.997

228.21, 0.0001

0.647, 0.723 4.36, 0.113 11.58, 0.003 < 0.00012

Table 13.5: Logistic regression analyses: Explanatory variables of use of contraception by age at first coital activity (ie. < 16 or 16+ years old)

Age at first coital activity ( < 16 years old)1

Age at first coital activity ( 16 years old)2

Dependent variable: Method of contraception Age of respondents Upper class Lower class (reference group) In sexual union (1=yes) Currently pregnant (1=yes) Rural Urban (references) Constant
1

coefficient -0.02 -0.33 1.63 -5.51 0.08

Std error 0.01 0.14 0.14 1.01 0.22

Wald Lower 7.90 6.05 135.10 29.82 0.11

Odds ratio 0.98 0.72 1.0 5.09 0.01 1.08

CI (95%) 0.97 -1.00 0.55 0.94 3.87 6.69 0.00 0.03 -

coefficient -0.03 2.24 -4.72 0.23 -0.20

Std error 0.01 0.10 0.46 0.08 0.18

Wald Lower 32.07 533.33 105.74 7.93 1.44

Odds ratio 0.97 9.37 0.03 1.26 1.00 0.80

CI (95%) 0.96 - 0.98 7.75 - 11.38 0.00 - 0.02 1.07 - 1.47 -

Model chi-square = 320.74, P<0.0001 -2 Log likelihood = 1909.11 Nagelkerke r-squared = 0.234 n = 1728 Hosmer and Lemeshow test, 2 = 8.22, P = 0.412; Overall correct classification = 74.1% Correct classification of cases in sexual union = 90.7% Correct classification of cases not in sexual union = 42.4%
2

Model chi-square = 951.90, P<0.0001 -2 Log likelihood = 3737.40 Nagelkerke r-squared = 0.319 n = 3588 Hosmer and Lemeshow test, 2 = 7.95, P = 0.439; Overall correct classification = 77.3% Correct classification of cases in sexual union = 91.4% Correct classification of cases not in sexual union = 52.2%

367

14
Reproductive health matters: Women whose first sexual intercourse occurred at 20+ years old
The evidence is in that public health interventions have failed to effectively address HIV/AIDS infections and lowering the age at first sexual intercourse in the developing nations, particularly in Jamaica despite the amount received and spent on those programmes. The new way of addressing the issues identified earlier is to examine those issues from the perspectives of those who wait until 20+ years old to have sexual intercourse. This study seeks to elucidate information on the reproductive health matters of those who whose first sexual engagement starts at 20+ years old. The current paper found that 9 in every 100 women aged 15-49 years commenced having sexual intercourse at least 20 years. Of those whose sexual relations begin at 20+ years old, 2 out of every 5 are married; 13 out of every 25 are frequent church attendees (at least once per week); 4 out of every 5 have never had a non-steady sexual partner; 14 out of every 25 were in the upper class; 1 out of every 10 shared sanitary convenience; and they began using contraceptives on average at 24 years old. Frequent church attendees on average start having sexual intercourse at 22.7 years, which is 1.2 years later than those who infrequently visit church. The new paradigm is on education and creating economic dependency and not first on safe sex, abstinence and/or on consistency condom usage among young women.

Introduction
Health and reproductive health literature is filled with studies that have examined age at first sexual intercourse (or sexual relations, coitus, sexual debut or sexual initiation) [1-5], and rightfully so because of its association in explaining HIV/AIDS infection, unwanted pregnancies, teenage pregnancy, sexual promiscuity, sexual and reproductive health matters, and general health status [6-8]. In Jamaica, statistics showed that the median age of first coitus among
368

females was 16.0 years in 2001, which fell from 17.3 years in 1997 [9]. Early sexual relation is an adolescent phenomenon, and it is falling more during the adolescence years in Jamaica. First sexual intercourse during the adolescence years is not atypical to Jamaica as this is equally the case in Antigua and Barbuda, Haiti, Guyana, Trinidad and Tobago, Dominica Republic [10], America [11], China [12] and many other developing countries, particularly in Africa [13,14] as well as in the United States of America [3,15]. The prevalence of sexually transmitted infections (STIs) has been on the rise over the decades in Jamaica [16], China [12, 17] and the wider developing nations [18]. Given that most of these occur in individuals aged 15-44 years, particularly 50 percent among people less than 25 years old [19], this is undoubtedly a public health concerns in many respects. In 2007, 1 in 4 Jamaicans were under 25 years old [20]; of those females aged 15-24 years old, only 24.6% reporting never having sex and 67.8% had at least one sexual partner. Previous studies have examined reproductive health matters of adolescents and age at sexual debut [1-5, 10, 19], but there is none which investigated the reproductive health matters of those who commenced sexual intercourse at least 20 years old. For decades, the developing nations have been suffering from increased HIV/AIDS infections, teenage pregnancies, unwanted pregnancies, and lowering of the age at sexual intercourse, yet plethora of studies which have been conducted have not resulted in a fundamental change in the public health problems previously identified. The answers to changing those issues are not beyond us, it is just that a new avenue should be taken in understanding the phenomena. Clearly, the evidence is in that public health interventions have failed to effectively address HIV/AIDS infections and lowering the age at first sexual intercourse in the developing nations, particularly in Jamaica despite the amount received and spent on intervention
369

programmes. This study emerged out of the wanting to provide answers to public health practitioners to change the old approach in viewing a problem that continues to reoccur in the developing nations. The current work will elucidate information on the reproductive health matters of women who delay their first sexual encounter until 20+ years old as this will offer some explanation that can be used to curb the reproductive health practices of those who commenced sexual intercourse during the adolescence years. The present study is therefore a part of larger initiative to change the old approach in examining reproductive health matters of adolescents in wanting to modify (1) age at first sexual intercourse, (2) gynaecological examination; (3) currently used a method of contraception; and (4) the role of church attendance influence sexual behaviour. The new way of addressing the issues identified earlier is examine those issues from the perspectives of those who wait until 20+ years old.

Methods and material


Sample This descriptive cross-sectional study used a secondary dataset from the National Family Planning Board (Reproductive Health Survey, RHS). There are two sets of inclusion criteria, which are females and ages. The eligibility criterion for age was 15 to 49 years at last birthday. Since 1997, the National Family Planning Board (NFPB) has been collecting information on women (ages 15-49 years) in Jamaica regarding contraception usage and/or reproductive health. In 2002, the Reproductive Health Survey (RHS) collected data on Jamaican men ages 15-24 years as well as women 15-49 years old. The current paper extracted 649 females who began having sexual intercourse at 20+ years old. The study population from which the current sample is drawn was 7,168 women of the reproductive ages [9].
370

Stratified random sampling was used to design the sampling frame from which the sample was drawn. Using the 2001 Census sector (or sampling frame), a three-stage sampling design was used. Stage 1 was the use of a selection frame of 659 enumeration areas (or enumeration districts, EDs). This was calculated based on probability proportion to size. Jamaica is classified into four health regions. Region 1 is composed of Kingston, St. Andrew, St. Thomas and St. Catherine; Region 2 comprises Portland, St. Mary and St. Ann; Region 3 is made up of Trelawny, St. James, Hanover and Westmoreland, with Region 4 being St. Elizabeth, Manchester and Clarendon. The 2001 Census showed that Region 1 comprised 46.5% of Jamaica compared to Region 2, at 14.1%; Region 3 at 17.6% and Region 4 at 21.8% [9]. In stage 2, the households were clustered into primary sampling units (PSUs), and each PSU constituted an ED, which in turn was comprised of 80 households. The previous sampling frame was in need of updating, and so this was performed between January and May 2002. The previous sampling frame was in need of updating, and so this was carried out between January 2002 and May 2002. The new sampling frame formed the basis upon which the sampling size was computed for the interviewers to use. Stage 3 was the final selection of one eligible female from each sampled household and this was done by the interviewer on visiting the household. The Statistical Institute of Jamaica (STATIN) provided the interviewers and supervisors, who were trained by McFarlane Consultancy, to carry out the survey. The instrument administered was a 35-page questionnaire. The data collection began on Saturday, October 26, 2002 and was completed on May 9, 2003. Prior to the date of the final data collection, pre-testing of the instrument was conducted between March 16 and 20, 2002. A total of 175 instruments
371

were pre-tested, of which 40.6% were given to eligible men. Modifications were made to the pretested instrument (questionnaire), after which the final exercise was carried out. Validity and reliability of the data were conducted by many statisticians, statistical agency, and university scholars before the data was used as the data are for national policy planning. After which it was released to the University of the West Indies, Mona, Data Bank for use by scholars. The data was weighted in order to represent the population of female aged 15 to 49 years in the nation [9]. Statistical analyses Data were entered, stored and retrieved using SPSS for Window, Version 16.0 SPSS Inc; Chicago, IL, USA). Descriptive statistics were performed on particular sociodemographic characteristics of the sample (frequency, mean, standard deviation (SD), and range). All metric variables were tested for normality (age at first sexual debut, crowding, age, and years of schooling). Where skewness was found to be less than 0.5, the variable was used in its current form and a value more than 0.5 was normalized by natural log, or another method. Independent sample t-test was used to examine differences in age at sexual debut between those who frequently attend churches and those who infrequently visit churches and F-statistic for age of respondents by age at sexual debut. Finally, ordinary least square (OLS) regression was used to fit the data because the dependent variable (age at sexual debut) was a continuous one. Stepwise multiple linear regression was used to fit the one outcome measure (age at first sexual debut) by different sociodemographic variables. Thus, only explanatory variables (i.e. statistically significant variables) are shown in Table 14.3. Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final model construction. To derive accurate tests of statistical significance, we used SUDDAN
372

statistical software (Research Triangle Institute, Research Triangle Park, NC), and this adjusted for the surveys complex sampling design. A p-value < 0.05 (two-tailed) was used to establish statistical significance. Measures Age at first sexual debut (or initiation or intercourse) was measured based on a respondents answer to the question At what age did you have your first intercourse? Crowding is the total number of persons in a dwelling (excluding kitchen, bathroom and verandah). Age is the number of years a person is alive up to his/her last birthday (in years). Contraceptive method comes from the question Are you and your partner currently using a method of contraception? , and if the answer is yes Which method of contraception do you use? Age at which began using contraception was taken from How old were you when you first used contraception? Area of residence is measured from In which area do you reside? The options were rural, semi-urban and urban (1 = rural, 0 = otherwise; 1 = semiurban, 0 = otherwise, and urban is the reference group). Currently having sex is measured from Have you had sexual intercourse in the last 30 days? (1=yes, 0 = otherwise). Education is measured from the question How many years did you attend school? Marital status is measured from the following question Are you legally married now?, Are you living with a common-law partner now? (that is, are you living as man and wife now with a partner to whom you are not legally married?), Do you have a visiting partner, that is, a more or less steady partner with whom you have sexual relations?, and Are you currently single? Age at menarche is measured from How old were you when your first period started (first started menstruation)? Gynaecological examination is taken from Have you ever had a gynaecological examination? (1 = yes, 0 = no). Pregnancy was assessed by Are
373

you pregnant now? (1=yes, 0 = otherwise or no). Religiosity was evaluated from the question With what frequency do you attend religious services? The options range from at least once per week to only on special occasions (such as weddings, funerals, christenings et cetera) (1=frequent attendance from response of at least once per week, 0 = otherwise). Subjective social class is measured from In which class do you belong? The options are lower, middle or upper social hierarchy (1 = middle class, 0 = otherwise; 1 = upper class, 0 = otherwise; reference group is lower class). Forced to have sexual relations was assessed from the question Were you forced to have sex at your first intercourse? and the options were yes, no, dont know and refused to answer (1= yes, 0 = otherwise). Age at first sexual debut, age at menarche, age at first contraceptive use, and years of schooling were used as continuous variables. In stable union measured (1) being legally married or (2) in a common-law union (are you living with a common-law partner now that is, are you living as man and wife now with partner to whom you are not legally married).

Results
Demographic characteristics of study population Table 14.1 summarizes the demographic characteristics of the studied population. Furthermore, 59.4% of the population currently use a method of contraception, and the mean age at sexual debut was 22.1 years (SD = 2.8 years). Almost 89% of the sample had their first sexual encounter before 26 years old, 57.3% at least 21 years and 2.3% at least 30 years old. None of the sample had their first sexual experience after 36 years old. Marginally more of the sample indicated currently using a method of contraception (59.4%). The methods were pill (29.9%); condom (29.3%); female sterilization (22.2%);
374

injection (9.0%); and other. Of those who used a condom, with a steady partner, 42.7% remarked always, 51.7% mentioned most times and 5.6% said seldom (Table 14.2). Bivariate analyses Of the sample, on average women in the lower social class had 2.9 children (SD = 2.3) compared to 2.1 children for those in the middle class (SD = 1.8) and 1.5 children for those in the upper class (SD = 1.4) (F statistic = 319.4, P < 0.0001). Those with primary or below education had 3.3 children (SD = 2.9) compared to 2.4 children (SD = 2.1) among those with secondary level education and 1.7 children (1.6) among those with tertiary level education (F statistic = 185.9, P < 0.0001). There exists a statistical association between educational levels and subjective social class (2 = 507.48, P < 0.0001). Seventy percentages of those in the upper class had tertiary level education compared to 52.6% of those in the middle class and 37.2% of those in the lower class. Table 14.3 presents information on marital status, employment status, raped, currently using a method of contraception, shared sanitary convenience, subjective social class, area of residence, age at sexual debut, age of respondent, years of schooling, age of menarche, age began using method of contraception, crowding and age at marriage by frequency of church attendance. Multivariate analyses Age at sexual debut can be explained by 6 explanatory variables (F statistic = 38.05, P < 0.0001, R2 = 0.376, Table 14.4). These are age, frequent church attendance, number of live births, gynaecological examination done in the last 12 months, education and in a stable union. Age at marriage can be explained by age of respondent, area of residence, age at sexual debut and crowding (F statistic = 6.422, P < 0.0001, R2 = 0.075, Table 14.5).

375

Three explanatory variables account for 29.1% of the variance in currently using a method of contraception: In a stable union; subjective social class and age at sexual debut (Model chi-square = 98.48, P < 0.021; -2 log likelihood = 447.86). Almost 75% of the data were correctly classified (Table 14.6). Six variables emerged as statistically significant correlates of a women having had a gynaecological examination in the past 12 months (Model chi-square = 160.28, P < 0.0001, -2 Log likelihood = 611.18). The factors (area of residence; subjective social class; employment status; age of respondent, education and Pap smears in the last 12 months) account for 32.3% of the variance in gynaecological examination in the last 12 months. Seventy-three percentage of the data were correctly classified (Table 14.7).

Discussion
The current paper found that 9 in every 100 women aged 15-49 years commenced having sexual intercourse at least 20 years. Of those whose sexual relations begin at 20+ years old, 2 out of every 5 are married; 13 out of every 25 years are frequent church attendee (at least once per week); 4 out of every 5 have never had a non-steady sexual partner; 14 out of every 25 were in the upper class; 1 out of every 10 shared sanitary convenience; and they began using contraceptives on average at 24 years. Among the factors that positively influences age at first sexual intercourse are frequency in church attendance, age, educational attainment and in a stable union. According to Bourne and Charles [21], church attendance is among the factors which account for young mens (aged 15-24 years) lowered age at first sexual intercourse. This research concurs with Bourne and Charles study that frequent church attendance is responsible for increased age at sexual debut among those who began having sexual intercourse at least 20
376

years. In this work, it was revealed that 13 out of every 25 women who delay sexual intercourse for 20+ years attend church on a regular basis. It was also found that among the studied population, those who are frequent church attendees on average commenced their sexual encounter age 22.7 years which is 1.2 years later than those who are infrequent attendees. One study which explored sexual initiation of persons within the age range of 15-44 years, found that protestants (similar to those of non-religion) were more likely to have their first sexual initiation within their 16th year, compared to the Catholics (within their 17th year) and those of other religion (18th year) [22]. It can be concluded that the cultural values and orientation of the churches that occasional attendance do not change women delaying sexual debut, but that frequent attendance is one of the media that increase delaying age at sexual relations. Another justification which account for delaying age at sexual debut among women is a stable sexual union. Embedded in this finding is fact that women who starts in stable unions such as marriage or common-law sexual unions are least likely to search for such a union as in the case of women who are in visiting relationship. Furthermore, the cultural values of the church is such that frequent membership is more fostered in marriage and therefore accounts for why 29 out of every 50 women who frequently attend church are married compared to 10 out of every 50 of those who infrequently visit churches. This is also embedded in the current finding which showed that 4 out of every 5 women who delay having sexual intercourse at 20+ years old indicated that they have never had a non-steady partner, suggesting that visiting union are more about sex than stable union. Thus, when Wilks et al. [16] found that 8 out every 25 women aged 15-24 years had never had a sexual partner (or in the last 12 months), it follows that these females are seeking for stable than transitional sexual union that are likely to result in another sexual relationship.
377

Another issue which emerged from current work about the studied population is subjective social class. It was found that 14 out of every 25 women who begin having sexual relationship at 20+ years old are in the upper class, and that 3out of every 25 are in the lower class. One scholar postulated that money is positively associated with health [23], but is appears that economic disparity accounts for the delaying of sexual intercourse, advanced educational attainment and number of live births. The current research found that women who delay sexual intercourse for 20+ years old is money, which fosters accessing higher level of education, less children, and have a greater sexual autonomy of their live than those in the lower class. Like Marmot, money matters for women health as well as their reproductive health and the age at which they begin having sexual intercourse. Poverty, economic deprivation and the social settings among those who are poor is account for the early sexual relation. Those in the sample have higher educational attainment are among the upper class. This work shows that educational level is positively associated with increased age at first sexual intercourse, suggesting that poverty increases people search for social relationship as a source of material goods. Those individuals have less sexual autonomy, and sexual intercourse is left up to the male who wants this earlier than later and not for the purpose of desiring a stable union. Because those sexual unions are mainly visiting and/or transitional partnerships, knowing the vulnerability of some females, males will dictate condom usage as a price for economic support. The results of inconsistent condom usage are STIs, unwanted pregnancies, and females being caught in the cycle of more such relationship for financial support because the last one offered less and they desire continuous assistance owing to their economic deprived status. Thus, this explains the negative association between number of live births and age at sexual debut, meaning that as more economic independency or family with economic resources have more sexual and
378

reproductive rights which justifies them delaying sexual intercourse for a later time (20+ years). If Yan et als postulations holds true in developing nations outside of China that Safe sexual behaviors include having a single sex partner and using condoms in every sexual encounter, and these behaviors also reduce risk of HIV/STDs. [12, p. 2], then women in instable unions will continue to inconsistently use condoms as they are economically dependent on a male partner for some level of survivability. In addition to the aforementioned, thus, this supports the finding that stable unions influence age at sexual intercourse and that money matters in reproductive health matters as well as sexual autonomy. This is supported by the a World Health Organizations (WHO) postulation that stated, In high-income countries, communicable diseases account for only 8% of years of life lost, compared with 68% in low-income countries [24, p. 47], suggesting economic deprivation is retarding the real accomplishments which could have materialized in public health interventions if there were income equality among people and within countries. The Pan American Health Organization (PAHO) offered another angle to the discussion that culturally there is greater tolerance for premarital sexual relationships among those in the lower socioeconomic strata [25], which provides yet another argument for money in addressing health, reproductive health matters and sexual behaviour. PAHO [25] noted that chronic poverty enforced promiscuity because of overcrowding, sharing sanitary convenience and living accommodations with other households, and less marriage occurred in this cohort compared with those in the middle socioeconomic strata and wealthy socioeconomic class. This paper revealed a positive statistical association between age at marriage and age at sexual debut, suggesting that women who delay sexual intercourse to 20+ years old are older meaning that they are more economic independent, career oriented, and are able to way for stable
379

union in which they have an equal state in the union. The economic dependency of females is accounting for early marriage among rural women as they seek the assistance for the male partner. Clearly the age at which a female commenced sexual intercourse may hold some explanation for her choice in currently using a method of contraception. It was found that women who begin sexual relations at 20+ years old are 8% less likely to currently use a method of contraception as they seek family because of the stability of their unions. This is reinforced by Yan et al [12] who stated that Adolescents typically engage in short-lived relationships that make them more likely than adults to have sex with multiple partners, thereby placing them at greater risk for contracting HIV/STDs [12, p. 2] and that attitudes to sex have an enormous influence on sexual behavior [12, p. 9]. However, embedded in Yan et al.s work is the fact that early sexual partnerships are more likely to be visiting, less knowledgeable and/or educated females, more economically dependent young people, and so their males partners are more able to veto their reproductive rights and/or choices. The reality here coupled with the current findings is that young females in economically vulnerable homes will not delay sexual intercourse as this provides survivability that cannot be found by the family unlike those in wealthy households. Money is not only influence risky sexual behaviour; it is also guiding health choices such as Papanicolaou (Pap) smear examination. Wilks et al. [16] found that 18% of women aged 15-74 years old in Jamaica had done a Pap smear in the last 12 months, but in this study it was 32%. The high cervical screening among the studied population has something to do with their educational awareness, income, and understanding of their bodies more so than the general female population. This is reinforced in a World Health Organizations publication [26] which stated that All women who have had
380

sexual intercourse are potentially at risk because they might have been infected with HPV [human papillomavirus]. It should be noted here that all the women are currently sexually active, and because of their awareness of cervical issues and their sexual practices account for the substantially greater percentage having done a Pap smear or gynaecological examination in the last 12 months. It was found that women in the upper class were almost 2 times more likely to had a gynaecological examination, those with tertiary level education were 1.2 times more likely to have this test, and being employed was another critical factors that fosters having the test done.

Conclusion
Females delaying their first sexual intercourse to 20+ years in Jamaica can provide a comprehensive insight into increasing age at sexual debut and a guide to public health practitioners in the way forwarding for intervention programmes. The current paper highlights that education, income (or social class ie. upper class), stable union, are frequent church attendance critical for increasing age at first sexual intercourse. Money does matter in delaying the time at which women become engaged in sexual relations, and this cannot be ignored in any intervention programme. Money increases women economic independency and it also does educational advancement, which must be included in public health intervention programmes that must be a part of the solutions in the way forward. In summary, public health programmes which are not geared towards economic independency and educational advancement will be futile. The way forward should not be about abstinence (or say no to sex) or consistency condom usage as the current paper should a road map on issues that account for why some women delay sexual intercourse. The thrust of any new
381

intervention programme that is geared towards changing sexual behaviour of Jamaican women should be on education and economic independency as those are the key tenants and not the current approach. The new paradigm is on education and creating economic independence and not first on safe sex, abstinence and/or on consistency condom usage or increasing knowledge about a reproductive or HIV/AIDS among young women.

Disclosures The author report not conflict of interest with this work.

Disclaimer
The researcher would like to note that while this study used secondary data from the Reproductive Health Survey, none of the errors in this paper should be ascribed to the National Family Planning Board, but to the researcher.

Acknowledgement
The author thanks the Data Bank in Sir Arthur Lewis Institute of Social and Economic Studies, the University of the West Indies, Mona, Jamaica for making the dataset (2002 Reproductive Health Survey, RHS) available for use in this study, and the National Family Planning Board for commissioning the survey.

382

References
1. Slaymaker, E, Bwanika, J B, Kasamba, I, Lutalo, T, Maher, D, Todd, J (2009). Trends in age at first sex in Uganda: evidence from Demographic and Health Survey data and longitudinal cohorts in Masaka and Rakai. Sex. Transm. Infect. 85: i12-i19 2. Fatusi AO, and Blum, RW. Predictors of early sexual initiation among a nationally representative sample of Nigerian adolescents. BMC Public Health; 8, 2008. 3. Santelli JS, Kaiser J, Hirsch L, Radosh A, Simkin L, Middlestadt S: Initiation of sexual intercourse among middle school adolescents: the influence of psychosocial factors. J Adolesc Health 2004, 34:200-208. 4. Penfold SC, Teijlingen ERV, Tucker JS. Factors associated with self-reported first sexual intercourse in Scottish adolescents. BMC Research Notes 2009; 2:42. 5. Rosenthal DA, Smith AMA, De Visser R. Personal and social factors influencing age at first sexual intercourse. Archives of Social Behavior 1999; 28(4):319-333. 6. Miller KS, Clarke LF, Moore JS. Sexual initiation with older male partners and subsequent HIV risk behavior among female adolescents. Family Planning Perspectives; 29, 1997:212-214. 7. Louie KS, de Sanjose S, Diaz M, et al. Early age at first sexual intercourse and early pregnancy are risk factors for cervical cancer in developing countries. Br J Cancer 2009; 100(7):119-7. 8. Mardh PA, Creatsas G, Guaschino S, et al. Correlation between early sexual debut, and reproductive health and behavioral factors: a multinational European study. Eur J Contracept Reprod Health Care 2000; 5:177-82. 9. Jamaica National Family Planning Board (NFPB). Reproductive Health Survey, 2002. Kingston: NFPB; 2005. 10. Rawlins J. Teenage Pregnancy: A study in three communities in Trinidad and Tobago. Paper presented at the Caribbean Health Research Conference 2007, Jamaica. 11. Sexual Initiation. American Sexual Behaviour. http://www.newstrategist.com/productdetails/Sex.SamplePgs.pdf (Accessed on April 1, 2010). 12. Yan H, Chen W, Wu H, et al. Multiple sex partner behavior in female undergraduate students in China: A multi-campus survey. BMC Public Health 2009; 9:305. 13. McGrath N, Nyirenda M, Hosegood V, et al. Age at first sex in rural South Africa. Sex Transm Infect 2009; 85:i49-i55. 14. Cremin I, Mushati P, Hallett T, et al. Measuring trends in age at first sex and age at marriage in Manicaland, Zimbabwe. Sex Transm Infect 2009; 85(Suppl 1):i34-i40. 15. Rotermann M. Sex, condoms and STDs among young people. Health Reports 2005;16:339-42. 16. Wilks R, Younger N, Tulloch-Reid M, McFarlane S, Francis D. Jamaica health and lifestyle survey 2007-8. Kingston: Tropical Medicine Research Institute, University of the West Indies, Mona; 2008. 17. State Council AIDS Working Committee Office and UN Theme Group on HIV/AIDS in China: A joint assessment of HIV/AIDS prevention, treatment and care in China (2005). Beijing 2005.
383

18. Douglas DL. Perspectives on HIV/AIDS in the Caribbean. In: Morgan O. ed. Health issues in the Caribbean. Kingston: Ian Randle Publisher; 2005: pp. xv. 19. Frederick J, Hamilton P, Jackson J, et al. Issues affecting reproductive health in the Caribbean. In: Morgan O. ed. Health issues in the Caribbean. Kingston: Ian Randle Publisher; 2005: pp. 41-50. 20. Statistical Institute of Jamaica (STATIN). Demographic statistics, 2007. Kingston: STATIN; 2008. 21. Bourne PA, Charles CAD. Contraception usage among young adult men in a developing country. Open Access J of Contraception 2010; 1:51-59. 22. Sexual Initiation. American Sexual Behaviour. http://www.newstrategist.com/productdetails/Sex.SamplePgs.pdf (Accessed on April 1, 2010). 23. Marmot M: The influence of Income on Health: Views of an Epidemiologist. Does money really matter? Or is it a marker for something else? Health Affairs. 2002; 21: 3146. 24. World Health Organization (WHO). World health statistics, 2009. Geneva: WHO; 2009. 25. Pan American Health Organization (PAHO). Health of women in the Americas. Washington D.C.: PAHO; 1985. 26. World Health Organization (WHO). Comprehensive cervical cancer control: A guide to essential practice. Geneva: WHO; 2006

384

Table 14.1: Demographic characteristic of study population, n= 649 Characteristic Shared sanitary convenience No Yes Want to be pregnant No Yes Employment status Unemployed (including sick, students, etc.) Employed Pap smear (in last 12 months) No Yes Raped No Yes Had sex (in last 30 days) No Yes Marital status Married Common-law Visiting Previously in union (divorced, separated, widowed) Single Social class Lower Middle Upper Age cohort (in years) 20-24 25-29 30-34 35-39 40-44 45-49 Educational level Primary or below Secondary Tertiary Area of residence Urban Semiurban Rural Frequent church attendance No Yes 385

n 578 70 611 38 244 405 437 212 561 86 0 649 261 83 140 153 12 82 201 366 41 109 149 149 124 77 32 127 482 145 223 281 307 342

% 89.2 10.8 94.1 5.9 37.6 62.4 67.3 32.7 86.7 13.3 0.0 100.0 40.2 12.8 21.6 23.6 1.8 12.6 31.0 56.4 6.3 16.8 23.0 23.0 19.1 11.9 4.9 19.6 74.3 22.3 34.4 43.3 47.3 52.3

Table 14.2: Particular reproductive health matters Characteristic Currently using a method of contraception No Yes First method of contraception Condom Other (modern methods) Withdrawal, Rhythm or calendar Did not respond Current method of contraception Female sterilization Implant Injection Pill Emergency contraception IUD/Coil Diaphragm Withdrawal Rhythm, calendar Condom Frequent condom usage (with steady partner) Always Most times Seldom Never Frequent condom usage (with non-steady partner) Always Most times Seldom Never Never had a non-steady partner Age began using method of contraception, mean (SD)

n 249 365 22 4 2 621 81 1 33 109 1 10 2 14 7 107 61 74 8 0 14 6 0 7 114

% 40.6 59.4 3.4 0.6 0.3 95.7 22.2 0.3 9.0 29.9 0.3 2.7 0.5 3.8 1.9 29.3 42.7 51.7 5.6 0.0 9.9 4.3 0.0 5.0 80.9 23.9 years (4.0 yrs)

386

Table 14.3: Particular demographic characteristics of sample by church attendance, n = 649 Frequent church attendance 2, Pvalue Characteristic Infrequent Frequent n (%) n (%) Marital status 141.0, < 0.0001 Married 63 (20.5) 198 (57.8) Common-law 69 (22.5) 14 (4.1) Visiting 104 (33.9) 36 (10.5) Separated, widowed, divorced 66 (21.5) 87 (25.4) Single 5 (1.6) 7 (2.0) Employment status 0.053, 0.818 Unemployed (including student, sick, etc) 114 (37.1) 130 (38.0) Employed 193 (62.9) 212 (62.0) Raped 3.168, 0.075 No 273 (89.2) 288 (84.5) Yes 33 (10.8) 53 (15.5) Currently using method of contraception 1.26, 0.262 No 112 (38.2) 137 (42.7) Yes 181 (61.8) 184 (57.4) Shared sanitary convenience 12.496, < 0.0001 No 259 (84.6) 319 (93.3) Yes 47 (15.4) 23 (6.7) Subjective social class 1.617, 0.445 Lower 44 (4.3) 38 (11.1) Middle 95 (30.9) 106 (31.0) Upper 168 (54.7) 198 (57.9) Area of residence 17.487, <0.0001 Urban 90 (29.3) 55 (16.1) Semiurban 102 (33.2) 121 (35.4) Rural 115 (37.5) 166 (48.5) t-test, Pvalue Age at sexual debut 21.5 yrs (2.3) 22.7y yrs (3.2) -5.582, < 0.0001 Age 34.8 yrs (7.0) 36.0 yrs (7.1) -2.184, 0.029 Years of schooling 14.4 yrs (3.2) 14.5 yrs (3.6) -0.140, 0.889 Age of menarche 13.6 yrs (4.1) 13.6 yrs (3.8) -0.214, 0.831 Age began using method of contraception 23.6 yrs (4.0) 24.2 yrs (4.1) -1.777. 0.076 Crowding 1.5 persons (0.5) 1.5 persons (0.5) -0.165, 0.869 Age at marriage 32.9 yrs (26.1) 33.4 yrs (24.9) -0.239, 0.812 Number of live births mean (SD) 2.1 children (1.9) 2.0 children (2.0) 0.912, 0.362

387

Table 14.4: Ordinary least square regression: Explanatory variables of age at sexual debut

Dependent variable: Age at sexual debut Constant Age Frequent church attendance (1=yes) Number of Live Births Gynaecological exam (1=yes) Education (in years) In a stable union (1=yes) F statistic = 38.05, P < 0.0001 R2 = 0.376, adjusted R2 = 0.366 N = 612

Unstandardized coefficient 12.94 0.35 0.90 -0.44 -0.72 0.07 0.54

Std. error 0.87 0.03 0.23

Beta

CI (95%) 11.24 - 14.64

0.52 0.17

0.30 - 0.41 0.45 - 1.35 -0.61 - -0.26 -1.17 - -0.28 0.01 - 0.13 0.08 - 1.00

0.09 -0.20 0.23 -0.13 0.03 0.23 0.10 0.10

388

Table 14.5: Ordinary least square regression: Explanatory variables of age at marriage Unstandardized coefficient -14.92 0.64 -7.56 -6.10 Std error 11.85 0.18 0.18

Dependent variable: Age at marriage Constant Age In stable union (1=yes) Rural Urban (reference group) Age at sexual debut Crowding F statistic = 6.422, P < 0.0001 R2 = 0.075, adjusted R2 = 0.063 N = 612

Beta

CI (95%) -38.21 - 8.38 0.30 - 0.99 -12.32 - -2.80 -11.05 - -1.16

2.42 -0.15 2.52 -0.12

1.05 4.82

0.42 2.39

0.12 0.10

0.22 - 1.88 0.11 - 9.52

389

Table 14.6: Logistic regression analyses: Explanatory variables of currently using a method of contraception Dependent variable: Currently using a method of contraception In a stable union (1=yes) Middle class Lower class (reference group) Age at sexual debut Constant coefficient 2.48 -0.64 -0.09 0.61 Std. error 0.30 0.26 0.04 0.90 Odds ratio 11.95 0.53 1.00 0.92 1.84

CI (95%) 6.68 - 21.37 0.32 - 0.87 0.85 - 0.99

Model chi-square = 98.48, P < 0.021 -2 Log likelihood = 447.86 Nagelkerke r-squared = 0.291 Hosmer and Lemeshow test, 2 = 1.734, P = 0.973 Overall correct classification = 74.5% Correct classification of cases of currently using a method of contraception = 92.3% Correct classification of cases of not currently using a method of contraception = 47.8%

390

Table 14.7: Logistic regression analyses: Explanatory variables of those who had done a gynaecological examination in the last 12 months, n=603

Dependent variable: Gynaecological exam Rural Urban Upper class Lower class (reference group) Employment (1= employed) Age of respondent Years of schooling Pap smear in last 12 months Constant

coefficient -0.76 0.68 0.45 0.06 0.15 1.363 -4.20

Std. error 0.21 0.21 0.21 0.02 0.04 0.25 0.87

Odds ratio 0.48 1.00 1.99 1.00 1.56 1.06 1.16 3.91 0.02

CI (95%) 0.31 - 0.70 1.31 - 3.00 1.04 - 2.36 1.03 - 1.10 1.07 - 1.26 2.38 6.42

Model chi-square = 160.28, P < 0.0001 -2 Log likelihood = 611.18 Nagelkerke r-squared = 0.323 Hosmer and Lemeshow test, 2 = 5.95, P = 0.653 Overall correct classification = 74.3% Correct classification of cases of currently using a method of contraception = 85.1% Correct classification of cases of not currently using a method of contraception = 52.5%

391

15
On sexual and non-intimate unions among the general reproductive population of women in Jamaica: A cross-sectional survey

This study evaluates the demographic, reproductive and health characteristics of women aged 15-49 years old who are currently in a sexual union or not in intimate unions in Jamaica and difference in area of residence as well as factors which determine (1) good-to-very good health status; (2) method of contraception; (3) had sex in the last 3 months; (4) high church attendance; and (5) physically forced to have sexual intercourse, with emphasis on those who are in a sexual union or not in Jamaica. A secondary analysis of 8,259 women in the reproductive ages obtained from a 2008 Reproductive Health Survey. Multivariate analyses were used to established factors which account for particular dependent variables (health status; method of contraception; had sex in the last 3 months; high church attendance; physically forced to have sexual intercourse and currently in an intimate union). Eighteen out of every 25 women of the reproductive ages reported that they are currently in a sexual union. Those in a sexual union were older (OR = 1.03, 95% CI = 1.01 1.04), more likely to have had sexual relations in the last 3 months (OR = 29.85, 95%CI = 23.27 38.20), using a method of contraception on the last sexual activity (OR = 1.76, 95%CI = 1.37 2.26), more likely to dwell in a crowded household (OR = 1.29, 95% CI = 1.02 1.63), did a HIV/AIDS test (OR = 1.6, 95% CI = 1.29 1.99), and less likely to be educated (OR = 0.94, 95%CI = 0.88 0.99). Many differences emerged between women aged 15 to 49 years old who are currently in a sexual union compared with those who are not. Those differences can be used to guide policy intervention, and shape future research. Biological sex is fundamentally a part of cohabitation among women aged 15 to 49 years who indicated being currently in an intimate union, suggesting that sexual unions are highly based on sexual intercourse for gains (emotional and material) and that this can be used to guide policy formulation and intervention programmes.

392

Introduction
Some people are inclined to assume that reproductive health matters are overly studied. But these matters affect many areas of the individuals life and/or that of a nation such as life expectancy, quality of life, health status, population compositions, mortality patterns, production, productivity, economic growth and economic development, and therefore speak to their importance. Furthermore, reproductive health issues persist on a yearly basis and as such justify the continuation of reproductive health research in an attempt to provide incessant answers on issues that affect people as well as the nation. In Caribbean nations, HIV/AIDS, other sexually transmitted infections and unwanted pregnancies are continuously rising as well as the lowering of the age of sexual debut; despite the designed sexual education intervention programmes, developing nations have not effectively address those challenges which persist on a yearly basis and so they cannot be set-aside to other sociomedical problems that may appear atop of the social and medical hierarchy of challenges. For years, Jamaican policy makers have been using different designed sexual education programmes to address various reproductive health matters. Inspite of their efforts, in 1997, the Jamaican National Family Planning Board found that the median age at first sexual intercourse for women was 17.3 years and this fell to 16.0 years in 2002.1 Two in every 5 Jamaican women have been pregnant prior to reaching the age of 20 years, and most of the pregnancies were unplanned, especially during the adolescent years (80%).1 Continuing, in 2002, the mean age of sexual debut in Jamaica was 15.8 for females and 13.5 for males,1 much of which were forced and is seen as a direct link with violence, as well as one of the roots of sexual and reproductive health problems in the international community.2 These matters go against the principles of the

393

ICPD 1994, which stipulates that when it comes to matters of sexual relations, full respect for the integrity of the individuals involved should be of the utmost.2 Empirical evidence showed that the First sexual intercourse almost always take place outside of a formal union3 and with older men (for the females). 4 Those occurrences are likely to result in health situations relating to STIs and HIV, as well as drug abuse.5 Warren et al.,6 the high fertility population in Jamaica was women ages 14-24 years, indicating a high degree of premarital sexual activities and inconsistent condom use within the context of reduced age at first sexual intercourse.7 A study of some sub-Saharan African and South-East Asian nations show similar sexual behaviour and attitude of young people8 whereas one by Henry-Lee9 found that 66% of Jamaican women used method of contraception, but only 34% of pregnancies were planned for. This indicates that inconsistent contraceptive use is accounting for increased HIV/AIDS and STIs in Jamaica that is typical in the wider developing countries, as young adults are engaged in risky sexual practices.10,11 Within the context of the lowered age of sexual debut in the world11-16 and the aforementioned identified reproductive health matters, it follows that reproductive health issues must be continuously studies as they affect many areas in the life of the individual and the nation that do not cease but progress and become problematic for nation building. With research showing the First sexual intercourse almost always take place outside of a formal union,3 STIs on the rise17-19 and that 22 out of every 25 Jamaicans aged 15-24 years have had sexual intercourse as well as 24.1% of Jamaicans aged 15-74 years indicated having at least 2 sexual partners (females, 8.4%; males, 41.0%),20 it follows that an investigation on the reproductive health matters of those in a sexual union versus those who do not is timely. And this would

394

provide invaluable insights into both cohorts, and how policies can be better implemented to address the identified challenges which emerged. Poverty is associated with illness, material deprivation, low educational attainment and other issues. In Jamaica, with rural poverty being at least twice that of urban poverty and poverty being greater among females than males,21 then females who reside in rural areas are more vulnerable to reduced sexual autonomy than their urban counterparts as they rely more on males for financial assistance than their urban or periurban counterparts because of low educational status, poverty, and material deprivation. The World Health Organization (WHO) 22 postulated that 80% of chronic illnesses were in low and middle income countries, indicating that illness interfaces with poverty and other socio-economic challenges. The WHO noted this aptly when it stated that ...People who are already poor are the most likely to suffer financially from chronic diseases, which often deepen poverty and damage long term economic prospects.22 Statistics for revealed that 15.1% of Jamaicans reported having had an illness in a 4-week period, and this was 17.8% for females and 17.2% for rural residents who are more likely to be in poverty compared to their male, and urban or periurban counterparts.21 Clearly, there is a poverty gender disparity as well as a poverty area of residence disproportionality in Jamaica, but no study has sought to elucidate whether there are differences in reproductive health matters among women in sexual union or not, and the area of residence that they dwell. Statistics showed that in 2009, 53.7% of Jamaicans were aged 15-49 years, and 54.7% of females were between the ages of 15 to 49 years old.23 The percentage of population ages 15 to 49 years old, particularly women, is a substantial group which cannot be left unresearched moreso because of the prevalence of HIV/AIDS (8th leading cause of morality among Jamaican females for 2006 and 200723) virus and unwanted pregnancies among this
395

cohort. This age cohort is a critical part of the productive age for employed, indicating its importance (or lack of) to production and development of the nation. Outside of poverty and the aforementioned reproductive health issues, we still do not know about the reproductive health matters of women aged 15-49 years who are in a sexual union versus those in non-sexual unions because no study emerged than elucidate important on these cohorts. The current paper will evaluate the demographic, reproductive and health characteristics of women aged 15-49 years old who are in a sexual union or not in Jamaica and their area of residence as well as factors which determine (1) good-to-very good health status; (2) method of contraception; (3) had sex in the last 3 months; (4) high church attendance; and (5) physically forced to have sexual intercourse, with emphasis on those who are in a sexual union or not in Jamaica.

Methods and materials


Sample The current research used the dataset from a national descriptive cross-sectional survey. The survey was conducted by the National Family Planning Board in 2008 on Jamaican women among the reproductive ages and males aged 15-24 years old (2008 Reproductive Health Survey, RHS). This study extracted only females aged 15-49 years from 2008 Reproductive Health Survey. The study population was 8,259 women. Stratified random sampling was used to design the sampling frame from which the sample was drawn. Using the 2001 Census sector (or sampling frame), a three-stage sampling design was used. Stage 1 was the use of a selection frame of 659 enumeration areas (or enumeration districts, EDs). This was calculated based on probability proportion to size. Jamaica
396

is classified into four health regions. Region 1 is composed of Kingston, St. Andrew, St. Thomas and St. Catherine; Region 2 comprises Portland, St. Mary and St. Ann; Region 3 is made up of Trelawny, St. James, Hanover and Westmoreland, with Region 4 being St. Elizabeth, Manchester and Clarendon. The 2001 Census showed that Region 1 comprised 46.5% of Jamaica compared to Region 2, at 14.1%; Region 3 at 17.6% and Region 4 at 21.8%.1 In stage 2, the households were clustered into primary sampling units (PSUs), and each PSU constituted an ED, which in turn was comprised of 80 households. The previous sampling frame was in need of updating, and so this was performed between January and May 2002. The previous sampling frame was in need of updating, and so this was carried out between January 2002 and May 2002. The new sampling frame formed the basis upon which the sampling size was computed for the interviewers to use. Stage 3 was the final selection of one eligible female from each sampled household and this was done by the interviewer on visiting the household. The Statistical Institute of Jamaica (STATIN) provided the interviewers and supervisors, who were trained by McFarlane Consultancy, to carry out the survey. The instrument administered was a 35-page questionnaire. A total of 175 instruments were pre-tested. Modifications were made to the pre-tested instrument (questionnaire), after which the final exercise was carried out. Validity and reliability of the data were conducted by many statisticians, statistical agency, and university scholars before the data was used as the data are for national policy planning. After which it was released to the University of the West Indies, Mona, Data Bank for use by scholars. The data was weighted in order to represent the population of female aged 15 to 49 years in the nation.1
397

Statistical analyses Data were entered, stored and retrieved using SPSS for Window, Version 17.0 (SPSS Inc; Chicago, IL, USA). Descriptive statistics were performed on particular sociodemographic characteristics of the sample (frequency, mean, standard deviation (SD), and range). All metric variables were tested for normality (age at first sexual debut, crowding, age, and years of schooling). Where skewness was found to be less than 0.5, the variable was used in its current form and a value more than 0.5 was normalized by natural log, or another method. Independent sample t-test was used to examine differences in age at sexual debut between those who frequently attend churches and those who infrequently visit churches and F-statistic was employed for age of respondents by age at sexual debut. Chi-square was used to examine the statistical association between two non-metric variables. Finally, stepwise multiple logistic regression analyses were used to fit the data because the dependent variable is a dichotomous nominal measure. Thus, only explanatory variables (i.e. statistically significant variables) are shown in each Table. Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final model construction. To derive accurate tests of statistical significance, we used SUDDAN statistical software (Research Triangle Institute, Research Triangle Park, NC), and this adjusted for the surveys complex sampling design. A p-value < 0.05 (two-tailed) was used to establish statistical significance. Measures Key variables: Age at first sexual debut (or initiation or intercourse) was measured based on a respondents answer to the question At what age did you have your first intercourse? Crowding is the total number of persons in a dwelling (excluding kitchen, bathroom and verandah). Age is
398

the number of years a person is alive up to his/her last birthday (in years). Contraceptive method comes from the question Are you and your partner currently using a method of contraception? , and if the answer is yes Which method of contraception do you use? Age at which began using contraception was taken from How old were you when you first used contraception? Area of residence is measured from In which area do you reside? The options were rural, semiurban and urban (1 = rural, 0 = otherwise; 1 = semi-urban, 0 = otherwise, and urban is the reference group). Currently having sex is measured from Have you had sexual intercourse in the last 30 days? (1=yes, 0 = otherwise). Education is measured from the question How many years did you attend school? Marital status is measured from the following question Are you legally married now?, Are you living with a common-law partner now? (that is, are you living as man and wife now with a partner to whom you are not legally married?), Do you have a visiting partner, that is, a more or less steady partner with whom you have sexual relations?, and Are you currently single? Age at menarche is measured from How old were you when your first period started (first started menstruation)? Gynaecological examination is taken from Have you ever had a gynaecological examination? (1 = yes, 0 = no). Pregnancy was assessed by Are you pregnant now? (1=yes, 0 = otherwise or no). Religiosity was evaluated from the question With what frequency do you attend religious services? The options range from at least once per week to only on special occasions (such as weddings, funerals, christenings et cetera) (1=frequent attendance from response of at least once per week, 0 = otherwise). Subjective social class is measured from In which class do you belong? The options are lower, middle or upper social hierarchy (1 = middle class, 0 = otherwise; 1 = upper class, 0 = otherwise; reference group is lower class). Forced to have sexual relations was assessed from the question Were you forced to have sex at your first intercourse? and the options were yes, no, dont
399

know and refused to answer (1= yes, 0 = otherwise). Age at first sexual debut, age at menarche, age at first contraceptive use, and years of schooling were used as continuous variables. Early sexual debut is having sexual intercourse before the statutory legal age to do so (in Jamaica, this is 16 years old).

Sexual union denotes the pairing of male and female for the purpose of reproductive matters.

Results
Descriptive statistics: Table 15.1 presents information on the demographic characteristics of sample by currently in or not in a sexual union. The findings revealed that 2.5 times more women aged 15-49 years were in a sexual union (71.2%) compared to those not in a sexual union. Of those in a sexual union 38.8% were in visiting relationships followed by common-law relationships (36.7%) and 24.5% were married women. Women who are in sexual unions enter in these unions 4 months earlier than those who were currently not in a sexual union, but latter began having sexual relations 5.9 years early than those currently in a sexual union. Figure 1 shows that respondents in the poorest 20% were most likely to be in sexual unions (74.5%) compared with those in other income quintiles. However, 62.3% of those in the poorest 20% were in common-law or visiting relationship than married people compared with 46.8% of those in the wealthiest 20% (2 = 347.53, P < 0.0001). Table 15.2 depicts information on the reproductive health matters of the sample by those currently in or not in a sexual union. Table 15.2 showed that women in a sexual union were less likely to report having 2+ sexual partners (6.0%) compared to those who were currently not in a sexual union (10.9%). However, the latter were more likely to use a condom (56.1%) and
400

reported having HIV/AIDS (25.7%) compared to those in a sexual union condom usage (44.1%), HIV/AIDS (14.2%). Table 15.3 exhibits information on the demographic characteristic of sample by area of residence. The findings indicate that women in periurban areas were wealthier (33.5%) than those in urban areas (29.7%) as well as rural locations (17.3%). In addition to the aforementioned issue, rural women enter in sexual unions 6 months earlier than periurban women and 7 months later than those who reside in urban zones (Table 15.3). However, periurban women begin having sexual intercourse the earliest (24.3 23.5 years) compared to those in urban (28.128.5 years) and rural women (25.6 25.9 years). Multivariate analyses: Table 15.4 presents information on logistic regression analyses of explanatory variables for good-to-very good health status of sampled population. Women in periurban areas were 18% less likely to report good-to-very good health compared to rural women. However, no statistical difference emerged in the self-rated health status between those in or not in a sexual union. In addition, women who were physical forced into sexual intercourse were 37% less likely to report good-to-very good health status. Table 15.5 shows the logistic regression of method of contraception. Based on the stepwise regression results eight variables emerged as statistical significant factors of method of contraception, and they explain 34.4% of method of contraception. The findings revealed that women who had sexual intercourse in the last 30 days were 7.7 times more likely to use a method of contraception the last time they had sexual relation, the employed were 1.2 times more likely, those in the wealthiest 20% were 1.4 time more likely, those who reported good-tovery good health status were 1.2 times more likely as well as those in a sexual union (OR = 1.8, 95% CI = 1.4 2.4).
401

Seven variables emerged as statistical significant factors explaining 63% of women aged 15-49 years having had sexual relations in the last 30 days (Table 15.6). Furthermore, women who indicated that they are currently in a sexual union were 73.8 times more likely to have had sex in the last 30 days. Table 15.7 shows explanatory variables of high church attendance. The findings revealed that seven variables emerged as statistical significant factors of high church attendance. Rural women exhibited the greatest church attendance. Furthermore those who reported having had sexual intercourse in the last 30 days were 34% less likely to attend church on a regular basis (at least once per week). Using stepwise logistic regression, nine variables emerged as statistically significant variables accounting for 24% of those who were physically forced into sexual relations (Table 15.8). Women who reside in periurban areas were 1.3 times more likely to be physically forced into sexual relations compared to those in rural areas (OR = 1.3, 95% CI = 1.03 - 1.68). Furthermore, those who were in sexual unions were 38% less likely to be physical forced into sexual relations compared to those who did not indicated being in a sexual union. Table 15.9 shows variables that explain women who are currently in or not in a sexual union. Twelve of the initial variables emerged as statistically significant factors explaining who are currently in or not in an intimate union (R2 = 59.0%).

Discussion
In Jamaica, intimate unions (sexual unions) among women aged 15 to 49 years were 2.5 times more likely than non-intimate unions (in 2008). The sexual unions were (1) marriage (24.5%), common-law (36.7%), and visiting relationships (38.8%). Previous empirical studies revealed
402

that many Jamaican males are in multiple sexual relationships and that 1 in every 3 males consistently used a condom with their steady intimate partners24-26, suggesting that females who are in sexual unions are exposed to sexually transmitted infections, particularly HIV/AIDS. The aforementioned issue justifies statistics on the prevalence of the HIV/AIDS virus in Jamaica.27, 28 Despite the history of millions of dollars spent on the HIV/AIDS virus, reproductive health education and other sex education programmes in the Caribbean, particular Jamaica, the virus has been increasing27, 35 embodied here is the resultant effect of intimate encounters. Statistics revealed that the mean age at first coitus among females Jamaicans in 2002 was 15.8 years and 13.5 years for males.1 Clearly from the aforementioned finding, intimate unions commence at an early age in Jamaica, which is also typically the case in the United States29 as well as other developing nations. Previous studies have shown that in Antigua and Barbuda, Haiti, Guyana, Trinidad and Tobago and Dominica Republic, one in six women between the ages of 15 and 24 become sexually active before the age of 15 years.30, 31 However, researchers found that the First sexual intercourse almost always take place outside of a formal union32 and with older men (for the females).33 The intimate unions that people enter and/or remain in are accounting for many reproductive health matters such as HIV/AIDS and pregnancy. A dyadic of scholars (or group) found that, 2 in every 5 Jamaican women have been pregnant at least once, 4 in every 5 adolescent women pregnancies were unplanned and 74% of females ages 15-17 years old were sexually active compared to 47% of males of the same age.34 And that 1 in every 50 people in the Caribbean being infected with the HIV/AIDS virus; AIDS being the main cause of deaths among people aged 15-44 years.28 The World Health Organization (WHO) offered some explanation for the reproductive health issues which emerged in many nations when they opined that unsafe
403

sexual practices are a part of risk factors which account for increased mortality and morbidity in the world.35 Inspite of the prevalence of the HIV/AIDS virus and unwanted pregnancy, women in sexual unions (being it, visiting, common-law or marriage partnership) were not protecting themselves from the probability of contracting the virus as a study found that 43.3% of Jamaican women aged 15-19 years old and about 66% of those aged 15-49 years reported using a condom in the last 30 days1 as well as 41.2% of females aged 15-74 years old used a condom20, indicating not only premarital sexual relations, but also risky lifestyle practices.1 The current work found that those who are currently in intimate unions (cohabiting unions) were 29.9 times more likely to have sexual intercourse; 1.8 times more likely to use a method of contraception; 1.4 times more likely to want to have more children; mostly likely to be in the poorest 20% (62.3%) compared with those in the wealthiest 20% (40.0%), 1.6 times more likely to have done a HIV test than a those in non-intimate unions (non-cohabiting), but they were less likely to use a condom. Knowing that women aged 15-49 years old who are involved in sexual unions want to have more children, inconsistent condom usage is inevitable, which means that they are exposed to sexually transmitted infections (as well as human papillomavirus (HPV)), cervical cancers, pregnancy and other risky sexual issues, which is supported by the literature.9 One of the issues which emerged from this work is the fact that women aged 15-49 in intimate union years old were less educated and more likely to be employed than those in noncohabiting unions. Being less educated than their non-intimate counterparts and dwelling in crowded household, they are having to subscribing to the dictates of their partners because of economic vulnerability. A study by Wilks and colleagues comparing results for 2000 and 2008 found that self-reported unemployment increased by 7.1%20 and within the context that rate of
404

unemployment is greater for female than males21, the economic vulnerability of women in sexual unions means that they are handing the vetoing powers of their reproductive rights to their male counterparts in exchange for socioeconomic assistance or survivability. It is this reality that justifies the high rates of contracting the HIV/AIDS virus because they desire more children, suggesting a likelihood of inconsistent condom usage among them and their sexual partners. Within the context that biological sex is a fundamental structural part of the social life of women in intimate unions (22 out of every 25 had sex in the last 30 days), these highly sexed individuals coupled with data showing that 66% of them (Jamaican women) used a method of contraception, then economic vulnerability among these women denotes that the gains made in sex education is likely to be eroded because of material deprivation. Women who are currently in an intimate union although they are more likely to be employed compared with those who are currently not in a sexual union, this fact does not provide economic independency because the occupational type will be lower as a result of the lower level of education. Poverty is, therefore, accounting for the educational disparity between those in sexual union status and those current not in an intimate relationship, and thereby creating other socioeconomic problems. For the current work, almost 10 in every 25 women aged 15-49 years old who are involved in sexual unions are in visiting cohabitations, 3 out of every 4 women in the poorest 20% were in intimate unions and 14 out of every 25 being unemployed, the economic dependency is fostering the reduced sexual autonomy of these people. Embedded in the current findings is the reality that cohabitation among Jamaican women is continuously changing because of money and material resources. Such an issue highlights the challenge of restoring sexual rights among women, and how sexually transmitted infections can be easily transmitted because men are still dominating reproductive rights of women owing to their economic power.
405

This is also offering some explanations why low-income and/or lowly educated women opt for intimate relationship than those in the wealthy income groups and their reproductive health matters are controlled by the male as intimate unions are used as an avenue of escape from economic challenges and material deprivation. The current work revealed that 22.3% more women aged 15-49 years who are in common-law or visiting unions were in the poorest 20% compared to those in the wealthiest 20% and that they had the lowest rate of involvement marriage than women in other income quintiles. Those results have implication for theorizing on reproductive health intervention programmes, the role of gender in reproductive health choices as well as retardation of economics in reproductive choices of women. Almost 15 out of every 25 respondents dwelled in rural areas, and rural poverty was 1.9 times more than urban poverty, rural poverty was 1.6 times more than periurban poverty. Money (or the lack of it) explains the high cohabitation as women seek assistance from older males who are more accomplished financial. Because one in every 2 women aged 15-49 years old were unemployed rural residents, the economic challenge will be greater in among rural women which are offset by assistance from different males. Thus, it is for this very reason why women enter into sexual unions earlier than those in non-cohabiting unions. Although women who are currently in sexual unions enter these 4 months earlier than those who are not in one, they commence sexual intercourse 5.9 years later. Delaying sexual intercourse does not mean that these women asexual compared to those not in intimate unions, but the contrary is the case. Continuing, they were higher sexual being, less educated, more likely to be unemployed and in the poorest 20%, making them apart of the economically vulnerable group. Thus, this study is forward a perspective that intimate unions among Jamaican

406

women aged 15-49 years old, therefore, is for economic gains and thereby justifies the high prevalence of sexual intercourse in these relationships. Women in cohabiting union, despite their economic vulnerability, are less likely to have multiple sexual partners (2+) and reported having HIV/AIDS, suggesting that they fear for partnership dissolution because if the male becomes aware of their infidelity, the union may cease from operating. Such information envelope the risky sexual lifestyle practices of those who are currently not in an intimate union, which is owing to their economic independency. Thus, they have more control over their reproductive health matters, choice of sexual activity and sexual freedom than those in intimate unions. Despite a greater degree of educated women being among those who are in non-sexual union, they are more engaged in risky sexual practices that speaks to the disjoined between knowledge or education and sexual behaviour. This could be ascribed to culture in the Caribbean as emerged from Chevannes work, which revealed that subtly cohered into risky sexual practices in order to established there capacity to bear children following entering into puberty.36 The sexual relationship that is entered into by women aged 15-49 years is a byproduct of early socialization. This is captured into aptly in an ethnographic studies carried out in some Caribbean communities. Chevannes opined that By the time small children reach the age of seven or eight, and are in primary school, their sexual socialization would have begun in earnest, though it is probably in the immediate prepubescent period that they begin to exhibit personal, emotional interest in sex.36 Chevannes findings highlight the rationale behind the high prevalence of sexual relations among women aged 15-49 years old who are currently involved in an intimate union. In 2007/08, a nationally representative probability survey of 2, 843 Jamaicans aged 15-74 years old revealed that 96.2% of males had sex compared with 93.3% of females;
407

40.9% of male sample had multiple sexual relationships (2+ partners) compared to 8.4% of females; and of those aged 15-24 years old, 36.1% of males had multiple partners to 15.4% of females.20 Intimacy is not only a as a result of the socialization of the individual, but it is also an explanation of the choices which are made by people. Women are making particular choices (or not) on reproductive health matters based on their socioeconomic situation, and these affect their entrance into sexual partners and intimate unions. This work unearths that material deprivation and socioeconomic challenges of women affect intimate unions, choices on reproductive health matters and their current realities. It is for this reason that periurban residents, who are the wealthiest among the different area of residents, and justifies why those in the wealthiest 20% reported a greater good-to-very good health status in reference to those in the poorest 20%, and they are more likely to use a condom. Money (or, insufficient financial resource), therefore, is fostering better health status as well as lifestyle practices of women, which concurs with the literature.38-42 Previous studies have shown that those in the lower socioeconomic status are less healthy than those in the wealthy socioeconomic groups41, 42 which is supported by the current findings. Another research found that poverty was greater among chronically ill people than the nonchronically ill, 43 and the WHO39 opined that 80% of chronic illnesses were in low and middle income countries. Poverty is not only associated with illness and ill-health, but also with higher rates of mortality and intimate partnership. This work ascertained that 25.4% of women who were in intimate unions were classified in the poorest 20% of socioeconomic income compared with 21.4% of those in non-intimate unions. Thus with poverty being greater in rural area, it is affecting health status of rural women and it is also influencing their reproductive health choices
408

as they are unable to effectively address those matters without the assistance of males, thereby accounting for the higher indulgence into sexual unions. A group of scholars, using Grounded theory found that 96% of the mothers voiced a general distrust of men, yet that distrust did not deter them from involvement in intimate unions.37 The rationales for such behaviours are embedded in (1) the culture, (2) economic vulnerability and (3) material deprivation. With the downturn in the global economy and its effect on the Jamaican economy, the new economic reality of material deprivation and economic hardship among women is creating increased premarital cohabitation, and less commitment to marriage or the permanence of intimate unions. Another reality which is occurring in Jamaica is the prevalence of multiple sexual relationships, particular among men, and this is not resulting in high condom usage among women in sexual unions. Chevannes noted that in the Caribbean males are given sexual freedom, sexual autonomy and sexual promiscuity is a part of the social setting36, and this is known by women. While women may distrust men because of their sexual promiscuity, which is culturally based, the current work showed that low condom usage is higher in intimate relationships, which concurs with Burton and colleagues study.37 Statistics revealed that in 2008 of the 22,152 marriages that were entered into, 94% of the females were less than 50 years compared to 89% for males, and 44.1% of females less than 30 years old compared with 32.5% of males of the same ages. The information provide an insight into the cultural disparity of marriage between the genders, and the reality that women enter into intimate unions with older men who are likely to be promiscuous and more financial secure. This also holds true in non-married sexual unions which was found by scholars who stated that the First sexual intercourse almost always take place outside of a formal union32 with older men.
409

Using data which were collected in 2000 as part of the Ministry of Healths HIV/AIDS/STD Survey in Jamaica, Gibbison24 opined that Males in general have multiple sexual partners and less than a third use condoms on a consistent basis with their regular sex partners.Clearly, women in certain regions or subpopulations face an increased risk of contracting sexually transmitted infections due to the sexual choices of their partners. Chevannes work showed that the cultural underpinnings in the Caribbean offers much reason for the lifestyle practices and choices of men and women, and creates a justification for the rationale of women subscribing to the culture instead of wanting an idealistic world.37 This research offers some explanation for the aforementioned issues as poverty is eroding the good lifestyle and sexual practices of women irrespective of knowing that their choices are risky, and understanding the sexual freedom on their male partners. Such issues correspond with high distrust of women for men, but do not militate against intimate unions or inconsistent condom usage because of the economic power of men. The disproportionate economic power between the genders dictates reproductive health matters of women as the economic benefits outweigh the risk of HIV/AIDS or other sexually transmitted infections. Thus, the risk associated with the choices made by women is enveloped within the disproportionate economic power between the genders, and this means that they are likely to experience morality associated with economic vulnerability. Douglas postulated that the major cause of mortality among women aged 15-44 years in the Caribbean is AIDS18 and Wilks et al forwarded that sexually transmitted infections (STI) is greater among males (18.1%) than females (11.0%) and that 41% of males had 2+ partners compared to 8.4% of females, 20 yet inconsistent condom usage and low condom usage is found among women who indicated being in an intimate union. This research is forwarding that sexual
410

promiscuity among women as well as inconsistent condom usage and reduced sexual autonomy is associated with economic hardships. Within the context of their socioeconomic realities, in an attempt to address the economic hardship, women enter into non-stable intimate unions (visiting unions) as a source of survivability. Thus, they will engaged into those unions, knowing that the male is already involved in a sexual union (visiting or extra marital relationship) as a means of providing for themselves and children, and not because of sexual freedom. It is this reality which justifies why 3 out of every 4 women in a sexual union were engaged in either common-law or visiting unions as these are easier to form, and provide some source of income because they are involved with older men who are able to offer assistance compared to younger men. The matter becomes even more complex in rural communities with less males, more women of the reproductive ages who are unemployed, of low educational attainment, living with families who are unable to offer much financial assistance and they females are unable to move to urban centres because their families residing in those areas are living in inner city communities with little opportunities. There is another side to the discourse on intimate or non-intimate unions, which is emotional satisfaction and physical pleasure. Sexual unions are not merely about economic stability, but this relates to emotional satisfaction and physical pleasure.44 It can be argued that lower prevalence of multiple sexual partners among women in intimate unions that the degree of emotional satisfaction would be greater as women sex is associate with higher emotional satisfaction. Koo, Rie & Parks study revealed that being married was a good cause for an increase in psychological and subjective wellbeing in old age.45 Delbs & Gaymu study that reads The widowed have a less positive attitude towards life than married people, which is not an unexpected result.46 Another research, using a sample of
411

1049 Austrians from ages 14 years and over, found that married individuals reported better subjective health - related quality of life index (8.3 ) than divorced persons (7.6) or singles (7.7).47 Other studies have shown that married people have a lower mortality risk in the healthy category than the nonmarried48 which explains why they take less life-threatening risks49,50 and are happier than non-married people.51 Intimate unions, therefore, are providing economic opportunities for women aged 15-49 years as well as positive affective psychological conditions (emotional satisfaction) as explained by Soons and Kalmijn, which supports longer and better quality of life.52 Because married women have more access to shared financial resources as a result of their male partners resources as well as the stability of those unions, many women seek this partnership. With more material resources in married and common-law union, women in such relationship are able to seek more medical care, which accounts for better health and more productive contribution to economic growth. A study by Bourne offered more of an explanation of interconnectedness among poverty, unemployment, not seeking medical care, morality and economic conditions in the nations.53 He found that there was a (1) positive statistical correlation between poverty and unemployment (R2 = 0.48); (2) positive association between not seeking medical care and unemployment (R2 = 0.58); (3) positive relationship between not seeking medical care and poverty (R2 = 0.58); (4) positive correlation between poverty and inflation (R2 = 0.73); (5) negative statistical association between seeking medical care and inflation (R2 = 0.67); and (6) a negative correlation between not seeking medical care and mortality (R2 = 0.56). Bournes work has provided some insight into the justification for high negative affective conditions and poverty, poverty and seeking sexual partnerships, poverty and death, economic

412

hardship and poor health, disparity in reproductive rights owing to gender differences, and acceptance of mens sexual promiscuity. While poverty is retarding womens sexual autonomy, economic independency and reproductive health freedom, the positive affective psychological conditions which emerged from economic independency is equally creating challenges for policy makers. As women who are economic independent, employed, educated and are able to seek medical care as well as decide what they want for life are engaged in more risky sexual practices than those who involved in intimate unions out of (or not) economic vulnerability. Thus, the economic power that men acquire from material resources explains their power over vulnerable women who seek financial assistance, and clearly this is equally the case among economically power women. These economically power women are more likely to reside in periurban areas, wealthier, more educated, more likely to currently not in an intimate unions, but they are more likely to have 2+ sexual partners not use a condom on the last sexual encounter, which are feature of Jamaican males. Chevannes argued that the culture accounts for sexual expressions of males, and while this is true, economics appears to be at the root of these behaviours as the sexual freedom of males in Caribbean as this is found to be the case among economically power women who between 14 and 50 years in Jamaica.

Conclusion
Biological sex is fundamentally structured into cohabitation between men and women, particularly among women aged 15-49 years old in Jamaica. The current findings showed that those in the poorest 20% were most likely to be in intimate unions, it can be extrapolated from that finding that sexual intercourse is used by women as a means of addressing economic and material deprivation. Poverty is greatest in rural zones, which accounts for the lower health status
413

among the sample as well as cohabitation, lower educational attainment, and greater household crowding. Continuing, lower educational achievement and household crowding were found among those in intimate unions, justifying the rationale why they are in highly sexed unions as a means of obtaining economic resources. On the other hand, women who are currently not in intimate unions were more educated, wealthier and this provides them with economic independency. The economic independency provides them with sexual freedom, sexual autonomy and choice of reproductive health matters than those in romantic unions, which explains their risky sexual practices and higher HIV infections. Involvement into sexual unions among women in the reproductive ages is highly based on sexual intercourse and economic gains, which should be used to guide policy formulation and intervention programmes. Even though biological sex is lower among women aged 15-49 years who are currently not in a romantic union, their greater degree of education is not influencing better sexual practices as they are more likely to have multiple sexual partners (2+) and less likely to use methods of contraception. As such, as women become economically independent, the gains to intimate unions decline, meaning that employment and education provide women with the same power and choice over their reproductive rights and sexual freedom as those current had by men. Thus, sexual union is engaged into because of its gains (emotional satisfaction, positive affective psychological conditions, and economic benefits).54 Those issues highlight the need to institute intervention programmes geared towards both those in sexual and not in romantic unions. Because poverty can reduce health status and sexual autonomy38,55 and a group of researchers went further to say that money buys health,56 although this is not necessarily the case among Jamaican females as Bourne55 found that income does not reduce health conditions, a sex
414

education programmes must recognize these facts. Poverty reduces the gains of an intervention programmes as it incapacities an individual from good nutrition, educational advancement, adequate physical milieu, long life and better choices.57,58 As such, poverty, unemployment and area of residence impinge on how women live58 dictate entrance into intimate partnership, which highlight the multisectoral approach that must be taken in order to establish sexual education or lifestyle intervention programmes because economic disparity is critical in a thrust against risky sexual practices and sexual autonomy. In summary, intimate unions that are entered into among women aged 15 to 49 years are driven by sexual intercourse and economic situation, and are entered into because of gains. On the contrary, women who are currently not in a sexual union were more likely to be more educated, use a condom, wealthier and these are responsibility for their economic independency, sexual autonomy and risky sexual practices. Despite women distrust for men as a result of the cultured sexual freedom, this has not fundamentally affected the structure of intimate partnership because economics is important to entrance into sexual relationship. Those findings highlight a need to institute measures to alleviate gendered poverty and gendered economic inequality. But economics independency among women of the reproductive ages must be met by intervention programmes fashioned to address sexual promiscuity, sexual freedom, and risky sexual practices. As such, a safe-sex social norm intervention programme needs to be developed for women aged 15 to 49 years old; particular those who are wealthy, educated, live in periurban and rural areas as well as those in sexual unions. Disclosures The authors report no conflict of interest with this work.

415

Disclaimer
The researchers would like to note that while this study used secondary data from the Reproductive Health Survey, none of the errors in this paper should be ascribed to the National Family Planning Board, but to the researchers.

Acknowledgement
The authors thank the Data Bank in Sir Arthur Lewis Institute of Social and Economic Studies, the University of the West Indies, Mona, Jamaica for making the dataset (Reproductive Health Survey, RHS) available for use in this study, and the National Family Planning Board for commissioning the survey.

416

References
1. Jamaica, National Family Planning Board (NFPB). Reproductive Health Survey, 2002. Kingston: NFPB; 2005 2. Family Care International. Commitments to Sexual and Reproductive Health and Rights for All. Framework for Action. New York, NY: Family Care International; 1995 3. US Department of Health and Services (2006). Gender Differences in Reproductive Health. Department of Health and Human Services, Centre for Disease Control, USAID and Jamaica National Family Planning Board. 4. Sexual Initiation. American Sexual Behaviour. http://www.newstrategist.com/productdetails/Sex.SamplePgs.pdf (Accessed on April 1, 2010). 5. Miller KS, Clarke LF, Moore JS. Sexual initiation with older male partners and subsequent HIV risk behavior among female adolescents. Family Planning Perspectives; 29, 1997:212-214. 6. Warren CW, Powell D, Morris L, Jackson J, Hamilton P. Fertility and family planning among young adults in Jamaica. Int Family Planning Perspectives 1988; 14(4):137-141. 7. Eggleston E, Jackson J and Hardee K. Sexual attitudes and behavior among young adolescents in Jamaica. Guttmacher. International Family Planning Perspectives; 25(2), 1999. 8. Williamson LM, Parkes A, Wight D, Petticrew M, Hart GJ. Limits to modern contraceptive use among young women in developing countries: A systematic review of qualitative research. Reproductive Health 2009; 6:3 9. Henry-Lee A. Womens reasons for discontinuing contraceptive use within 12 months: Jamaica. Reproductive Health Matters 2001; 9(17):213-220. 10. World Health Organization (WHO). Reproductive health research at WHO: A new beginning. Biennial report 1998-1999. Geneva: WHO; 2000. 11. World Health Organization (WHO). World health statistics, 2009. Geneva: WHO; 2009. 12. Fatusi AO, and Blum, RW. Predictors of early sexual initiation among a nationally representative sample of Nigerian adolescents. BMC Public Health; 8, 2008. 13. Santelli JS, Kaiser J, Hirsch L, Radosh A, Simkin L, Middlestadt S: Initiation of sexual intercourse among middle school adolescents: the influence of psychosocial factors. J Adolesc Health 2004, 34:200-208. 14. Penfold SC, Teijlingen ERV, Tucker JS. Factors associated with self-reported first sexual intercourse in Scottish adolescents. BMC Research Notes 2009; 2:42. 15. Rosenthal DA, Smith AMA, De Visser R. Personal and social factors influencing age at first sexual intercourse. Archives of Social Behavior 1999; 28(4):319-333. 16. Louie KS, de Sanjose S, Diaz M, et al. Early age at first sexual intercourse and early pregnancy are risk factors for cervical cancer in developing countries. Br J Cancer 2009; 100(7):119-7. 17. State Council AIDS Working Committee Office and UN Theme Group on HIV/AIDS in China: A joint assessment of HIV/AIDS prevention, treatment and care in China (2005). Beijing 2005. 18. Douglas DL. Perspectives on HIV/AIDS in the Caribbean. In: Morgan O. ed. Health issues in the Caribbean. Kingston: Ian Randle Publisher; 2005: pp. xv.
417

19. Yan H, Chen W, Wu H, et al. Multiple sex partner behavior in female undergraduate students in China: A multi-campus survey. BMC Public Health 2009; 9:305. 20. Wilks R, Younger N, Tulloch-Reid M, McFarlane S, Francis D. Jamaica health and lifestyle survey 2007-8. Kingston: Tropical Medicine Research Institute, University of the West Indies, Mona; 2008. 21. Planning Institute of Jamaica (PIOJ), Statistical Institute of Jamaica (STATIN). 2008. Jamaica Survey of Living Conditions, 2007. Kingston: PIOJ, STATIN. 22. World Health Organization (WHO). Preventing Chronic Diseases a vital investment. Geneva: WHO; 2005. 23. Statistical Institute of Jamaica (STATIN). Demographic statistics, 2007, 2009. Kingston: STATIN; 2008, 2010. 24. Gibbison GA. Attitude towards intimate partner violence against women and risky sexual choices of Jamaican males. West Indian Med J 2007; 56(1):66-71. 25. Ward M The reproductive and sexual health of Jamaican youth. Advocates for Youth. Washington DC. Advocates for Youth; 2001. 26. Bourne PA, Charles CAD. Contraception usage among young adult men in a developing country. Open Access J of Contraception 2010; 1:51-59. 27. Pan American Health Organization (PAHO). Health in the Americas, 2007, volume II Countries. Washington, D.C.; 2007: pp.448-464. 28. Bain B. HIV/AIDS the rude awakening/stemming the tide. In: Morgan O. ed. Health issues in the Caribbean. Kingston: Ian Randle Publisher; 2005: pp. 62-76. 29. Santelli JS, Brener ND, Lowry R, Bhatt A, Zabin LS. Multiple sexual partners among U.S. adolescents and young adults. Family Planning Perspectives 1998; 30:271-275. 30. Rawlins J. Teenage Pregnancy: A study in three communities in Trinidad and Tobago. Paper presented at the Caribbean Health Research Conference 2007, Jamaica. 31. Jamaica Observer. Study shows Jamaican Girls Encounter Violent Sexual Relationships. Jamaica Observer (15 April 2009). 32. US Department of Health and Services (2006). Gender Differences in Reproductive Health. Department of Health and Human Services, Centre for Disease Control, USAID and Jamaica National Family Planning Board. 33. Sexual Initiation. American Sexual Behaviour. http://www.newstrategist.com/productdetails/Sex.SamplePgs.pdf (Accessed on April 1, 2010). 34. Crawford TV, McGrowder DA, Crawford A. Access to contraception by minors in Jamaica: a public concern. North Am J of Med Sci 2009; 1(5):247-255. 35. World Health Organization (WHO). World health statistics, 2009. Geneva: WHO; 2009. 36. Chevannes B. Learning to be a man: Culture, socialization and gender identity in five Caribbean communities. Kingston, Jamaica: The Univer. Of the West Indies Press; 2001. 37. Burton LM, Cherlin A, Winn DM, Estacion A, Holder-Taylor C. The role of trust in lowincome mothers intimate unions. J Marriage Fam 2009; 71(5):1107-1124. 38. Marmot M. The influence of Income on Health: Views of an Epidemiologist. Does money really matter? Or is it a marker for something else? Health Affairs. 2002; 21: 3146. 39. World Health Organization. Preventing Chronic Diseases a vital investment. Geneva: WHO; 2005.
418

40. Pan American Health Organization, (PAHO). Investment in health: Social and economic returns, Scientific and Technical Publication, No. 582. Washington DC: PAHO, WHO; 2001. 41. Fox J ed. Health inequalities in European Countries. Aldershot: Gower Publishing Company Limited; 1989. 42. Illsley R, Svenson PG, ed. Health inequalities in Europe. Soc Sci Med 1990; 31(special issue):223-420. 43. Van Agt HME, Stronks K, Mackenbach JP. Chronic illness and poverty in the Netherlands. Eur J of Public Health 2000; 10:197-200. 44. Waite LJ, Joyner K. Emotional satisfaction and physical pleasure in sexual unions: Time horizon, sexual behavior, and sexual exclusivity. J of Marriage and Family 2001; 63(1): 247-264. 45. Koo J, Rie J, Park K. Age and gender differences in affect and subjective wellbeing. Geriatrics and Gerontology International, 2004; 4:S268-S270. 46. Delbs C, Gaymu J. The shock of widowed on the eve of old age: Male and female experience. Demography 2002; 3: 885-914. 47. Prause W, Saletu B, Tribl GG, Rieder A, Rosengerger A, Bolitschek J, Holzinger B, et al. Effects of socio-demographic variables on health-related quality of life determined by the quality of life indexGerman version. Human psychopharmacology Clinical and Expremental. 2005; 20:359-365. 48. Goldman N. Marriage selection and mortality patterns: Inferences and fallacies. Demography 1993; 30:189-208. 49. Smith KR, Waitzman NJ. Double jeopardy: Interaction effects of martial and poverty status on the risk of mortality. Demography 1994; 31:487-507. 50. Umberson D. Family status and health behaviors: Social control as a dimension of social integration. Journal of Health and Social Behavior 1987; 28:306-19. 51. Diener E. Subjective wellbeing. Psychological Bulletin, 1984; 95:542-575. 52. Soons JPM, Kalmijn M. Is marriage more than cohabitation? Wellbeing differences in 30 European Countries. Journal of Marriage and Family 2009; 71(5):1141-1157. 53. Bourne PA. Impact of poverty, not seeking medical care, unemployment, inflation, selfreported illness, health insurance on mortality in Jamaica. North American Journal of Medical Sciences 2009; 1(3):99-109. 54. Becker GS. A Theory of Marriage: Part I. Journal of Political Economy, 1973; 81, 812-846. 55. Bourne PA. Health of Females in Jamaica: using two cross-sectional surveys. North Am J Med Sci. 2009;1: 272-278. 56. Smith JP, Kington R. Demographic and Economic Correlates of Health in Old Age. Demography 1997; 34:159-70. 57. Bourne PA. Health status and Medical Care-Seeking Behaviour of the poorest 20% in Jamaica. International Journal of Collaborative Research on Internal Medicine & Public Health, 2009;1(6&7):167-185. 58. Bourne PA, Rhule J. Good Health Status of Rural Women in the Reproductive Ages. International Journal of Collaborative Research on Internal Medicine & Public Health, 2009;1(5):132-155.

419

Figure 1: Intimate unions (married, common-law and visiting) by population income quintiles

420

Table 15.1: Demographic characteristics of sample by currently in or not in a sexual union Characteristic Sample Currently in or not in a sexual union Not in In sexual sexual union union n (%) n (%) 354 (14.9) 656 (27.6) 1371 (57.6) NA NA 510 (21.4) 469 (19.7) 476 (20.0) 544 (22.8) 382 (16.0) 0 (0.0) 0 (0.0) 0 (0.0) 1344 (56.4) 1037 (43.6) 1260 (52.9) 1121 (47.1) 1625 (68.2) 756 (31.8) 18.9 (4.2) 19.0 (3.8) 10.8 (8.5) 12.8 (1.5) 1.6 (0.5) 28.1 (11.5) 844 (14.3) 1627 (27.7) 3407 (58.0) NA NA 30.00,< 0.0001 2001 (24.2) 1716 (20.8) 1668 (20.2) 1650 (20.0) 1224 (14.8) 1441 (17.4) 2158 (26.1) 2279 (27.6) 1344 (16.3) 1037 (12.6) 5403 (65.4) 2856 (34.6) 4857 (58.8) 3402 (41.2) 18.7 (4.3) 19.0 (3.9) 15.0 (5.8) 13.0 (1.5) 1.7 (0.5) 31.6 (9.9) 1491 (25.4) 1247 (21.2) 1192 (20.3) 1106 (18.8) 842 (14.3) 8259.0, <0.0001 1441 (24.5) 2158 (36.7) 2279 (38.8) 0 (0.0) 0 (0.0) 231.09, <0.0001 4143 (70.5) 1735 (29.5) 123.07, <0.0001 3232 (55.0) 2646 (45.0) 18.6 (4.4) 19.1 (4.1) 16.7 (2.6) 13.0 (1.6) 1.7 (0.5) 32.9 (8.8) t-test, P t= 2.40, 0.017 t= 1.29,0.195 t=-31.99,<0.0001 t=-4.82,<0.0001 t=-8.13,<0.0001 t=-18.27,<0.0001 2 , P

n (%) Area of residence Urban Periurban Rural Currently in a sexual union No Yes Income quintile Poorest 20% Second poor Middle Second wealthy Wealthiest 20% Marital status Married Common-law Visiting Separated, divorced & widowed Single High church attendance No Yes Employed No Yes Age enter into first sexual union Age at first contraceptive use Age of sexual debut Age of menarche Crowding Age of respondents 1198 (14.5) 2283 (27.6) 4778 (57.9) 2381 (28.8) 5878 (71.2)

0.36, 0.837

421

Table 15.2: Reproductive health matters by currently in or not in a sexual union Currently in a sexual union Characteristic Sample 2 , P Not in In a sexual sexual union union n (%) n (%) n (%) Had sex in last 30 days 4121.57,<0.0001 No 2873 (34.8) 2087 (87.7) 786 (13.4) Yes 5386 (65.2) 294 (12.3) 5092 (86.6) Gynaecological examination 210.99, <0.0001 No 4664 (56.5) 1643 (69.0) 3028 (51.5) Yes 3588 (43.4) 738 (31.0) 2850 (48.5) Physically forced to have sex 3.21,0.073 No 6622 (91.7) 1216 (90.5) 5406 (92.0) Yes 600 (8.3) 128 (9.5) 472 (8.0) Ever done HIV test 467.05,<0.0001 No 3016 (45.8) 1404 (64.5) 1600 (36.3) Yes 3560 (54.2) 769 (35.3) 2797 (63.5) Currently used method of contraception 867.07, <0.0001 No 2580 (35.7) 1011 (68.3) 1571 (27.3) Yes 4647 (64.3) 469 (31.7) 4178 (72.7) Want to have more children 0.007, 0.934 No 3898 (58.4) 845 (58.3) 3053 (58.4) Yes 2780 (41.6) 605 (41.7) 2175 (41.6) Sexually abused 83.87, <0.0001 No 7844 (95.0) 2179 (91.5) 5665 (96.4) Yes 415 (5.0) 202 (8.5) 213 (3.6) Have HIV/AIDS (self-reported) 155.07, <0.0001 No 6815 (82.5) 1770 (74.3) 5045 (85.8) Yes 1444 (17.5) 611 (25.7) 833 (14.2) Last method of contraception 35.83, <0.0001 Injection 393 (15.4) 132 (13.3) 261 (16.8) Pill 723 (28.4) 242 (24.3) 481 (31.0) Condom 1243 (48.8) 559 (56.1) 684 (44.1) Emergency oral contraception 45 (1.8) 13 (1.3) 32 (2.1) Withdrawal or natural method 139 (5.5) 48 (4.8) 91 (5.9) Other 5 (0.2) 2 (0.2) 3 (0.2) Number of sexual partners in last 12 39.16, <0.0001 months 1 6129 (93.0) 721 (88.1) 5408 (93.7) 2+ 438 (6.6) 89 (10.9) 349 (6.0) No response 22 (0.3) 8 (1.0) 14 (0.2)

422

Table 15.3: Demographic characteristic of sample by area of residence Area of residence Characteristic Urban n (%) Socioeconomic status Lower class Middle class Upper class High church attendance No Yes Employed No Yes Currently in a sexual union No Yes Have HIV/AIDS No Yes Dont know Had sex in last 30 days No Yes Health status Very good Good Moderate Poor Marital status Married Common-law Visiting Separated, divorced & widowed Single Want more children No Yes Age enter into first sexual union Age at first contraceptive use Age of sexual debut Age of menarche Crowding Age of respondents 172 (14.4) 670 (55.9) 356 (29.7) 854 (71.3) 344 (28.7) 598 (49.9) 600 (50.1) 354 (29.5) 844 (70.5) 268 (74.9) 87 (24.3) 3 (0.8) 415 (39.6) 783 (65.4) 234 (19.6) 679 (56.8) 256 (21.4) 26 (2.2) 148 (12.4) 293 (24.5) 403 (33.6) 187 (15.6) 167 (13.9) 551 (58.2) 395 (41.8) 18.0 (4.0) 19.1 (4.1) 28.1 (28.5) 12.8 (1.6) 1.7 (0.5) 31.4 (10.3) 423 Periurban n (%) 407 (17.8) 1111 (48.7) 765 (33.5) 1475 (64.6) 808 (35.4) 1220 (53.4) 1063 (46.6) 656 (28.7) 844 (70.5) 345 (57.5) 246 (41.0) 9 (1.5) 801 (35.1) 1482 (64.9) 705 (31.2) 937 (41.4) 535 (23.6) 86 (3.8) 429 (18.8) 582 (25.5) 616 (27.0) 420 (18.4) 236 (10.3) 1049 (55.9) 827 (44.1) 19.1 (4.7) 19.2 (3.8) 24.3 (23.5) 12.9 (1.6) 1.6 (0.5) 32.1 (9.8) Rural n (%)

2, P

319.06, 0.0001 1320 (27.6) 2633 (55.1) 825 (17.3) 21.36, <0.0001 3074 (64.3) 1704 (35.7) 111.64, <0.0001 3039 (63.6) 1739 (36.4) 0.36, 0.835 1371 (28.7) 3407 (71.3) 33.35, <0.0001 831 (60.0) 540 (39.0) 13 (0.9) 0.13, 0.939 1657 (34.7) 3121 (65.3) 89.79, <0.0001 1256 (26.4) 2290 (48.1) 1032 (21.7) 178 (3.7) 64.12, <0.001 864 (18.1) 1283 (26.9) 1260 (26.4) 737 (15.4) 634 (13.3) 7.04, 0.030 2298 (59.6) 1558 (40.4) 18.6 (4.3) 18.9 (3.8) 25.6 (25.9) 13.0 (1.5) 1.7 (0.5) 31.3 (9.9) F statistic, P 26.17 <0.0001 4.69, 0.009 7.94, <0.0001 9.83, <0.0001 29.22, <0.0001 5.86, 0.002

Table 15.4: Logistic regression analyses: Explanatory variables of good-to-very good health status
Explanatory variables coefficient Periurban Reference group (rural area) Physically forced to have sex (1=yes) Currently using method of contraception Age of sexual debut Years of schooling Age Employed (1=yes) Logged fertility Wealthiest 20% Second poor Reference group (poorest 20%) -0.195 Std. error 0.08 Wald statistic 5.42 Odds ratio 0.82 1.00 0.63 1.26 1.04 1.05 0.96 1.22 0.82 1.44 1.24 1.00

CI (95%) 0.70 - 0.97

-0.463 0.232 0.042 0.044 -0.045 0.198 -0.200 0.364 0.213

0.13 0.08 0.02 0.02 0.01 0.08 0.08 0.13 0.10

13.70 8.10 6.32 4.22 57.59 6.52 6.94 8.45 5.05

0.49 - 0.80 1.08 - 1.48 1.01 - 1.08 1.00 - 1.09 0.95 - 0.97 1.05 - 1.42 0.71 - 0.95 1.13 - 1.84 1.03 - 1.49

Model chi-square = 198.25, P<0.0001 -2 Log likelihood = 4133.85 Nagelkerke r-squared = 0.18 n = 5781 Hosmer and Lemeshow test, 2 = 4.87, P = 0.772 Overall correct classification = 71.9% Correct classification of cases in good-to-very good health status = 97.9% Correct classification of cases not in good-to-very good health status = 60.0%

424

Table 15.5: Logistic regression analyses: Explanatory variables of method of contraception


Wald statistic Odds ratio

Explanatory variable Had sex in last 30 days (1=yes) Age Employed (1=yes) Logged fertility Wealthiest %20 Reference group (1=poorest 20%) Visiting Reference group (1=single) Health status (1=good-to-very good) Currently in a sexual union (1=yes) Constant

coefficient

Std. error

CI (95%)

2.04 -0.05 0.19 0.80 0.30

0.12 0.01 0.09 0.08 0.13

294.50 69.29 4.74 94.34 5.79

7.66 0.95 1.21 2.22 1.35 1.00 1.31 1.00 1.23 1.82 0.83

6.07 - 9.66 0.94 - 0.96 1.02 - 1.43 1.89 - 2.60 1.06 - 1.73

0.27

0.10

6.90

1.07 - 1.61

0.21 0.60 -0.19

0.09 0.14 0.25

4.85 17.92 0.59

1.02 - 1.47 1.38 - 2.41

Model chi-square = 1021.31, P<0.0001 -2 Log likelihood = 3530.81 Nagelkerke r-squared = 0.344 n = 5781 Hosmer and Lemeshow test, 2 = 4.87, P = 0.772 Overall correct classification = 71.9% Correct classification of cases in good-to-very good health status = 97.9% Correct classification of cases not in good-to-very good health status = 60.0%

425

Table 15.6: Logistic regression analyses: Explanatory variables of had sex in last 3 months
Explanatory variables coefficient Gynaecological exam (1=yes) Age High church attendance (1=once per wk) Second poor Reference group (poorest 20%) Visiting Reference group (Single) Currently used method contraception Currently in a sexual union (1=yes) Std. error Wald statistic Odds ratio

CI (95%)

0.30 -0.03 -0.39 0.50

0.12 0.01 0.13 0.15

6.81 13.57 9.76 11.75

1.35 0.97 0.68 1.66 1.00 0.29 1.00 7.47 73.77 0.14

1.08 - 1.70 0.96 - 0.99 0.53 - 0.86 1.24 - 2.21

-1.25

0.13

86.47

0.22 - 0.37

2.01 4.30

0.12 0.16 0.32

295.75 714.89 38.28

5.94 - 9.40 53.82 101.11

Constant -1.99 Model chi-square = 2005.66, P<0.0001 -2 Log likelihood = 2097.69 Nagelkerke r-squared = 0.63 n = 5781 Hosmer and Lemeshow test, 2 = 11.31, P = 0.0.19 Overall correct classification = 88.9% Correct classification of cases in had sex in last 3 months = 95.8% Correct classification of cases not in had sex in last 3 months = 68.8%

426

Table 15.7: Logistic regression analyses: Explanatory variables of high church attendance
Wald statistic 4.29 13.58 Odds ratio 0.83 0.64 1.00 1.05 1.06 1.04 0.12 0.13 0.26 1.00 0.80 0.66 0.20

Explanatory variables Peri-Urban Urban Reference group (rural) Age of sexual debut Years of school Age Common-law Visiting Separated, divorced & widowed Reference group (single) Currently used method contraception Had sex in last 30 days (1=yes) Constant

coefficient -0.19 -0.45

Std. error 0.09 0.12

CI (95%) 0.69 - 0.99 0.50 - 0.81

0.05 0.06 0.04 -2.11 -2.04 -1.34

0.02 0.02 0.01 0.11 0.11 0.15

10.11 7.73 39.71 346.30 318.09 79.09

1.02 - 1.09 1.02 - 1.11 1.03 - 1.05 0.10 - 0.15 0.10 - 0.16 0.20 - 0.35

-0.22 -0.42 -1.63

0.10 0.13 0.42

5.05 9.84 15.25

0.66 - 0.97 0.51 - 0.86

Model chi-square = 811.16, P < 0.0001 -2 Log likelihood = 3814.64 Nagelkerke r-squared = 0.28 n = 5781 Hosmer and Lemeshow test, 2 = 4.46, P = 0.81 Overall correct classification = 74.6% Correct classification of cases in high church attendance = 53.5% Correct classification of cases not in high church attendance = 85.3%

427

Table 15.8: Logistic regression analyses: Explanatory variables of physically forced to have sex
Explanatory variable PeriUrban Reference group (rural) Gynaecological exam (1=yes) Age of menarche Age of sexual debut Age Logged fertility Visiting Reference group (single) Health status (1=good-to-very good) Currently in a sexual union (1=yes) Constant 0.25 0.08 -0.08 -0.04 0.40 0.40 0.12 0.04 0.03 0.01 0.12 0.14 4.29 4.01 8.36 16.68 11.47 8.82 coefficient 0.27 Std. error 0.13 Wald statistic 4.72 Odds ratio CI (95%)

1.32 1.03 - 1.68 1.00 1.29 1.01 - 1.63 1.08 1.00 - 1.16 0.93 0.88 - 0.98 0.96 0.95 - 0.98 1.50 1.19 - 1.89 1.50 1.15 - 1.96 1.00 0.62 0.49 - 0.80 0.62 0.46 - 0.84 0.52

-0.47 -0.47 -0.65

0.13 0.15 0.66

14.30 9.35 0.97

Model chi-square = 70.36, P<0.0001 -2 Log likelihood = 215.37 Nagelkerke r-squared = 0.24 n = 5781 Hosmer and Lemeshow test, 2 = 6.99, P = 0.54 Overall correct classification = 91.0% Correct classification of cases in physically forced to have sex = 100.0% Correct classification of cases not in physically forced to have sex = 100.0%

428

Table 15.9: Logistic regression: Explanatory variables of currently in a sexual union


Characteristic Years of schooling Age Crowding High church attendance (1=yes) Age of menarche Had sex in last 30 days Age of sexual debut Good-to-very good health status (1=yes) Did HIV/AIDS test Physically forced to have sexual intercourse Periurban Urban Reference group (Rural) Currently used method of contraception Like to have more children Wealthiest 20% Second wealthy Middle class Second poor Reference group (Poorest 20%) Age began using contraception Age into first sexual union Gynaecological examination Employed (1=yes) Number of sexual partners in last 12 months Constant 0.00 -0.01 0.22 0.64 -0.833 -2.29 0.02 0.02 0.12 0.05 0.19 0.71 0.001 0.07 3.77 149.09 19.57 10.47 0.982 0.798 0.045 <0.0001 <0.0001 0.001 0.56 0.37 -0.28 -0.36 -0.16 -0.10 0.13 0.13 0.08 0.07 0.08 0.08 19.30 7.83 12.39 24.55 4.30 1.62 <0.0001 0.005 <0.0001 <0.0001 0.038 0.203 coefficient -0.07 0.03 0.25 -0.13 0.06 3.40 -0.01 -0.01 0.47 -0.06 -0.22 -0.23 Std. error 0.03 0.01 0.12 0.11 0.03 0.13 0.02 0.12 0.11 0.19 0.12 0.16 Wald statistic 4.66 12.35 4.48 1.26 3.34 713.41 0.35 0.01 18.17 0.09 3.13 2.05 P 0.031 <0.0001 0.034 0.263 0.068 <0.0001 0.554 0.953 <0.0001 0.763 0.007 0.152 Odds ratio 0.94 1.03 1.29 0.88 1.06 29.85 0.99 0.99 1.60 0.95 0.81 0.80 1.00 1.76 1.44 0.75 0.70 0.86 0.91 1.00 1.00 1.00 1.25 1.9 0.44 0.10 0.97 - 1.03 0.97 1.03 1.00 1.57 1.72 2.11 0.30 0.63 1.37 - 2.26 1.12 - 1.86 0.64 0.88 0.60 0.80 0.74 0.99 0.79 1.05 CI (95%) 0.88 - 0.99 1.01 -1.04 1.02 - 1.63 0.71 - 1.10 1.00 - 1.14 23.27 - 38.30 0.94 - 1.03 0.78 - 1.26 1.29 - 1.99 0.65 - 1.37 0.63 - 1.02 0.59 - 1.09

Model chi-square = 1672.81, P < 0.0001 -2 Log likelihood = 1633.47 Nagelkerke r-squared = 0.59 n = 5781 Hosmer and Lemeshow test, 2 = 5.96, P = 0.65 Overall correct classification = 89.4% Correct classification of cases in sexual union = 91.1% Correct classification of cases not in sexual union = 82.0%

429

16
Sociodemographic correlates of age at sexual debut among women of the reproductive years in a middle-income developing nation
Early sexual debut poses both health and social risks, which justifies this study. The main objective of this paper was to test a hypothesis of socioeconomic variables which explains age at first sexual intercourse of Jamaican women aged 15-49 years. The mean age at first sexual intercourse of the sample was 16.8 years with 80% of the subjects having had their first sexual encounter before their 19th birthday. On average, those who frequently attended church start having sexual intercourse 12 months later (17.4 years, SD = 3.2 years) than those who did not (16.4 years, SD = 2.4 years). Only 11% of those who had their first sexual intercourse were forced to do so. Eleven variables emerged as statistically significant predictors of age of first sexual relations (F-statistic = 176.2, P < 0.0001): Age began using method of contraception; education; subjective social class; forced sexual activity; church attendance; crowding; employment; age of first menarche; shared sanitary convenience; marital status, and area of residence. Public health policies have failed to effectively address the reduction in age at first sexual intercourse of women in Jamaica, and this study shows a need for a multifactorial approach to intervention.

Introduction
In 1997, statistics revealed that the median age at first sexual intercourse for Jamaican women was 17.3 years and this fell to 16.0 years in 2002.1 Embedded in this finding is the lowering of premarital sexual relations with the passing of time, and the reproductive health problems associated with early sexual debut among women aged 15-49 years. Early sexual debut poses both health (STIs, HIV, HPV, pregnancy) and social (school drop-outs) risks, and continues to be
430

a public health concern among several nations.1 Inconsistent contraceptive use coupled with the continuous lowering of the age of sexual relations offers an explanation of the failure of public health programmes to effectively address sexual behaviour of females in many developing countries, particularly in Jamaica. This is embedded in statistics which showed that only 43.3% of Jamaican women aged 15-19 years old and about 66% of women aged 15-49 years reported using a condom in the last 30 days1, indicating not only premarital sexual relations, but also risky lifestyle practices and the likely to the spread of HIV/AIDS and other sexually transmitted infections.1 The lowering of the age at sexual debut further goes beyond unwanted pregnancies to health problems such as cervical cancers, human papillomavirus (HPV) and genital or anal ulceration, unsafe abortions, psychological trauma and the socioeconomic challenges for the society in the future, which makes it a public health problem worth studying. Almost 2 in every 5 Jamaican women have been pregnant at least once prior to reaching the age of 20; most of pregnancies are unplanned, especially during the adolescent years (80%). 1 The average age at first sexual initiation in Jamaica is 15.8 for females and 13.5 for males,1 much of which is forced and is seen as a direct link with violence, as well as one of the roots of sexual and reproductive health problems in the international community.2 Such problem goes against the principles of the ICPD 1994, which stipulates that when it comes to matters of sexual relations, full respect for the integrity of the individuals involved should be of the utmost.2 First sexual intercourse almost always take place outside of a formal union3 and with older men (for the females) 4, this occurrence is likely to result in health situations relating to STIs and HIV, as well as drug abuse.5

431

Inspite of the reality of the lowering of age at first sexual debut, particularly with regard to premarital sex of adolescents, the developing societies, in particular Jamaica, do not frown upon this practice.6,7 Although teenage fertility is not actively condoned in the Caribbean,6 the churches and family planning interventions have been actively campaigning against this practice as well as early sexual debut, but the practice continues. Early sexual debut, inconsistent condom usage and teenage pregnancy are not atypical in the developing world, more specifically Jamaica. A study of some sub-Saharan African and South-East Asian nations show similar sexual behaviour and attitude of young people.8 According to Warren et al.,9 the high fertility population in Jamaica was women ages 14-24 years, indicating a high degree of premarital sexual activities and inconsistent condom use within the context of reduced age at first sexual intercourse.10 A study by Henry-Lee11 showed that 66% of Jamaican women used contraceptives, but only 34% of pregnancies were planned indicating that inconsistent contraceptive use is accounting for increased HIV/AIDS and STIs in Jamaica and on a wider scale in other

developing countries, as young adults are engaged in risky sexual practices.12,13 In Antigua and Barbuda, Haiti, Guyana, Trinidad and Tobago and Dominica Republic, one in six women between the ages of 15 and 24 became sexually active before the age of 15 years.14, 15 According to Crawford, McGrowder and Crawford,
16

2 in every 5 Jamaican women

have been pregnant at least once, 4 in every 5 adolescent women pregnancies were unplanned and 74% of females ages 15-17 years old were sexually active compared to 47% of males of the same age. Moreover, Crawford and colleagues found that of the sample of adolescents, none of the females were having sexual intercourse with males within their age cohort compared to 39% of adolescent males.16 Ninety-five percentages of adolescent females sexual partners were 17+ years old compared to 78.2% of adolescent males. It can be extrapolated from the afore432

mentioned findings that premarital sexual relations are on the rise in developing nations, in particular Jamaica, and the lowering of age at first sexual intercourse among young women in the developing world is a public health concern. In a study which looks at sexual initiation of persons within the age range of 15-44 years, it was seen that protestants (similar to those of non-religion) were more likely to have their first sexual initiation within their 16th year, when compared to the Catholics (within their 17th year) and those of other religion (18th year).4 In addition to the factor of religion, the said study pointed out that young individuals who resided with both parents encountered sexual initiation later than those in other family situations.4 Another study conducted by Fatusi and Blum,17 using a sample of 2,070 adolescents who were never married, found that condom efficacy, positive attitude to family planning use, condom access, alcohol use, and higher level of religiosity were associated with age at first sexual debut. Fatusi and Blums work concurs with some of the findings of an earlier study, which found self-efficacy, alcohol and drug use, norms about having sexual intercourse, poor academic performance and gender to be factors that explain sexual initiation among middle-school, inner city youth.18 Penfold et al.,
19

using a sample of 4,379 Scottish

adolescents, found that family (parental monitoring), school life (enjoyment), gender, selfesteem, religion, and informal sexual health intervention were associated with self-reported first sexual intercourse. Penfold et al.s work added more variables to the existing body of literature on age at first sexual debut. Rosenthal et al.20 added to the afore-mentioned factors which are also associated with age at first sexual initiation. They found that the perception of greater physical maturity, expectations of earlier autonomy among gender, and the use of illicit drugs to be statistically associated with age at first sexual debut among high schoolers.

433

Inspite of the lowering of age at first sexual relations and statistics showing that HIV, and other malignant neoplasm are among the 10 leading causes of mortality among Jamaican women;
21

as well as the direct association between early age of sexual debut, the increased risk
22

of cervical cancers,

the relationship between cervical cancer and STI, in particular human

papillomavirus (HPV), and age at first sexual debut23-25. The issue of factors explaining age at first sexual initiation is unresearched in Jamaica. Most studies that have examined factors associated with age at first sexual intercourse have used young people between ages 10-30 years. In this study, we seek to elucidate correlates which account for age at sexual debut of women aged 15-49 years in Jamaica. This study is not far fetched as a previous study in Europe used ages 16-44 years.26 The main objective of this paper was to elucidate the socioeconomic variables which explain age at first sexual initiation of Jamaican women (ages 15-49 years). It explored variables relating to early sexual debut such as age of menarche, contraception, religion, education, crowding, shared sanitary convenience, forced sexual experience, marital status, employment status, subjective social class, and area of residence among women in the reproductive years.

Methods
Sample This descriptive cross-sectional study used a secondary dataset from the National Family Planning Board (Reproductive Health Survey, RHS). There are two sets of inclusion criteria, which are females and ages. The eligibility criterion for age was 15 to 49 years at last birthday. Since 1997, the National Family Planning Board (NFPB) has been collecting information on women (aged 15-49 years) in Jamaica regarding contraception usage and/or reproductive health.
434

In 2002, the Reproductive Health Survey (RHS) collected data on Jamaican men ages 15-24 years as well as women 15-49 years old. The current paper extracted only females aged 15-49 years from 2002 Reproductive Health Survey to carry out this research. The study population was 7,168 women of the reproductive ages. Stratified random sampling was used to design the sampling frame from which the sample was drawn. Using the 2001 Census sector (or sampling frame), a three-stage sampling design was used. Stage 1 was the use of a selection frame of 659 enumeration areas (or enumeration districts, EDs). This was calculated based on probability proportion to size. Jamaica is classified into four health regions. Region 1 is composed of Kingston, St. Andrew, St. Thomas and St. Catherine; Region 2 comprises Portland, St. Mary and St. Ann; Region 3 is made up of Trelawny, St. James, Hanover and Westmoreland, with Region 4 being St. Elizabeth, Manchester and Clarendon. The 2001 Census showed that Region 1 comprised 46.5% of Jamaica compared to Region 2, at 14.1%; Region 3 at 17.6% and Region 4 at 21.8%.1 In stage 2, the households were clustered into primary sampling units (PSUs), and each PSU constituted an ED, which in turn was comprised of 80 households. The previous sampling frame was in need of updating, and so this was performed between January and May 2002. The previous sampling frame was in need of updating, and so this was carried out between January 2002 and May 2002. The new sampling frame formed the basis upon which the sampling size was computed for the interviewers to use. Stage 3 was the final selection of one eligible female from each sampled household and this was done by the interviewer on visiting the household.

435

The Statistical Institute of Jamaica (STATIN) provided the interviewers and supervisors, who were trained by McFarlane Consultancy, to carry out the survey. The instrument administered was a 35-page questionnaire. The data collection began on Saturday, October 26, 2002 and was completed on May 9, 2003. Prior to the date of the final data collection, pre-testing of the instrument was conducted between March 16 and 20, 2002. A total of 175 instruments were pre-tested.. Modifications were made to the pre-tested instrument (questionnaire), after which the final exercise was carried out. Validity and reliability of the data were conducted by many statisticians, statistical agency, and university scholars before the data was used as the data are for national policy planning. After which it was released to the University of the West Indies, Mona, Data Bank for use by scholars. The data was weighted in order to represent the population of female aged 15 to 49 years in the nation.1 Statistical analyses Data were entered, stored and retrieved using SPSS for Window, Version 16.0 (SPSS Inc; Chicago, IL, USA). Descriptive statistics were performed on particular sociodemographic characteristics of the sample (frequency, mean, standard deviation (SD), and range). All metric variables were tested for normality (age at first sexual debut, crowding, age, and years of schooling). Where skewness was found to be less than 0.5, the variable was used in its current form and a value more than 0.5 was normalized by natural log, or another method. Independent sample t-test was used to examine differences in age at sexual debut between those who frequently attend churches and those who infrequently visit churches and F-statistic was employed for age of respondents by age at sexual debut. Finally, ordinary least square (OLS) regression was used to fit the data because the dependent variable (age at sexual debut) was a
436

continuous one. Stepwise multiple linear regression was used to fit the one outcome measure (age at first sexual debut) by different sociodemographic variables. Thus, only explanatory variables (i.e. statistically significant variables) are shown in Table 16.3. Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final model construction.27 To derive accurate tests of statistical significance, we used SUDDAN statistical software (Research Triangle Institute, Research Triangle Park, NC), and this adjusted for the surveys complex sampling design. A p-value < 0.05 (two-tailed) was used to establish statistical significance. Measures Age at first sexual debut (or initiation or intercourse) was measured based on a respondents answer to the question At what age did you have your first intercourse? Crowding is the total number of persons in a dwelling (excluding kitchen, bathroom and verandah). Age is the number of years a person is alive up to his/her last birthday (in years). Contraceptive method comes from the question Are you and your partner currently using a method of contraception? , and if the answer is yes Which method of contraception do you use? Age at which began using contraception was taken from How old were you when you first used contraception? Area of residence is measured from In which area do you reside? The options were rural, semi-urban and urban (1 = rural, 0 = otherwise; 1 = semi-urban, 0 = otherwise, and urban is the reference group). Currently having sex is measured from Have you had sexual intercourse in the last 30 days? (1=yes, 0 = otherwise). Education is measured from the question How many years did you attend school? Marital status is measured from the following question Are you legally married now?, Are you living with a common-law partner now? (that is, are you living as man
437

and wife now with a partner to whom you are not legally married?), Do you have a visiting partner, that is, a more or less steady partner with whom you have sexual relations?, and Are you currently single? Age at menarche is measured from How old were you when your first period started (first started menstruation)? Gynaecological examination is taken from Have you ever had a gynaecological examination? (1 = yes, 0 = no). Pregnancy was assessed by Are you pregnant now? (1=yes, 0 = otherwise or no). Religiosity was evaluated from the question With what frequency do you attend religious services? The options range from at least once per week to only on special occasions (such as weddings, funerals, christenings et cetera) (1=frequent attendance from response of at least once per week, 0 = otherwise). Subjective social class is measured from In which class do you belong? The options are lower, middle or upper social hierarchy (1 = middle class, 0 = otherwise; 1 = upper class, 0 = otherwise; reference group is lower class). Forced to have sexual relations was assessed from the question Were you forced to have sex at your first intercourse? and the options were yes, no, dont know and refused to answer (1= yes, 0 = otherwise). Age at first sexual debut, age at menarche, age at first contraceptive use, and years of schooling were used as continuous variables. Early sexual debut is having sexual intercourse before the statutory legal age to do so (in Jamaica, this is 16 years old).

Result
Table 16.1 presents the demographic characteristics of the sample, which comprises 7,168 respondents (women who are ages 15-49 years at their last birthday). Most of the women in the survey have been pregnant (84.3%) prior to this study and (4.4%) were pregnant at the time of the study. Only 40.6% of the sample indicated that they had wanted to become pregnant, when
438

they realized they were. The mean age of menarche was 13.5 years (SD = 4.4 years), with the median age of first sexual relations being 16.0 years (Range = 36 years). The mean age at which the sample began using a contraceptive method was 19.8 years (SD = 4.3 years). Twenty-five percentages of women began having sex at 15 years, fifty percentages at 16 years and seventyfive percentages at 18 years. One-half of the sample indicated that they began learning about sex education at 13.0 years (Range: 10, 29 years). The mean age for those who had their first sexual intercourse was 15.2 years (SD = 5.9). One-half of the sample stated that they were dating their partners for 2 years (Range: 0, 15 years) prior to their first sexual encounter. Almost 38% of the sample attended church at least once per week; 19% at least once per month and 7.3% attended church even on special occasions such as christening, wedding, funerals or graduation. Eight-four percent (84%) of those who were married were living with their husbands at the time of this study, five percent (5%) of those who have been pregnant had still births, 12.1% had miscarriages, and 11.4% have been forced to have sexual relations with another person. Fifty-six percentages of the respondents are currently using a contraceptive to prevent pregnancy. The study also shows that the condom was the most prevalent contraceptive method (40.5%) among the respondents. This result was followed by the pill (32.9%), tubal ligation (23.8%) and injection (22.9%). Figure 16.1 provides information on the relations between the respondents and the persons with whom they (respondents) had her first sexual encounter. Majority of the sample indicated that they used a contraceptive method on their first sexual relations (64.1%). These methods include the condom (95.1%); rhythm or knaus-ogino method (2.3%); pill (1.9%);
439

injection and intra-uterine device (0.1%) each. Two percent of the sample who had an abortion did so once (13.6%), twice (2.4%), thrice (0.8%) and four times (0.0%). The reasons given for the abortion were risk to mothers health (22.5%); risk of birth defects (2.9%); financial challenges (29.4%); unwanted pregnancy by mother (12.7%); unwanted pregnancy by partner (4.9%); the absence of a partner (2.0%) and other issues (22.5%). Thirty-five percentage of the respondents indicated that they became pregnant while attending school, of which 28.3% continued their education after the birth of their child. When the respondents were asked How many weeks after _________ birth of [last child] did you resume sexual relations?, 25% of them said 2 weeks, 50% indicated 3 weeks and 75% claimed at most 14 weeks. Two-thirds of the sample used private health care facilities (private clinician, 64.6%; private hospitals, 0.7% and private clinics, 1.3%) when compared with 31.1% of those who used public/government facilities (public hospitals, 8.9%; government clinics, 22.2%). Frequent attendees to church begin having sexual relations on average (mean) at 17.4 years (SD = 3.2) compared to 16.4 years (SD = 2.4 years) for non-frequent attendees t-test = 12.6, P < 0.0001. A significant statistical difference emerged among age at sexual relations of residence of particular geographical areas (F-statistic = 32.4, P < 0.0001). On average rural women began having sexual intercourse at 16.5 years (SD = 2.6 years) compared to 1.7 years (should this be 17.4 years or another year) (SD = 2.9 years) for residence of semi-urban areas and 17.2 years (SD = 3.0) for those in urban zones.

440

Table 16.2 shows information on the age of the respondents and age at sexual debut. Statistical difference was found among the age of respondents and age at sexual debut of the studied population (F statistic = 47.3, P < 0.0001). Table 16.3 examines factors that are associated with the age of first sexual relations of women ages 15 to 49 years in Jamaica. Using multiple regressions analyses, of the 17 variables that were tested in the model, 11 variables emerged as statistically significant predictors of age of first sexual relations (F-statistic [11, 5720] = 176.2, P-value < 0.0001). The factors explained 27.8% of the variability in age of first sexual relations.

Discussion
The sociodemographic related evidence of early sexual initiation has been put forward in this study and shows consonance with the literature. It is realized that sexual intercourse at an early age is usually by someone older and who is outside of a union. The risk associated with this factor is that older male partner presents a greater HIV transmission risk because they are more likely than adolescent men to have had multiple partners; to have had varied sexual and drug use experience and to be infected with HIV.5 Sometimes the young female is persuaded by the their older male perpetrators or partner, from using condom because of varying personal ideologies and are therefore, less likely to use condom at first sexual intercourse (82%),5 unlike the findings of this study (64.1%). This not only result in STIs but also unwanted pregnancy (which affects more than 80 million people worldwide
28

), thus the high possibility of school drop-out, most

times after receiving up to approximately 12 years of formal education (similar to the findings of this study). Where females are persuaded from using condom at first sexual intercourse, this may
441

be explained by the fact that males tend to be more casual about sexual relations and are more willing to emphasize sexual aspects than their female counterparts, who are more likely to romanticize sexual relationships. This view point bears consistency with the findings of this study, whereby drop-outs were more prevalent among those who became pregnant while attending school (35%) when compared to those who continued their education after the pregnancy (28.3%). Other schools of thought postulated that sexual activity and pregnancy among adolescents or teenagers in Jamaica, Guatemala, and Latin America have been thought to be associated with poor education, poverty and other social factors.10, 29,
30

The current findings highlighted that rural women on

average began having sex 8 months earlier than other women (at 16.5 years) that is the age in which they would be in grades 10 and/or 11. Those grades are pivotal for the completion of secondary level education, which means that lower level education will be greater among rural women than those in other geographic areas. It is this lowered age of sexual debut and ignorance of contraception that accounts for higher fertility and unwanted pregnancies among rural women. Research has shown that at least 120 million women would have used contraceptives if information was available.2 Therefore, the lack of knowledge and available options

undermines the right of couples and individuals to exercise control over their fertility and to have children in health and by choice.2 Knowledge about contraception and the various services available regarding its access is considered an obligation of national governments, especially from a human rights perspective.31 In Jamaica, many youths lack accurate sexual health information, especially with regard to the possibility of pregnancy at first intercourse; protection against STIs via the correct use of the correct contraceptives; the effectiveness of oral
442

contraceptives against pregnancy; fallacies relating to contraceptive methods.32 Such asymmetric information result in unintended pregnancies, STIs, and abortion. Where abortion is illegal and access to contraception is limited, more than half of the pregnancies end in abortion.28 Take for instance, in the cases of Chile, Hungary, Russia, Turkey, Czech Republic, abortion rates declined significantly owing to access to modern forms of contraception.28 Similarly, in Canada, access to user-friendly reproductive health services, high quality sex education and the increase use of oral contraception has resulted in a decline in teen pregnancy rate.33 In Jamaica, a research by McNeil concurs with the aforementioned studies that teenage pregnancy fell by 14.6% (from 1997 to 2000) because of sex education programmes, training, counseling, skills training and increased contraceptive use.34 Many scholars view early pregnancies as a potential population problem as this increases the chance of larger family size. This has contributed to 30% birth in islands such as St. Kitts and Nevis, Dominica, St. Lucia and 32% for Jamaica.35 In an effort to avoid poor education or school drop-outs, pregnancies are sometimes interrupted (induced abortion), which is about 60% among the average teenager.14 In South Africa, a study found that 32% of pregnant teenagers complete high school,36 suggesting and agreeing that medical or surgical abortions reduce the probability of poor educational attainment. Another study shows that adolescent girls contribute 55% of all clandestine abortions in Nigeria.37 While abortion still remains a public policy and public health debate, in some countries it is considered a human right (Sweden),
31

legal (Guyana and Haiti

and illegal (Jamaica, Nicaragua and Chile). The reality is Over 19 million women globally resort to unsafe abortion each year, largely among the worlds poorest and most vulnerable women, especially young women,38 indicating that the illegality of abortion does not abate its practices, but it becomes a public health concern. In Jamaica, abortion is considered a serious
443

offence under the Offences Against the Person Act 1973, Section 73, 39 and this goes to reducing their reproductive health choice and open avenues of them seeking the service in unsafe conditions. The reality is, with poverty being greater in rural areas and among females in Jamaica,40 unwanted pregnancies which are arising from ignorance of contraception and earlier sexual initiation means that educational disparity and income inequality if not abated will see a higher fertility, adoption and unsafe abortions among those women. Worldwide, more than half a million women die every year from pregnancy-related causes.2 Many deaths resulted from approximately 20 million unsafe abortions that occur yearly, especially among adolescent girls and young women in developing countries.2 In many developing countries, abortion (if unsafe) is considered a common cause of maternal mortality, hence a serious social problem.31 Nevertheless, a lack of access to safe and legal abortions is an obstacle to their enjoyment of human rights.31 The goal of the World Summit on Social Development (WSSD) Declaration and Programme of Action 1995, the ICPD 1994 and the World Conference on Human Rights (WCHR), Declaration and Programme of Action 1993 is to .reduce maternal mortality and morbidity and greatly reduce the number of deaths from unsafe abortion.2 Women in Jamaica like other Caribbean islands (such as Antigua and Barbuda, Haiti, Guyana, Trinidad and Tobago and Dominica Republic) show a similar age of sexual debut. One in six women in other Caribbean nations between the ages of 15 and 24 became sexually active before the age of 15 years,
14,15

and 1 in 4 women in Jamaica begin at 15 years, and this is even

lower among non-frequent religious women (14.7 years). The current research shows a marginal difference in the Crawford, McGrowder and Crawfords16 which had that the mean age of sexual
444

debut for female was 15.8 years in Jamaica. Disaggregating the age at which women aged 15-49 years had their first sexual intercourse, we found that the mean age at sexual debut was lowest for women aged 15-19 years old (15.2 years (SD = 1.6)) compared to other aged women. While Crawford, McGrowder and Crawford found that much of earlier sexual debut was out of violence, this research disagrees as we found that only 11.4% of those who have had earlier sexual intercourse were raped,16 which indicates that the majority of first sexual debut was a consensual act although by statutes all sexual relations below 16 years is a rape.41 A study in New Zealand found that 7% of first sexual intercourse was forced,
42

which is marginally

lower than that of Jamaica. The time difference may account for this dissimilarity as Dickson et al.s work
42

was in 1993-1994, while the current paper used data for 2004. Moreover, First

sexual intercourse almost always took place outside of a formal union3 and with older men (for the females).4 We found that the majority of first sexual relations took place with a boy friend in a visiting relationship with the respondent. Based on the foregoing, The timing of sexual debut among adolescents is influenced by a wide range of factors including: age, gender, poverty, family structure, educational level, pubertal timing, socio-economic status, self-efficacy, peer influences, religiosity, knowledge and perceived risk of sexually transmitted infections, parenting practices and parental supervision, community, media and health inequalities.43 Outside of those factors which explain early first coitus in the developing nations, particularly the Caribbean is the masculine orientation and culture.44 Research demonstrates that the role of culture in the socialization of children is critical to fashioning the adult, and as soon as females begin to grow breast and to menstruate there is a

445

perception of womanhood. During this growth and development process, the female adolescents physiology of reproduction sometimes begins in order to establish womanhood. The validity of recall of age of first sexual intercourse has been established by a group of researchers in 1997. They found that the test-retest correlations for the recall of age at first sexual relations was 0.85 for females and 0.91 for males,45 which indicates the validity of usage of recall data to measure the phenomenon. Hence, there is legitimacy in the use of cross-sectional survey data to examine age at first sexual intercourse in Jamaica, and the findings therefore provide invaluable insight into the attitude, behaviour and practices of women in Jamaica and those factors which are associated with age at first sexual debut. The current paper, therefore, have added variables to the literature: gender, ethnicity, income, mothers education, family structure, interpersonal relationship and other socioeconomic conditions are associated with age at first sexual intercourse.46-48 It also concurs with other studies that sexual activity is no longer strongly predicted by marriage49-52 as the majority of women who had their first sexual experience, engaged in such activity with a boy friend, stranger or mere acquaintance (87 out of every 100 women). With the low condom usage on first sexual intercourse found in this research, young women are open to the risk of STI, pregnancy and psychological challenges of early sexual relations, and therefore this justifies the rationale for wanting to modify sexual practices of adolescents.53,54 While the current reality of age at first sexual intercourse in Jamaica appears low, this is equally the case in other nations as we found that 80% of a recent cohort of youths who had sex did so become 20 years.55 The image that is embodied in these figures is the sexual complaints which are likely to result from the adult sexual decision that will be taken by adolescents, 56, 57
446

and the possible life changing situations that are likely to result from a sexual encounter. Clearly, the current public health intervention programmes in Jamaica, as well as other geopolitical areas in the world are not reaching adolescents as they are have committed (under the ICPD 1994), and by extension have failed to reduce the lowering of age of first sexual intercourse. With the factors which emerged from the current paper as accounting for age at first sexual intercourse, as well as those from other studies,
58-60

like McGrath and colleagues,

61

we believe that a

multisectoral approach is needed to address these growing public health and legislative problems not as a single variable (age at first sexual intercourse) but other factors that are purported in the reviewed studies, 62-64 as well as the evidentiary support of Jamaica. Within the context of lowered age at first sexual intercourse of Jamaican women as well as the association between forced sexual relations and early age of sexual debut, 65 this would be contributing to the current public health problems of teenage pregnancies, high fertility, STIs, increased maternal and child mortality, and psychologically challenged young people as they undergo the difficulty of the experience.66-68 Clearly, this study highlighted the finding that the average age at first sexual debut for Jamaican women (median age was 16.0 years) was lower than that of women in rural South African (median age was 18.5 years)69 and eastern Zimbabwe (median age was 18.5 years).70 A study, using European women ages 16-44, found that the average age of first sexual debut was less than 16 years, and offers little solace for public health practitioners in Jamaica.23 Although South Africa had the highest HIV infection rate in the world69 and an age at first sexual debut lower than that of Jamaican women, public health specialists need to use the current findings to ensure that the premarital sexual relations, inconsistent condom use and STI infections, especially HIV, do not reach the levels of those in South Africa as previous studies have shown
447

the association between age at first sexual intercourse and having an STI.70 The rationale for this prescriptive recommendation for public health specialists is embedded in the association between early sexual debut and sexually transmitted infections as well as evidence which shows that STIs are a gateway to complications such as pelvic inflammatory infections, infertility, ectopic pregnancy, fetal abnormality and HIV/AIDS.70-72 Those are not the only issues of concerns at age at first sexual debut as many studies have shown that gender, illicit drugs, age at menarche, religiosity, area of residence and other factors are associated with this

phenomenon.19,20,22,25,69,70,73 This study concurs with the literature and added more variables such as age at contraceptive use, forced sexual relations, employment status, shared sanitary convenience, area of residence, and marital status, indicating that multi-variables are associated with age at first sexual initiation of Jamaican women.

Conclusion
Public health policies have failed to effectively increase the age at first sexual intercourse for women in Jamaica. This study shows that a multisectorial philosophy to the intervention is needed in order to address the multidimensional nature of the factors which are associated with age at first sexual debut. Sexual intercourse is commonly initiated in the adolescence years, and with the increased risk of sexually transmitted infections, teenage pregnancy and adoption with early sexual initiation, the public health consequences will be dire if they are felt unabated or the age at sexual debut allowed to fall lower than current value. Disclosures The author report no conflict of interest with this work.
448

Disclaimer
The researcher would like to note that while this study used secondary data from the Reproductive Health Survey, none of the errors in this paper should be ascribed to the National Family Planning Board, but to the researcher.

Acknowledgement
The author thank the Data Bank in Sir Arthur Lewis Institute of Social and Economic Studies, the University of the West Indies, Mona, Jamaica for making the dataset (2002 Reproductive Health Survey, RHS) available for use in this study, and the National Family Planning Board for commissioning the survey.

449

References
[1] Jamaica, National Family Planning Board (NFPB). Reproductive Health Survey, 2002. Kingston: NFPB; 2005 [2] Family Care International. Commitments to Sexual and Reproductive Health and Rights for All. Framework for Action. New York, NY: Family Care International; 1995 [3] US Department of Health and Services (2006). Gender Differences in Reproductive Health. Department of Health and Human Services, Centre for Disease Control, USAID and Jamaica National Family Planning Board. [4] Sexual Initiation. American Sexual Behaviour. http://www.newstrategist.com/productdetails/Sex.SamplePgs.pdf (Accessed on April 1, 2010). [5] Miller KS, Clarke LF, Moore JS. Sexual initiation with older male partners and subsequent HIV risk behavior among female adolescents. Family Planning Perspectives; 29, 1997:212-214. [6] Drayton VLC. Contraceptive use among Jamaican teenage mothers. Pan Am H Public Health 2002; 11(3):150-157. [7] Jagdeo T. The dynamics of adolescents fertility in the Caribbean. St. Johns, Antigua: Caribbean Family Planning Affiliation; 1992. [8] Williamson LM, Parkes A, Wight D, Petticrew M, Hart GJ. Limits to modern contraceptive use among young women in developing countries: A systematic review of qualitative research. Reproductive Health 2009; 6:3 [9] Warren CW, Powell D, Morris L, Jackson J, Hamilton P. Fertility and family planning among young adults in Jamaica. Int Family Planning Perspectives 1988; 14(4):137-141. [10] Eggleston E, Jackson J and Hardee K. Sexual attitudes and behavior among young adolescents in Jamaica. Guttmacher. International Family Planning Perspectives; 25(2), 1999. [11] Henry-Lee A. Womens reasons for discontinuing contraceptive use within 12 months: Jamaica. Reproductive Health Matters 2001;9(17):213-220. [12] World Health Organization (WHO). Reproductive health research at WHO: A new beginning. Biennial report 1998-1999. Geneva: WHO; 2000. [13] World Health Organization (WHO). World health statistics, 2009. Geneva: WHO; 2009. [14] Rawlins J. Teenage Pregnancy: A study in three communities in Trinidad and Tobago. Paper presented at the Caribbean Health Research Conference 2007, Jamaica. [15] Jamaica Observer. Study shows Jamaican Girls Encounter Violent Sexual Relationships. Jamaica Observer (15 April 2009). [16] Crawford TV, McGrowder DA, Crawford A. Access to contraception by minors in Jamaica: a public concern. North Am J of Med Sci 2009; 1(5):247-255. [17] Fatusi AO, and Blum, RW. Predictors of early sexual initiation among a nationally representative sample of Nigerian adolescents. BMC Public Health; 8, 2008. [18] Santelli JS, Kaiser J, Hirsch L, Radosh A, Simkin L, Middlestadt S: Initiation of sexual intercourse among middle school adolescents: the influence of psychosocial factors. J Adolesc Health 2004, 34:200-208. [19] Penfold SC, Teijlingen ERV, Tucker JS. Factors associated with self-reported first sexual intercourse in Scottish adolescents. BMC Research Notes 2009; 2:42.

450

[20] Rosenthal DA, Smith AMA, De Visser R. Personal and social factors influencing age at first sexual intercourse. Archives of Social Behavior 1999; 28(4):319-333. [21] Statistical Institute of Jamaica (STATIN). Demographic statistics, 2007. Kingston: STATIN; 2008. [22] Louie KS, de Sanjose S, Diaz M, et al. Early age at first sexual intercourse and early pregnancy are risk factors for cervical cancer in developing countries. Br J Cancer 2009; 100(7):119-7. [23] Population Reference Bureau. Preventing cervical cancer worldwide. Washington, D.C: Population Reference Bureau; 2004. [24] Pan American Health Organization (PAHO). A situational analysis of cervical cancer in Latin American and the Caribbean. Washington, D.C: PAHO; 2004. [25] Kahn JA et al., Mediators of the association between age of first sexual intercourse and subsequent human papillomavirus infection, Pediatrics, 2002, 109(01). [26] Mardh PA, Creatsas G, Guaschino S, et al. Correlation between early sexual debut, and reproductive health and behavioral factors: a multinational European study. Eur J Contracept Reprod Health Care 2000; 5:177-82. [27] Polit DF. Data analysis and statistics from nursing research. Stamford: Appleton & Lange Publisher; 1996. [28] Population Action International. Contraceptive use helps reduce the incidence of abortion. Fact Sheet. Washington DC., Population Action International. [29] Alan Gunmacher Institute. Women and reproductive health in Latin America and the Caribbean. Women, families and the future. NY: The Institute, 1994. [30] Castro MT. Guatemala: Encuesta Nacional de Salud Materno Infantil, 1995. Demographic and Health Survey. MD: Macro International, 1996. [31] Ministry of Foreign Affairs. Swedens International Policy on Sexual and Reproductive Health and Rights. Stockholm: Ministry of Foreign Affairs, Sweden; 2006. [32] Ward M The reproductive and sexual health of Jamaican youth. Advocates for Youth. Washington DC., Advocates for Youth; 2001. [33] McKay A. Adolescent sexual and reproductive health in Canada: A report card in 2004. The Canadian Journal of Human Sexuality; 13(2), 2004:67-81. [34] McNeil P. Coping with teenage pregnancy. In: Morgan O. ed. Health issues in the Caribbean 2005; pp. 51- 57. [35] Rawlins J. Parent-Child interaction and teenage pregnancy. Master of Science Degree Thesis. University of the West Indies, Jamaica; 1981. [36] Cooper D, Dickson K, Blanchard K, et al. Medical abortion: The possibilities for introduction in the public sector in South Africa. Reproductive Health Matters 2005;13:35-43. [37] Bankole A, Oye-Adeniran BA, Singh S, et al. Unwanted pregnancy and induced abortion in Nigeria: causes and consequences. New York: Guttmacher Institute; 2006 [38] Reproductive Health Matters. Consensus statement. Medical abortion: Expanding access to safe abortion and saving womens lives. Reproductive Health Matters 2005;13(26):11-12. [39] Jamaica Laws. Offences Against the Person Act, 1973. Kingston: Jamaica Government Printery;1973. [40] Planning Institute of Jamaica (PIOJ), Statistical Institute of Jamaica (STATIN). Jamaica Survey of Living Conditions, 2007. Kingston: PIOJ, STATIN: 2008.
451

[41] Jamaica Laws. Child Care and Protect Act, 2004. Kingston: Jamaica Government Printery; 2004. [42] Dickson N, Paul C, Herbison P, Silva P. First sexual intercourse: age, coercion, and later regrets reported by a birth cohort. BMJ 1998;316:39-30. [43] Lammers C, Ireland M, Resnick M, Blum R: Influences on adolescents decision to postpone sexual intercourse: a survival analysis of virginity among youths aged 13 to 18 years. J Adolesc Health 2000, 26:42-48 [44] Chevannes B. Learning to be a man: Culture, socialization and gender identity in five Caribbean communities. Kingston, Jamaica: The Univer. of the West Indies Press; 2001. [45] Dunne MP, Martin NG, Statham DJ, Pangan T, Madden PA, Heath AC. The consistency of recall age at first sexual intercourse. J of Biosocial Science 1997;29:1-7. [46] Upchurch DM, Levey-Storms, Sucoff CA, Aneshensel CS. Gender and ethnic differences in the timing of first sexual intercourse. Family Planning Perspectives, 1998, 30(3):121127. [47] DeLamater J. The social control of sexuality. Annual Review of Sociology, 1981; 7: 263 290. [48] Udry JR and Campbell BC, Getting started on sexual behavior, in: Rossi AS, ed., Sexuality Across the Life Course, Chicago: University of Chicago Press; 1994. [49] DEmilio J and Freedman ES, Intimate Matters: A Historyof Sexuality in America, New York: Harper and Row, 1988. [50] Nathanson CA, Dangerous Passage: The SocialControl of Sexuality in Womens Adolescence, Philadelphia:Temple University Press, 1991. [51] Hogan DP and Astone NM, The transition to adulthood, Annual Review of Sociology, 1986, No. 12, pp.109130. [52] Miller BC and Heaton TB, Age at first sexual intercourse and the timing of marriage and childbirth, Journal of Marriage and the Family, 1991, 53(3):719732. [53] Brindis CD et al., Complex Terrain: Charting a Course of Action to Prevent Adolescent Pregnancy, San Francisco: Center for Reproductive Health Policy Research, Institute for Health Policy Studies, University of California, 1997. [54] Moore KA et al., Adolescent Sex, Contraception, and Childbearing:A Review of Recent Evidence, Washington, DC: Child Trends, 1995. [55] Henshaw SK, U.S. teenage pregnancy statistics, NewYork: The Alan Guttmacher Institute, 1994. [56] Woo JST, Brotto LA. Age of first sexual intercourse and acculturation: Effects on adult sexual responding. The J of Sexual Medicine 2008;5:571-582. [57] Slaymaker, E, Bwanika, J B, Kasamba, I, Lutalo, T, Maher, D, Todd, J. Trends in age at first sex in Uganda: evidence from Demographic and Health Survey data and longitudinal cohorts in Masaka and Rakai. Sex. Transm. Infect. 2009; 85: i12-i19. [58] Gupta N, Mahy M. Sexual initiation among adolescent girls and boys: trends and differentials in sub-Saharan Africa. Arch Sex Behav 2003; 32:4153. [59] Hallett TB, Lewis JJ, Lopman BA, et al. Age at first sex and HIV infection in rural Zimbabwe. Stud Fam Plann 2007; 38:110. [60] Fatusi AO, Blum RW. Predictors of early sexual initiation among a nationally representative sample of Nigerian adolescents. BMC Public Health 2008; 8. [61] McGrath N, Nyirenda M, Hosegood V, Newell M-L. Age at first sex in rural South Africa. Sex Transm Infect 2009 85: i49-i55
452

[62] Hargreaves J, Boler T. Girl power: the impact of girls education on HIV and sexual behaviour. Johannesburg: ActionAid International, 2006. [63] Birdthistle IJ, Floyd S, Machingura A, et al. From affected to infected? Orphanhood and HIV risk among female adolescents in urban Zimbabwe. AIDS 2008; 22:75966. [64] Gregson S, Nyamukapa CA, Garnett GP, et al. HIV infection and reproductive health in teenage women orphaned and made vulnerable by AIDS in Zimbabwe. AIDS Care 2005; 17:78594. [65] Baumgartner JN, Geary CW, Tucker H, Wedderburn M. The influence of early sexual debut and sexual violence on adolescent pregnancy: A matched case-control study in Jamaica. Guttmacher Institute; 35(1), 2009. [66] Bachanas PJ, Morris MK, Lewis-Gess JK, Sarett-Cuasay, EJ, Sirl, K, Ries, JK, Sawyer MK. Predictors of Risky Sexual Behavior in African American Adolescent Girls: Implications for Prevention Interventions. Journal of Pediatric Psychology; 27(6), 2002: 519-530. [67] Caminis A, Henrich C, Ruchkin V, et al. Psychosocial predictors of sexual initiation and high-risk sexual behaviors in early adolescence. Child and Adolescent Psychiatry and Mental Health 2007; 1:14. [68] Majaraj RG, Nunes P, Renwick S. Health risk behaviours among adolescents in the Englishspeaking Caribbean: a review. Child and Adolescent Psychiatry and Mental Health 2009; 3:10. [69] McGrath N, Nyirenda M, Hosegood V, Newell M-L. Age at first sex in rural South Africa. Sex Tansm Infect 2009; 85(suppl 1):49-55. [70] Kaestle CE, Halpern CT, Miller WC, Ford CA. Young age at first sexual intercourse and sexual transmitted infections in adolescents and young adults. Am J Epidemiol 2005; 161:774780. [71] Moodley P, Sturm AW. Sexually transmitted infections, adverse pregnancy outcome and neonatal infection. Semin Neonatol 2000; 5:255-69. [72] Sorvillo F, Smith L, Kerndt P, et al. Trichomonas vaginalis, HIV, and African-Americans Emerg Infect Dis 2001;7:927-32. [73] Cremin I, Mushati P, Hallett T, et al. Measuring trends in age at first sex and age at marriage in Manicaland, Zimbabwe. Sex Transm Infect 2009; 85(Suppl 1):34-40.

453

Table 16.1. Demographic characteristic of studied population, n = 7, 168 n % Characteristic Shared sanitary convenience with other household No 5907 82.9 Yes 1219 17.1 Employment status Employed 3025 42.2 Unemployed (including students) 4143 57.8 Main source of financial support Partner 4129 57.6 Other 3039 42.4 Marital status Legally married 1542 21.5 Common-law 1733 24.2 Visiting 1959 27.3 Not currently in union 1934 27.0 Currently pregnant Yes 288 4.4 No 6219 94.6 Ever been pregnant Yes 5301 84.3 No 985 15.7 Forced to have sex Yes 747 11.4 No 5707 86.8 Health conditions Diabetes 284 12.2 Anemia 438 18.8 Heart disease 94 4.0 Pelvic inflammatory disease 125 5.4 Urinary tract infection 800 34.3 Asthma 587 25.0 Hepatitis B 6 0.3 Area of residence Urban 1144 16.0 Semi-urban 2079 29.0 Rural 3945 55.0 Socioeconomic class Lower 1705 23.8 Middle 3079 43.0 Upper 2384 33.2 No. of pregnancies that resulted in live births median (range) 2.0 (0, 14) Years of schooling mean (SD) 13.0 years (3.0 years) Age mean (SD) 31.3 years (9.3 years) Age at sexual debut median (Range) 16.0 years (29 years; max age 36 years)

454

Figure 16.1. Person with whom respondents had their first sexual relations

455

Table 16.2. Age cohort of respondents by age at sexual debut Age at sexual debut (in years) Age cohort of respondents (in years) Mean (SD1) 15 19 15.2 (1.6) 20 24 16.2 (2.0) 25 29 16.8 (2.4) 30 34 17.1 (2.9) 35 39 17.2 (3.1) 40 44 17.2 (3.2) 45 49 17.1 (3.0) Sample 16.8 (2.8) 1 SD denotes standard deviation F statistic = 47.3, P < 0.0001

456

Table 16.3. Multiple linear regression analyses: Explanatory variables of age at first sexual debut, n = 5,732 Explanatory variable Constant Age began using contraceptive method Years of schooling Lower class (reference group) Upper class Forced sexual relations (1= yes) Frequent church attendance (1= once or less per week) Crowding Employment status (1= employed) Age of first menarche Shared sanitary convenience (1=yes) Married or common-law union Urban area (reference group) Rural
NA Not applicable

Coefficient 8.377 0.266 0.166

CI (95%) 7.852 - 8.903 0.250 - 0.283 0.141 - 0.190

R2 NA 0.179 0.048

0.560 -0.650 0.511 0.409 0.347 0.048 -0.325 -0.175

0.385 - 0.735 -0.820 - 0.481 0.364 - 0.659 0.240 - 0.579 0.206 - 0.489 0.027 - 0.069 -0.504 - 0.147 -0.315 - 0.035 -0.856 - 0.452

0.021 0.009 0.006 0.005 0.003 0.003 0.002 0.001

-0.654

0.001

457

17
Current use of contraceptive method among women in a middle-income developing country
Paul A. Bourne, Christopher A.D. Charles, Maureen D. Kerr-Campbell & Cynthia G. Francis Jamaica is a middle income developing country with an increasing population and public resource constraints. Therefore, reproductive health issues are of critical importance to Jamaica. This study examines the use of contraceptives among women and the factors that influence these women to use contraceptives. The majority of participants used some method of contraception (64%). The most popular method of contraception was a condom (32%). The multivariate analysis suggests that the explanatory variables for the method of contraception used are age (OR = 0.98, 95% CI: 0.98-0.99), social class (OR = 0.83, 95% CI: 0.73-0.95), being in a relationship (OR = 3.35, 95% CI: 2.80-4.02), the rural-urban dichotomy (OR = 1.16, 95% CI: 1.02-1.32), being currently pregnant (OR = 0.01, 95% CI: 0.00-0.02), currently having sex (OR = 2.29, 95% CI: 1.95-2.70), the number of partners (OR = 1.85, 95% CI: 1.57-2.17), the age at which the women began using a contraceptive (OR = 0.99, 95% CI: 0.98-1.00) and crowding (OR = 1.40, 95% CI: 1.21-1.60). The findings are far-reaching and can be used to aid policy formulation and intervention.

Introduction
This article aims to explore the use of contraceptives among women in Jamaica in order to correct a paucity of information in the academic literature, and to provide information for policymakers, public health practitioners and educators. The rationales which influence this research are (1) the lack of a comprehensive study on contraceptive use, and (2) the public health
458

concerns which have arisen in the past decade. The mean age of first sexual intercourse has been falling since 1997 and this is coupled with (1) increased contraceptive use, (3) increased teenage pregnancy, (3) increased premarital sexual relations, and (4) increased HIV in the young adult population, as well as (5) the piecemeal approach to the study of contraceptive use in Jamaica. It should be noted that some adolescent and young adult females who engaged in unplanned sexual intercourse underestimated the risk of pregnancy and did not use contraceptives consistently. The prevention of pregnancy was deemed to be the responsibility of the women. Their decisionmaking was strongly influenced by friends, family and social norms. The most important forms of support these females received were from partners and parents1. College women having their first experience of sexual intercourse did so at an older age than men. Some 61% of the women used an unreliable method or no contraceptive at all. The most frequent reason reported by these women for non-use of contraceptives in their first sexual intercourse was that it was unplanned. The lack of knowledge and inaccessible sources of contraception also influenced the womens use of contraceptives.2 The use of contraceptives is also related to the length of birth intervals. A review of the literature connecting the length of birth intervals to the use of contraceptives reveals mixed findings. However, the use of contraceptives is a protective factor against short birth intervals.3 The variables of the theories of planned behaviour, coupled with family planning self efficacy, accounted for 65% of the outcome for intent to use oral contraceptives, and 27% of the variance in behaviour among women.4 A study of womens views on family planning services suggests that they find several factors important. These are, the providers showing empathy and respecting the womens autonomy, the provision of personalized care and comfort of the women, the information provided, the technical quality of care and the organization of the service.
459

Another important factor is the providers ability and willingness to communicate in the language of the women5. Moving from the clinic to their homes, the views of these women sometimes have little influence, even with the encouragement of their female friends. The encouragement wives receive from their social networks about the use of contraceptives does not influence the use of contraceptives by their husbands.6 However, some women who were experiencing domestic violence in their relationship with men stated that violence was not an important factor influencing their use of contraception.7 Illness also influences womens use of contraceptives. Women with bipolar disorder use contraception sub-optimally.8 Similarly, women who are depressed are more likely to choose an ineffective method of contraception. These women need contraceptive counselling which is tailored to improve their decision-making and choice of contraception.9 A review of the studies dealing with oral contraceptives and multiple sclerosis (MS) suggests that the use of oral contraceptives does not increase the risk of MS. On the contrary, it may delay the onset of the disease.10 Although knowledge of contraception is high among HIV sero-discordant couples, the use of contraceptives is low. Gender difference is an important factor, because many women engage in the clandestine use of contraceptives.11 The use of contraceptives among women is not only related to illness but also to religious factors. Religions differ in their dictates about contraception, which influence its use among religious people shaped by the history and politics of their particular religion.12 Given the range of factors outlined above, dealing with womens use of contraceptives, the promotion of reproductive health by the international donor community sometimes does not work. This failure occurs because the donor and the policy-makers of the target country define
460

reproductive health differently. The priorities and values of the donors and policy-makers are at odds, because cultural factors drive contraceptive use away from the expected outcome of the donors13. The purpose of this article is to understand the methods of contraception used by Jamaican women, and some of the factors that influence them to use these methods of contraception. The data collection method used in the current paper is outlined below.

Methods
Since 1997, the National Family Planning Board (NFPB) has been collecting information on women (ages 15-49 years) in Jamaica regarding contraception usage and/or reproductive health. In 2002, the Reproductive Health Survey (RHS) collected data on women ages 15-49 years and men 15-24 years. The current paper extracted the sample of only women (ages 15-49 years) given the nature of the research. The sample was 7,168 women, representing a response rate of 91.8%. Stratified random sampling was used to design the sampling frame from which the sample was drawn. Using the 2001 Census sector (or sampling frame), a three-stage sampling design was used. Stage 1 was the use of a selection frame of 659 enumeration areas (or enumeration districts, EDs). This was calculated based on probability proportion to size. Jamaica is classified into four health regions. Region 1 consists of Kingston, St. Andrew, St. Thomas and St. Catherine; Region 2 comprises Portland, St. Mary and St. Ann; Region 3 is made up of Trelawny, St. James, Hanover and Westmoreland, with Region 4 being St. Elizabeth, Manchester and Clarendon. The 2001 Census showed that region 1 comprised 46.5% of Jamaica, compared to Region 2, 14.1%; Region 3, 17.6% and Region 4, 21.8%. 14

461

Stage 2 saw the clustering of households into primary sampling units (PSUs), with each PSU constituting an ED, which in turn consisted of 80 households. The previous sampling frame was in need of updating, and so this was carried out between January 2002 and May 2002. The new sampling frame formed the basis upon which the sampling size was computed for the interviewers to use. Stage 3 was the final selection of one eligible female this was done by the interviewer on visiting the household. The Statistical Institute of Jamaica (STATIN) provided the interviewers and supervisors, who were trained by McFarlane Consultancy to carry out the survey. The interviewers administered a 35-page questionnaire. The data collection began on Saturday, October 26, 2002 and was completed on May 9, 2003. The data was weighted in order to represent the population of women ages 15 to 49 years in the nation.14 Statistical methods We used the Statistical Packages for the Social Sciences (SPSS) for Windows, Version 16.0 (SPSS Inc; Chicago, IL, USA). Frequencies and means were computed on the basis of sociodemographic characteristics, health conditions, pregnancy, Pap smears, gynaecological examinations and reasons for choices. We also performed 2 tests to compare associations, in particular sociodemographic variables, contraception, pregnancy, and gynaecological

examination. Stepwise multiple logistic regressions were used to analyze factors that explained gynaecological examinations undergone in the last 12-month period, and Pap smear tests done during the same period. Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final model construction.15 To derive accurate tests of statistical significance, we used SUDDAN statistical
462

software (Research Triangle Institute, Research Triangle Park, NC), and this was adjusted for the surveys complex sampling design. Measure Crowding is the total number of persons in a dwelling (excluding kitchen, bathroom and verandah). Age is the number of years a person is alive up to his/her last birthday (in years). Contraceptive method comes from the question Are you and your partner currently using a method of contraception? , and if the answer is yes Which method of contraception do you use? Age at which began using contraception was taken from How old were you when you first used contraception? Area of residence is measured from In which area do you reside? The options were rural, semi-urban and urban. Currently having sex is measured from Have you had sexual intercourse in the last 30 days? Education is measured from the question How many years did you attend school? Marital status is measured from the following question Are you legally married now?, Are you living with a common-law partner now? (that is, are you living as man and wife now with a partner to whom you are not legally married?), Do you have a visiting partner, that is, a more or less steady partner with whom you have sexual relations?, and Are you currently single? Age at first sexual intercourse is measured from At what age did you have your first intercourse? Gynaecological examination is taken from Have you ever had a gynaecological examination? Pregnancy was assessed by Are you pregnant now? Religiosity was evaluated from the question With what frequency do you attend religious services? The options range from at least once per week to only on special occasions (such as weddings, funerals, christenings et cetera). Subjective social class is measured from In which class do you belong? The options are lower, middle or upper social hierarchy.

463

Analytic Model Using logistic regression, this study seeks to examine factors associated with the method of contraception usage among women in Jamaica. Different social factors influence womens choices and their decision to use a method of contraception, and this study used Grossmans model16 which established the use of econometric analysis to determine the use of health demand. Grossmans model has been modified and used by many scholars to examine health, health outcome and other health-related issues. The current research will use the theoretical framework of Grossmans econometric analysis to examine factors associated with the method of contraception usage among women ages 15-49 years in Jamaica. The variables used in this econometric model are based on the literature as well as the dataset. We will test the hypothesis that the methods of contraception usage among women ages 15-49 years are determined by particular sociodemographic variables (Equation [1]). Cwi = f(Ai, EDi, Ui, SSi, ARi, Pi, Fi, GNi, ASi, Si, Ni, Ri, Ki, Mi, Wi, Ti i) Eqn [1]

where Cwi denotes method of conception usage among women i, Ai is age of woman i, EDi represents educational level of woman i, Ui, means employment status of woman i, SSi is social class of woman i, ARi indicates area of residence of woman i, Pi denotes current pregnancy status of woman i, Fi is forced to have sex (woman i), GNi means gynaecological examination in the last 12 months, woman i, ASi is age of first sexual intercourse of woman i, Si represents currently having sex (woman i), Ni is number of sexual partners of woman i, Ri denotes religiosity of woman i, Ki woman is currently in a sexual union i, Mi denotes age of first menstruation of woman i, Wi represents crowding in household of woman i, Ti denotes age at which contraceptive use for woman began i, and the parameter i is the models error term.
464

Using the data to test the hypothesis (Equation [1]), from the logistic regression analyses, we can write equation [2] to represent the function that explains the method of contraception for women ages 15-49 years in Jamaica. Cwi = f(Ai, SSi, ARi, Pi, Ki, Wi, Ni, Si, Ti, i) Eqn [2]

To make more sense of the function (Equation [2]), we can rewrite it into an equation (Equation [3]):
Log (P/1-P) = + 1Ai + 2SSi + 3ARi + 4Pi + 5Ki + 6Wi + 7Ni + 8 + 9Si + 10Ti + i Eqn [3]

Where P denotes the probability of currently using a method of contraception and 1- P is the probability of currently not using a method of contraception, represents the constant, 1-10 means the coefficient of each variable from 1 to 10. The predictive power of the model was tested using the omnibus test of model and Hosmer and Lemeshows17 technique was used to examine the models goodness of fit.

Result
Table 17.1 presents sociodemographic information on the sample. The sample was 7,168 women ages 15 to 49 years, and most of them were currently using a method of contraception (64%). Currently, 4.4% of the sample was pregnant and 84.3% had previously been pregnant. Almost 16% had at least one miscarriage, 2.2% at least one abortion and 5.4% at least one stillbirth. The mean age of the sample was 31.0 years (SD = 9.3 years). A detailed description of the age cohort of the sample revealed that 13.8% of the women were 15-19 years; ages 20-24 years, 13.1%; ages 25-29 years, 16.4%; 30-34 years, 18.3%; 35-39 years, 16.2%; ages 40-44 years, 12.8% and ages 45-49 years, 9.4%. Half of the sample began using a method of contraception at 19 years
465

(range = 33 years: 11, 44), and 4.5% indicated that they desired to be pregnant sometime in the future. Almost 62% of the respondents indicated that they had asked their partners to use a condom, 20.1% of the women reported that they refused to do so. On the other hand, 5% of women said that they insisted that their partners did not use a condom during sexual intercourse. Twenty-three percent of the respondents indicated that they had had multiple partners. Of those with non-steady partners, 24% provided information on the frequency of use of condoms with this/these person/s: Forty-nine percent of them indicated that they always used a condom, 26.1% claimed most times, 0.7% said seldom and 23.9% remarked never. Regarding women with a steady partner, 44.4% indicated that they always used a condom with their partners, 48% remarked most times, 6.0% said seldom and 0.2% reported that they had never done so. Only 2.4% of the women in the sample were sex workers (being paid for sex money or goods in exchange for sex), and 9.1% said that they had done this more than two times in their lives. Twenty-six percent of the sex workers indicated that they began while they were in school, and 57% said they commenced after leaving school. When the respondents were asked Are you and your partner currently using a method of contraception or doing something to prevent pregnancy, 63.8% indicated yes. Of those who responded to the method of contraception, it was revealed that most respondents used a condom (62%) followed by the pill (14.4%), female tubal ligation (10.4%), and injection (10%). This question was followed by Are you and your partner also using a second method of contraception, and to this 14.6% indicated yes. The methods were withdrawal (65.3%), rhythm, calendar or Billings, 26.5%, pill (2.1%), diaphragm (4.1%) and other (2.0%). When the sample
466

was asked Are you and your partner also using a second method at the same time for either sexually transmitted disease prevention or contraception, 14.6% indicated yes. The methods were withdrawal (65.3%), condom (7.5%), and other. Table 17.2 presents information on particular demographic characteristics of the sample by age group. Table 17.3 presents information on frequency of condom use (with a non-steady partner) by age group. Table 17.4 presents information on frequency of condom use (with a steady partner) by age group. Multivariate analyses Table 17.5 provides information on factors that explain the method of contraception usage of women ages 15 to 49 years. Using stepwise logistic regression analyses, eight variables emerged as statistically significant variables of women ages 15-49 years who are currently using a method of contraception. Women (ages 15-49) who are in the upper class are 17% less likely to use a method of contraception in reference to those in the lower class (OR = 0.83, (95% CI: 0.730.95). The older the women become, they are 2% less likely to use a method of contraception (OR = 0.98, 95% CI: 0.98 0.99), and if they are pregnant they are 99% less likely to use a method of contraception. The model had statistically significant predictive power (model 2 (df = 9) = 1684.75, P-value < 0.0001; Hosmer and Lemeshow goodness of fit 2 = 2.87, P = 0.94), and correctly classified 78.5% of the sample (Table 17.5).

Limitations of the study


One of the fundamental limitations of this study is the cross-sectional nature of the data collection. A cross-sectional study cannot be used to establish causality or predictability, and the results can change with time. Hence, although social policy formulation relies on this research
467

design, policy-makers should be cognizant of the aforementioned issues in designing interventions and strategic frameworks. Despite those limitations, cross-sectional data design is still a good way to collect social science data on a population.

Discussion
The current paper found that although 64% of the sample indicated that they or their partner used a method of contraception, consistency of use among those with a steady partner was relatively low (always, 44.4%; most times, 48.0%) and 73.9% of the respondents indicated that they had never had a non-steady partner. Of those who had a non-steady partner, 49.3% consistently used a condom and 23.9% indicated that they had never used a condom. Almost 60% of the respondents indicated that they had had sex in the last 30 days, and 58% were primarily financially supported by their partner(s). Some 5% of the respondents stated that they desired to become pregnant. Current methods of contraception used by the female or her partner were explained by age of respondent, subjective social class, whether or not in a sexual union, area of residence, currently pregnant, currently having sex (in the last 30 days), number of sexual partners, age at which individual began using contraception and crowding. In 1997, statistics revealed that the prevalence of women currently using contraceptives in Jamaica was 50.3%
21

and this increased to 64% in 2007. The majority of women reported

using a contraceptive, which is a very high rate of usage for a developing country. Using data from Kenya, Tanzania, and Trinidad and Tobago, Norman [22] found that only 19% reported consistently using a condom. In another research, using a sample of 212 respondents from clinics in Montego Bay (Jamaica) who had sexually transmitted infections, the study [23] found that 43% reported using a condom the last time they had sexual intercourse. The current paper
468

revealed a higher consistency prevalence of condom usage than the aforementioned studies, and showed that almost 92.4% of Jamaican women used a condom most times with their current partner, and 75.4% did so with non-steady partner. The current work provides information which shows that 21.4% of young women (ages 15-19 years) were forced into having sexual relations, 7.4% of adolescents (ages 15-19 years) desired to become pregnant, 51.4% had sexual intercourse, some of them were having it twice a month, and on average they were having sex with 6.6 men in 90 days. Although the mean age of the first sexual intercourse was 15.2 years of age, the mean age of the first sexual intercourse for women between the ages of 15-19 years was 7.7 years of age. Furthermore, 35.7% of women between the ages of 15-19 years had been pregnant in the past. The present work showed that 1 in every 2 Jamaican woman between the ages of 15-19 years had sexual intercourse in the last 30 days, and that 8 out of every 10 had been pregnant. Some 21.5% of these women experienced a miscarriage and stillbirth, with only 2.2 % of them having at least one abortion. This low rate of abortion reported is consistent with the relatively high use of contraceptives reported by the women, and the 62% of the respondents who declared that they asked their partner to use a condom. Taking the possibility of underreporting into account, the reported low use of abortion as a method of contraception contradicts recent media reports of widespread abortion in Jamaica, and the fierce activism of the church lobby against legalizing abortion. The assertiveness of the women in asking their men to use a condom is something that should be further encouraged within the national family planning strategy, rather than joining the moral panic against abortion that sometimes infuses national discussions of family planning. The reality which emerged from the current research is that a
469

little over 50% of women are mostly supported financially by their partners, which means that the males are still able to determine, or veto, contraceptive usage. This is not peculiar to Jamaica as it was also found to be the case in Sub-Saharan Africa, and the removal of spousal authorization (male) was associated with increased contraceptive usage.24 Despite the relatively high use of contraceptives among the women, what emerged from the current work gives rise to many public health and other concerns. Women as young as 7 years of age are having sexual intercourse, and 21 out of every 100 adolescents between the ages of 15-19 years are forced into sexual activities. It does not cease there, as 36 out of every 100 young women (ages 15-19 years) have been pregnant, which means that there would be a high fertility rate or prevalence of adoption among these individuals. A number of young adult women in Jamaica were not only having premarital sexual relations, but they were having sexual intercourse with multiple partners. There is currently a public health problem as adolescents (ages 15-19 years) had more multiple sexual partners in the last 3 months than other women, with some young women engaging in promiscuity. This concurs with the literature which shows high promiscuity, premarital sexual activity and high fertility among young adults.25-27 Less than 50% of the women with multiple partners who responded to the question about condom use indicated that they always used a condom. More than half of these participants do not use a condom with multiple partners, which increases their risk of contracting sexually transmitted infections (STIs). However, since only 24% of the women with multiple partners provided information about condom use, further research is required to explore this critically important health issue. Regarding women with one steady partner, only 44.4% stated that they always used a condom, which leaves them vulnerable to STIs if their male partner is unfaithful.
470

Therefore, although the overall use of contraceptives among the women was relatively high (64%), and 62% of the women asserted that their men used a condom, the breakdown of the data between women with steady and unsteady partners indicated that inconsistent condom use is high among Jamaican women. It is this inconsistent contraceptive usage, in particular condom utilization, that explains the HIV/AIDs epidemic in Jamaica and other developing countries.28-33 Only 56.1% of the sample stated their choice of contraceptives, among those who are currently using a method of contraception. Overall, the data suggests that among the women who responded about the method of contraception used, there is a greater concern for preventing an unwanted pregnancy than for contracting STIs. Only 32 % of these women used a condom, with the remaining 68% using, in descending order of importance, the pill, injection, tubal ligation, the withdrawal method, IUD, the rhythm method, emergency contraceptives, implants and other methods. More women ages 15-19 years used a method of contraception to avoid being pregnant (64%) than to prevent STIs (14.6%), indicating that contraception is more about preventing pregnancy then STIs. Embedded in this finding is the disconnect between awareness, knowledge and practice. Failure to consistently use a condom exposes one to HIV and/or STIs. It is clear from the current findings that women are exposing themselves to STIs by premarital sexual relations, promiscuity, and inconsistent condom usage. Young women having sex as early as 7 years, and becoming pregnant between 15-19 years, exposes many of these individuals to an HIV positive partner. HIV serodiscordant couples is a reality, as was noted by USAID and other scholars.34-37 Economic challenges, and mens economic supremacy, are among the reasons why womens reproductive health issues can be vetoed by males. Embedded in these findings is an explanation
471

of why condom usage and other contraceptive methods are inconsistently used by both women and men, as this is often based on the decision of the male partner. Henry-Lee38 opined that 34% of pregnancies in Jamaica are planned, and that poverty retards information on contraceptives and their usage. Another study found that 80% of adolescent pregnancies39 were unplanned, which reemphasizes the heavy involvement of this age cohort in premarital sexual activities and sometimes promiscuity. The lack of material power and economic independence means that some women will find it extremely difficult to dictate, insisting that their male partners have an HIV test, and that a condom is consistently used in sexual intercourse. Warren et al. opined that 40% of all females aged 14-24 years and 61% of those who are sexually experienced have been pregnant.40 Warren and his colleagues work was in 1988 and the current research found that 56% of women of the same age had already been pregnant. Disaggregating the age cohort (15-24 years), this work found that 67.1% of women ages 20-24 years, and 35.7% of those 15-19 years have already been pregnant. Despite the difference in years between 1988 and now, with increased knowledge, wider access to contraception and increased public health education campaigns, the number of young adult-women who are still having unplanned pregnancies is still higher than in 1988. The public health concern is not only with increased pregnancies among young women, but promiscuity, the increased incidence of HIV28, the problems of inconsistent condom usage, and the disparity between the widespread knowledge of HIV and continued inconsistent condom usage.23, 41 Women who are involved in a relationship are more likely to use a method of contraception. It is possible that these women are not ready to get pregnant or they believe their partner may not be ideal for them. Women in urban areas are more likely to use a method of
472

contraception than rural women. It is possible that rural women may have a lower level of education, and there are not enough professional women in the rural areas compared to urban areas, or rural women may be influenced by the culture of having a lot of children to help with agricultural work, and children may be seen as retirement planning, given the higher level of poverty in rural areas.18-20 Women currently having sex and those with multiple partners tend to use a method of contraception. It is possible that these women are not ready to have children. Similarly, there is a positive relationship between crowding and use of contraceptives. However, the younger the age at which the women start to use a method of contraception, the more likely it is that they are using one currently. Clearly, Jamaica has been struggling with premarital sexual relations and adolescent pregnancy for many decades42 and this continues unabated. In contemporary Jamaica, the issues are old, but there are also some new ones. Added to the old issues are increased HIV prevalence among young adults,43 early sexual relations of women ages 15-19 years, promiscuity among women ages 15-19 years, the prevalence of young women who are forced to have sexual relations, frequency of sexual relations in a 30-day period and the percentage of women ages 1539 years who want to become pregnant. Douglas43 opined that the major cause of mortality among women of 15-44 years in the Caribbean is AIDS, and that 1 in every 50 Caribbean nationals was infected with HIV/AIDS. There is no denying that inconsistent condom use accounts for high fertility, pregnancies and HIV/AIDS infections in the Caribbean and sub-Saharan African nations. Therefore, it is useful to encourage the early use of contraceptives, in particular the condom, as a second method among young women. In 2000, the Jamaican Ministry of Health used the Jamaica Reproductive
473

Health Survey for 1997, among other data sources, to write a Strategic Framework for Reproductive Health Programme for 2000-2005.44 New data indicate that the issues are more diverse than in 1997 and therefore a new policy framework is needed for 2005 and beyond. It is an absolute that consistent condom usage can stem the rise in teenage pregnancies, HIV/AIDS45 and abortions, and reduce sexually transmitted infections46 and the risk of transmission among young women and men in developing countries. Wilks et al.s findings revealed that 97% of those in the lower social hierarchy have had sexual intercourse, compared to 96% of the middle class and 95% of the upper social hierarchy, 60% of Jamaican students (ages 15+ years) have had sexual relations, as have 99% of those with primary or lower education level (secondary, 93%; post-secondary, 94%).47 In addition to the aforementioned, the new findings should be used to effectively frame policies to address the new realities. A multifaceted approach must be taken to address the new realities in Jamaica, and this must include (1) an intervention programme to address the information needs of adolescents, and to make reproductive health services more young people friendly (2) post-intervention surveys to assess the effectiveness of implemented measures, (3) a sensitization campaign against male supremacy in vetoing reproductive health choices of females, (4) identifying new areas for contraception inquiry, (5) formulation of an intact condom usage campaign, and (6) designing a programme to financially empower those in the lower class, the disadvantaged, orphans and young people, as well as to provide the same group with educational empowerment. A study conducted in Mexico City on a group of young females who had unplanned pregnancies emphasized the rationale of financial empowerment in contraceptive decision-making. The respondents indicated that they left contraceptive decisions to their partner as he looks after me,48 indicating the males vetoing power in reproductive

474

health decisions, and the justification for their involvement in womens reproductive health matters.

Conclusion
The majority of participants used some method of contraception. The most popular method of contraception among the women was condoms. Despite the relatively high use of contraceptives among the women, a breakdown of the data on women in a steady relationship and women with multiple partners, suggests that 44% of the women used a condom in the former group and 49% used a condom in the latter group. These findings have implications for the spread of STIs. There are several explanatory variables for contraceptive use among the women. These explanatory factors are age, social class, being in a relationship, the rural-urban dichotomy, being currently pregnant, currently having sex, the number of partners, the age at which the women began using a contraceptive, and crowding.

Conflict of interest
The authors have no conflict of interest to report.

Disclaimer
The researchers would like to note that while this study used secondary data from the Reproductive Health Survey, none of the errors in this paper should be ascribed to the National Family Planning Board, but to the researchers.

Acknowledgement
The authors thank the Data Bank in Sir Arthur Lewis Institute of Social and Economic Studies, the University of the West Indies, Mona, Jamaica for making the dataset (Jamaica Survey of Living Conditions, 2002) available for use in this study, and the National Family Planning Board for commissioning the survey.
475

References
[1] Ekstrand M, Tyden T, Darj E, Larsson M. An illusion of power: Qualitative perspectives on abortion decision-making among teenage women in Sweden. Perspectives on Sexual and Reproductive Health 2009; 41(3): 173-180. [2] Needle RH. The relationship between first sexual intercourse and ways of handling contraception among college students. Journal of the American College Health Association 1975; 24(2):106-111. [3] Yeakey MP, Ramachandran DV, Myint Y, Creanga AA, Tsui AO. How contraceptive use affects birth intervals: results of a literature review. Studies in Family Planning 2009; 40(3): 205214. [4] Peyman N, Oakley D. Effective contraceptive use: An exploration of theory-based influences. Health Education Research 2009; 24(4): 575-585. [5] Becker D, KLassen AC, Koenig MA, LaVeist TA, Sonenstein FL, Tsui AO. Womens perspectives on family planning service quality: An exploration of difference by race, ethnicity and language. Perspectives on Sexual and Reproductive Health 2009; 41 93): 158165. [6] Avogno W, Agadjanian V. Mens social networks and contraception in Ghana. Journal of Biosocial Science 2008; 40(3):413-429. [7] Ogunjuyiqbe Po, Akinlo A, Oni GO. Violence against women as a factor in unmet need for contraception in Southwest Nigeria. Journal of Family Violence 2010; 25(2): 123-130. [8] da Silva Magalhaes PV, Kapczinski F, Kauer-SantAnna M. Use of contraceptive methods among women treated for bipolar disorder. Archives of Womens Mental Health 2009; 12(3):183-185. [9] Garbers S, Correa N, Tobier N, Blust S, Chiasson MA. Association between symptoms of depression and contraceptive method choices among low income women at urban reproductive health centers. Maternal and Child Health Journal 2010; 14910:102-109. [10] Alonso A, Clark CJ. Oral contraceptives and the risk of multiple sclerosis: A review of Epidemiologic evidence. Journal of the Nuerological Sciences 2009; 286 (1): 73-75. [11] Grabbe K, Stephenson R, Vwalika B, Ahmed Y, Vwalika C, Chomba E, Karita E, Kayitenkore K, Tichacek A, Allen S. Knowledge, use, and concerns about contraceptive Methods among sero-discordant couples Rwanda and Zambia. Journal of Womens Health, 2009; 18 (9):1449-1456. [12] Agadjanian V, Yabiku ST, Fawcett L. History, community milieu, and Christian-Muslim Differentials in contraceptive use in Sub-Saharan Africa. Journal of the Scientific Study of Religion 2009; 48(3): 462-479. [13] Waller KA. Understanding policymakers perspectives. Dissertation Abstracts International: Section B: The Sciences and Engineering 2009; 69 (12B):7452. [14] National Family Planning Board. Reproductive Health Survey, 2002 (Jamaica). Kingston: NFPB; 2005. [15] Polit DF. Data analysis and statistics for nursing research. Stamford: Appleton & Lange Publisher, 1996. [16] Grossman M. The demand for health - a theoretical and empirical investigation. New York: National Bureau of Economic Research, 1972.
476

[17] Homer D, Lemeshow S. Applied Logistic Regression, 2nd edn. John Wiley & Sons Inc., New York, 2000. [18] Planning Institute of Jamaica, (PIOJ), Statistical Institute of Jamaica, (STATIN). Jamaica Survey of Living Conditions, 1989-2007. Kingston: PIOJ, STATIN; 1989-2008. [19] Statistical Institute of Jamaica (STATIN). Demographic Statistics, 1988-2007. Kingston; STATIN; 1989-2008. [20] Planning Institute of Jamaica (PIOJ). Economic and social survey of Jamaica, 1990-2007. Kingston: PIOJ;1991-2008. [21] National Family Planning Board (NFPB). Reproductive Health Survey 2002. Kingston: NFPB;2005. [22] Norman LR. Predictors of consistent condom use: A hierarchical analysis of adults from Kenya, Tanzania and Trinidad. Int J of STD & AIDS. 2003;14:584-590. [23] Nnedu ON, McCorvey S, Campbell-Forrester S, et al. Factors influencing condom use among sexually transmitted infection clinic patients in Montego Bay, Jamaica. The Open Reproductive Science J 2008;1:45-50. [24] Cook RJ, Maine D. Spousal veto over family planning services. Am J of Public Health 1987;77(3):339-344. [25] Feyisetan B, Pebley AR. Premarital sexuality in urban Nigeria. Studies in Family Planning 1989;20(6):343-354. [26] Hull TH, Hasmi E, Widyantoro N. Peer initiatives for adolescent reproductive health projects in Indonesia. Reproductive Health Matters 2004;12(23):29-39. [27] Sychareun V. Meeting the contraceptive needs of unmarried young people: Attitudes of formal and informal sector providers in Vietiane Municipality, Lao PDR. Reproductive Health Matters 2004;12(23):155-165. [28] Pan American Health Organization (PAHO). Health in the Americas, 2007 volume IICountries. Washington, D.C: PAHO;2007. pp.448-464. [29] Population Action International (2007). A Measure of Survival. Calculating Womens Sexual and Reproductive Risk. Washington DC: Population Action International. [30] World Health Organization (WHO). World health statistics, 2009. Geneva: WHO; 2009. [31] Rawlins, Joan and Crawford, Tazhmoye (2006). Womens Health in the English-Speaking Caribbean: The Case of Trinidad and Tobago. Journal of Social and Economic Studies; 55(4), 2006:1-31. [32] Camara, B., Lee, R., Garweed, J., Wagner, H., Cazal-Gamelisky, R., and Boisson, E. (2003); in Rawlins, Joan and Crawford, Tazhmoye (2006). Womens Health in the EnglishSpeaking Caribbean: The Case of Trinidad and Tobago. Journal of Social and Economic Studies; 55(4), 2006:1-31. [33] Thomas, Tara (2006). Youth Reproductive and Sexual Health in Jamaica. Washington DC., Advocates for Youth. [34] USAID. HIV prevention knowledge base: Emerging areas. HIV prevention for serodiscordant couples. New York: USAID; 2009. www.aidstarone.com/prevention/knowledgebase. [35]. Bunnel R, Opio A, Musinguzi J, et al. HIV transmission risk behavior among HIV-infected adults in Uganda: Results of nationally representative survey. AIDS 2008;22(5:617-24. [36] Bunnel R, Nassozi J, Marum E, et al. Living with discordance: Knowledge, challenges, and prevention strategies of HIV-discordant couples in Uganda. AIDS Care 2005;17(8):999-1002.
477

[37] Dunkle K, Stephenson R, Karita E, et al. New heterosexually transmitted HIV infections in married or cohabiting couples in urban Zambia and Rwanda: An analysis of survey and clinical data. The Lancet 2008;37:2183-91. [38] Henry-Lee A. Womens reasons for discontinuing contraceptive use within 12 months: Jamaica. Reproductive Health Matters 2001;9(17):213-220. [39] Crawford TV, McGrowder DA, Crawford A. Access to contraception by minors in Jamaica: a public health concern. North Am J of Med Scie 2009;1(5):247-255. [40] Warren DP, Morris L, Jackson J, Hamilton P. Fertility and family planning among young adults in Jamaica. Int Family Planning Perspectives 1988;14 (4):137-141. [41] Duncan J, Gebre Y, Grant Y, et al. HIV prevalence and related behaviors among sex workers in Jamaica. Sexually Transmitted Disease 2010. [42]. Drayton VLC. Contraceptive use among Jamaican teenage mothers. Rev Panam Salud Publica 2002;11(3):150-157. [43] Douglas DL. Perspectives on HIV/AIDS in the Caribbean. In: Morgan O (ed). Health issues in the Caribbean. Kingston: Ian Randle; 2005:pp. xv-xxi. [44] Jamaican Ministry of Health (MoH). Strategic framework for reproductive health within the Family Health programme 2000-2005. Kingston: MoH; 2000. [45] Steiner MJ, Cates W. Are condoms the answer to rising rates of non-HIV sexually transmitted infections? Yes. BMJ 2008;336(7637):184. [46] Carey RF, Lytle CD, Cyr WH. Implications of laboratory tests of condom integrity. Sex Transm Dis 1999;26:216-20. [47] Wilks R, Younger N, Tulloch-Reid M, McFarlane S, Francis D. Jamaica health and lifestyle survey 2007-8. Kingston: Tropical Medicine Research Institute, University of the West Indies, Mona; 2008. [48] World Health Organization (WHO). Reproductive health research at WHO: a new beginning. Biennial Report 1998-1999. Geneva: WHO;2000: p.31.

478

Table 17.1. Sociodemographic characteristic of sample, n = 7,168 n % Characteristic Religiosity At least once a week 2707 37.8 At least once a month 1368 19.1 Less than once a month 861 12.0 Only on special occasions (weddings, funerals, 1631 22.7 christening) Does not attend at all 524 7.3 No response 77 1.1 Marital status Legally married 1542 21.5 Common-law 1733 24.2 Visiting 1959 27.3 Not currently in union 1934 27.0 Currently pregnant Yes 288 4.4 No 6219 94.6 Ever been pregnant Yes 5301 84.3 No 985 15.7 Forced to have sex Yes 747 11.6 No 5707 88.4 Currently having sex (in the last 30 days) Yes 4289 59.8 No 2879 40.2 Currently using a method of contraception Yes 4027 63.8 No 2282 36.2 Employment status Unemployed 4143 57.8 Employed 3025 42.2 Are of residence Urban 1144 16.0 Semi-urban 2079 29.0 Rural 3945 55.0 Socioeconomic class Lower 1705 23.8 Middle 3079 43.0 Upper 2384 33.2 No. of pregnancies that resulted in live births median 2.0 (0, 14) (range) Years of schooling mean (SD) 13.0 years (3.0 years) Age mean (SD) 31.3 years (9.3 years) Age of first sexual intercourse median (range) 17.0 (15,49), mean =15.2 yrs (SD =5.8)

479

Table 17.2. Particular demographic characteristics by age group of respondents, n= 7,126 Age group Characteristic 15-19 yrs 20-24 yrs 25-29 yrs 30-34 yrs 35-39 yrs % % % % %
Currently pregnant Yes Ever been pregnant Yes Forced to have sex Yes Currently having sex (in the last 30 days) Yes Currently using a method of contraception Yes Want to be pregnant Yes Ever had sexual intercourse Yes Age at first sexual intercourse mean (SD) in years Frequency of sexual intercourse in last 30 days Mean (SD) No. of men had sexual intercourse with (in last 3 months) Age at first contraceptive use (in years)

P 40-44 yrs % 0.1 96.1 18.8 66.9 61.9 18.2 100.0


17.2 (3.2) 1.9 (8.6) 3.2 (13.7) 21.8 (5.6)

45-49 yrs % 0.1 95.4 18.9 56.2 53.9 11.6 99.7


15.1 (5.8) 1.5 (6.3) 2.1 (9.9) 22.6 (4.3)
2 = 111.5, P < 0.001 2 = 1289.8, P < 0.001 2 = 11.6, P = 0.071

9.3 35.7 21.4 51.8 62.9 7.4 51.4


7.7 (7.7) 2.0 (9.3) 6.6 (21.4) 15.9 (1.4)

5.7 67.1 23.8 66.3 67.4 21.9 94.8


15.3 (4.2) 1.5 (6.2) 2.7 (11.9) 17.8 (2.2)

6.6 83.5 19.0 71.2 64.5 31.0 97.4


16.4 (4.3) 1.8 (8.0) 2.7 (11.9) 19.2 (3.0)

5.4 91.8 21.0 69.1 66.7 35.6 98.8


16.9 (4.2) 1.5 (6.5) 3.2 (13.4) 20.0 (3.9)

3.3 93.6 19.1 64.9 64.6 36.8 99.0


17.0 (3.7) 1.5 (6.4) 3.1 (13.4) 20.7 (4.5)

2 = 92.8, P < 0.001

2 = 37.7, P < 0.001 2 = 75.6, P < 0.001 2 = 2497.1, P < 0.001 F = 11416.7, P <0.0001 F = 0.5, P <0.0001 F = 5.1, P <0.0001 F = 198.6, P <0.0001

480

Table 17.3: Frequency of condom use (with non-steady partner) by age group, n = 1,748 Age group Frequency of condom usage Always Most of the time Seldom Never Never had non-steady partner Total, n 2 = 20.9, P = 0.644 15-19 yrs 20-24 yrs 25-29 yrs 30-34 yrs 35-39 yrs 40-44 yrs 45-49 yrs % 11.8 7.0 0.0 3.9 77.3 229 % 13.6 9.6 0.3 6.3 70.2 332 % 14.9 6.5 0.0 5.9 72.8 323 % 14.2 5.3 0.6 6.5 73.4 338 % 10.5 4.9 0.0 7.9 76.7 266 % 12.0 6.9 0.0 6.3 74.9 175 % 9.4 8.2 0.0 7.1 75.3 85 12.9 6.8 0.2 6.2 73.9 1748 Total

481

Table 17.4: Table 17.3: Frequency of condom use (with steady partner) by age group, n = 1,767 Age group Frequency of condom use Always Most of the time Seldom Never Never no steady partner Total, n 2 = 63.6, P < 0.0001 15-19 yrs 20-24 yrs 25-29 yrs 30-34 yrs 35-39 yrs % 49.4 43.7 2.6 0.0 4.3 231 % 38.9 54.2 5.7 0.3 0.9 332 % 41.5 49.5 8.7 0.0 0.3 323 % 42.0 48.7 8.5 0.3 0.6 343 % 43.2 50.9 5.1 0.4 0.4 273 40-44 yrs % 54.2 38.5 6.7 0.0 0.6 179 45-49 yrs % 55.8 38.4 5.8 0.0 0.0 86 Total % 44.4 48.0 6.4 0.2 1.0 1767

482

Table 17.5. Logistic regression: Explanatory variables on method of contraception usage of women (ages 15 49 years) in Jamaica, n = 6,043

Explanatory variable Age Lower class (reference group) Upper class In union Urban (reference group) Rural Currently pregnant Currently having sex Number of partners Age at which began using contraception Crowding

Odds ratio 0.98*** 1.00 0.83** 3.35*** 1.00 1.16* 0.01*** 2.29*** 1.85*** 0.99** 1.4***

CI (95%) 0.98 - 0.99 0.73 - 0.95 2.80 - 4.02 1.02 - 1.32 0.00 - 0.02 1.95 - 2.70 1.57 - 2.17 0.98 - 1.00 1.21 1.60

R2 0.006 0.002 0.158 0.001 0.114 0.034 0.036 0.002 0.005

2Log likelihood = 5588.0 R2 = 0.358 Model 2 (df = 9) = 1684.75 P-value < 0.0001 Overall correct classification = 78.5% Correct classification of cases that currently use a method of contraception = 91.8% Correct classification of cases that are not currently using a method of contraception = 54.2% *P < 0.05, **P < 0.01, ***P < 0.001

483

18
Females with multiple sexual partners and their reproductive health matters: A comprehensive analysis of women aged 15-49 years in a developing nation
For millennia, females with multiple sexual partners have been called names including whores, prostitutes, sexually promiscuous, and other negative terminologies. However, there is a gap in the literature regarding their comprehensive reproductive health matters. The aims of the current paper are to elucidate (1) the reproductive health matters of females who have multiple sexual partners, (2) the socio-demographic characteristics of the study population, (3) any associations between age of respondent, years of schooling, age of menarche, age of first intercourse, age when the person began using contraception and age of person with whom they first had sexual intercourse, (4) the prevalence of those who were sexually assaulted, (5) the age cohort of females who were raped, and (6) factors which account for current contraceptive usage. On average females first sexual encounter was with males at least 9.5 years older than them. Fifty-seven percent of the study population had more than 11 sexual partners. Currently using a method of contraception can be explained by social class (middle class, Odds ratio (OR) = 0.08, 95% CI = 0.01 0.59); age at first sexual intercourse (OR = 0.90, 95% CI = 0.68 1.21); employment status (employed, OR = 5.07, 95% CI = 1.06 24.36); and marital status (married or common-law, OR = 0.09, 95% CI = 0.02 0.38). The early initiation of adolescent females into commercial sex work cannot be left unaddressed as the prevalence rate is high. Serious efforts are needed to comprehend, alleviate and rectify such a practice by many school aged females.

484

Introduction
For millennia, females with multiple sexual partners have been called names including whores, prostitutes, sexually promiscuous, and other negative terminologies. Irrespective of the social construction that is used to label these individuals; the reality is, they do exist and must be planned for among the female population. It is well established that sexual promiscuity, low condom usage and early sexual initiation account for some of the increase in HIV/AIDS, teenage pregnancies and abortions in the developing world [1-8]. This provides a rationale for the importance of understanding those who have multiple sexual partners (commercial or otherwise). Unlike some disciplines which are more concerned about the social behaviour of this cohort, public health, in seeking to promote a healthy lifestyle, needs a comprehensive knowledge of the reproductive health matters of all individuals. The reality is that people who are promiscuous and infrequently use condoms have a greater probability of contracting sexually transmitted infections, in particular HIV/AIDS and the human papillomavirus (HPV) [9]. Before public health practitioners can commence any elaborate health intervention programmes to address the reproductive health matters of a population or a sub-population, they must first understand the cohort in question. Studies which have examined commercial sex workers [10-12] are not the same as an inquiry into the health status, health care-seeking behaviour and reproductive health matters of those with multiple sex partners. While a commercial sex worker has multiple sexual partners, there are people who do not exchange sexual favours for money or any other transferable commodity. A commercial sex worker is, therefore, a sub-set of those with multiple sexual partners and not the other way around. In 2007/08, according to Wilks et al. [13], using a sample
485

of 2,848 randomly sampled Jamaicans aged 15-74 years; they found that 24.1% indicated having at least 2 sexual partners (females, 8.4%; males, 41.0%). In Kenya in 1999 it was estimated that 6.9% of women nationally were engaged in commercial sex activities (exchanging sex for money, gifts or favours). While there were no statistics on the prevalence of commercial sex workers in Jamaica, not all of the 24.1% of those with multiple sexual partners [13] are commercial sex workers. However, using figures from the Reproductive Health Survey for 2002, 2.2% of Jamaican women aged 15-49 years were involved in the commercial sex trade [14]. Clearly, there are substantially more females who have had multiple sexual partners compared to those who are engaged in the commercial sex trade. Studies which have examined commercial sex workers have researched reproductive health matters, in particular condom usage, STIs, HIV/AIDS, unwanted pregnancies [10,12,15], and violence against these individuals [12,16]. Wilks et al. [13] found that more females of ages 15-24 years had multiple sexual partners (15.2%) compared to females aged 25-34 years (11.1%); 35-44 years (6.6%); 4554 years (2.6%); 55-64 years (1.0%) and 65-74 years (0.2%); and that 41% of young females (ages 15-24 years) were reporting having sex once per week, whereas only 25% indicated that they never had sexual relations. While the aforementioned information provides pertinent material that can be used to understand promiscuity in females at a particular age cohort, reproductive health goes beyond this, suggesting the potency of more information. The literature showed that studies on females with multiple sexual partners have limited their inquiry to age and gender composition, risk factors, ever having had sexual intercourse, factors associated with the odds of having multiple sexual partners, the rationale for multiple sexual partners and the increased risk of contracting human papillomavirus [13, 17-20].

486

The empirical evidence on females with multiple sexual partners has omitted in a single study the (1) reproductive health matters of females who have multiple sexual partners, (2) socio-demographic characteristics of the study population, (3) associations between age of respondent, years of schooling, age of menarche, age of first intercourse, age when the person began using contraception and age of person with whom they first had sexual intercourse, (4) prevalence of those who were sexually assaulted, (5) age cohort of females who were raped, and (6) factors which account for current contraceptive usage. There is, therefore, a gap in the literature, and this research seeks to fill the void. The aims of the current paper are to elucidate (1) reproductive health matters of females who have multiple sexual partners, (2) sociodemographic characteristics of the study population, (3) associations between age of respondent, years of schooling, age of menarche, age of first intercourse, age when the person began using contraception and age of person with whom they first had sexual intercourse, (4) prevalence of those who were sexually assaulted, (5) age cohort of females who were raped, and (6) factors which account for current method of contraception.

Methods
The current paper extracted a sample of 225 respondents who indicated having had multiple sex partners. The only inclusion/exclusion criterion for this study was having two or more sexual partners. We used data from the Reproductive Health Survey, 2002 conducted by the National Family Planning Board. Since 1997, the National Family Planning Board (NFPB) has been collecting information on women (ages 15-49 years) in Jamaica regarding contraception usage and/or reproductive health. In 2002, the Reproductive Health Survey (RHS) collected data on women ages 15-49 years and men 15-24 years. The current paper extracted the sample of only
487

women (ages 15-49 years) given the nature of the research. The sample was 7,168 women, representing a response rate of 91.8%. Stratified random sampling was used to design the sampling frame from which the sample was drawn. Using the 2001 Census sector (or sampling frame), a three-stage sampling design was used. Stage 1 was the use of a selection frame of 659 enumeration areas (or enumeration districts, EDs). This was calculated based on probability proportion to size. Jamaica is classified into four health regions. Region 1 consists of Kingston, St. Andrew, St. Thomas and St. Catherine; Region 2 comprises Portland, St. Mary and St. Ann; Region 3 is made up of Trelawny, St. James, Hanover and Westmoreland, with Region 4 being St. Elizabeth, Manchester and Clarendon. The 2001 Census showed that region 1 comprised 46.5% of Jamaica, compared to Region 2, 14.1%; Region 3, 17.6% and Region 4, 21.8% [14]. Stage 2 saw the clustering of households into primary sampling units (PSUs), with each PSU constituting an ED, which in turn consisted of 80 households. The previous sampling frame was in need of updating, and so this was carried out between January 2002 and May 2002. The new sampling frame formed the basis upon which the sampling size was computed for the interviewers to use. Stage 3 was the final selection of one eligible female, and this was done by the interviewer on visiting the household. The Statistical Institute of Jamaica (STATIN) provided the interviewers and supervisors, who were trained by McFarlane Consultancy to carry out the survey. The interviewers administered a 35-page questionnaire. The data collection began on Saturday, October 26, 2002 and was completed on May 9, 2003. The data was weighted in order to represent the population of women ages 15 to 49 years in the nation [14].
488

Statistical methods We used the Statistical Packages for the Social Sciences (SPSS) for Windows, Version 16.0 (SPSS Inc; Chicago, IL, USA). Frequencies and means were computed on the basis of sociodemographic characteristics, and other variables. We also performed 2 tests to compare associations in non-metric variables and Pearsons Product Moment Correlation for metric variables. Multiple logistic regressions were used to analyze factors that explained current method of contraception. Where collinearity existed (r > 0.7), variables were entered independently into the model to determine those that should be retained during the final model construction [21]. To derive accurate tests of statistical significance, we used SUDDAN statistical software (Research Triangle Institute, Research Triangle Park, NC), and this was adjusted for the surveys complex sampling design. Measures Crowding is the total number of persons in a dwelling (excluding kitchen, bathroom and verandah). Age is the number of years a person is alive up to his/her last birthday (in years). Contraceptive method comes from the question Are you and your partner currently using a method of contraception? , and if the answer is yes Which method of contraception do you use? Age at which the person began using contraception was taken from How old were you when you first used contraception? Area of residence is measured from In which area do you reside? The options were rural, semi-urban and urban. Currently having sex is measured from Have you had sexual intercourse in the last 30 days? Education is measured from the question How many years did you attend school? Marital status is measured from the following question Are you legally married now?, Are you living with a common-law partner now?
489

(that is, are you living as man and wife now with a partner to whom you are not legally married?), Do you have a visiting partner, that is, a more or less steady partner with whom you have sexual relations?, and Are you currently single? Age at first sexual intercourse is measured from At what age did you have your first intercourse? Gynaecological examination is taken from Have you ever had a gynaecological examination? Pregnancy was assessed by Are you pregnant now? Religiosity was evaluated from the question With what frequency do you attend religious services? The options range from at least once per week to only on special occasions (such as weddings, funerals, christenings et cetera). Subjective social class is measured from In which class do you belong? The options are lower, middle or upper social hierarchy. Analytic Model Using logistic regression, this study seeks to examine factors associated with the method of contraception usage among females with multiple sexual partners in Jamaica. Different social factors influence womens choices and their decision to use a method of contraception, and this study used Grossmans model [22] which established the use of econometric analysis to determine the use of health demand. Grossmans model has been modified and used by many scholars to examine health, health outcome and other health-related issues.

Results
A significant statistical association existed between number of sexual partners and commercial sexual encounter (2 = 39.4, P < 0.0001, Figure 18.1). Figure 18.1 shows that 69.2% of females with 2 sexual partners were paid for the sexual encounter compared to 11.5% of those with 3 sexual partners, 3.8% of those with 4 sexual partners, and 11.5% of those with 11+ sexual partners.
490

The socio-demographic characteristics of the study population are presented in Table 18.1. Table 18.1 shows that on average females first sexual encounter was with males at least 9.5 years older than them. Almost 29% of the sample indicated that they had been sexually assaulted during their lifetime, while 51% indicated that they were sexually assaulted on their first sexual encounter. Almost 33% of the study population was 15-24 years of age, compared to 16.4% ages 2529 years, 20.9% ages 30-34 years, 14.2% ages 35-39 years, 10.7% aged 40-44 years and 4.9% aged 45-49 years. Half of the commercial sex workers indicated that they had been sexually assaulted, and 19.2% stated they were sexually assaulted on their first sexual encounter. Of those who had indicated being commercial sex workers, 32% had done this between 2-5 times, 16% between 610 times, and 44% at least 11 times. When the commercial sex workers were asked when they began this activity, 31% indicated during school, 46% reported after leaving school and 23% remarked both. Of the 54% of the respondents who indicated being currently sexually active (in the last 30 days), 8.5% were in urban zones, 25.8% were in semi-urban areas and 65.8% were in rural areas (2 = 13.09, P = 0.001). Furthermore, 48.3% of the currently sexually active females used a condom the last time (2 = 34.76, P < 0.0001). Table 18.2 presents information on the number of sexual partners, condom usage and frequency of condom usage with steady and non-steady partners. Fifty-seven percent of the study population had more than 11 sexual partners, and condom usage was relatively inconsistent. There was no significant statistical association between age cohort of study population and those who reported being forced (or not forced) to have sexual intercourse (2 = 16.3, P <
491

0.177). However, information can be provided on those who indicated being sexually assaulted by age cohort: 15-19 years, 12.1%; 20-24 years, 27.3%; 25-29 years, 12.1%; 30-34 years, 12.1%; 35-39 years, 21.2%; 40-44 years, 12.1%; and 44-49 years, 3.0%. Almost 17% of the sample shared sanitary conveniences, 11.6% reported having a pelvic or urinary tract infection, 23.4% had done a Pap smear, and 11.7% were commercial sex workers. Table 18.3 presents information on age of respondent, years of schooling, age of menarche, age of first intercourse, age when the person began using contraception and age of person with whom they had their first sexual intercourse. Based on Table 18.3, a positive statistical correlation existed between age of respondent and age of person with whom the individual had their first sexual intercourse (r = 0.28, P = 0.15).

Multivariate analyses Table 18.4 presents information on possible factors which account for using a method of contraception. Using logistic regression analyses, four variables emerged as statistically significant factors of method of contraception. The model had statistically significant predictive power (model (Chi-square (17) = 30.79, P < 0.021); Hosmer and Lemeshow goodness of fit test, 2 = 4.01, P = 0.80), and correctly classified 97.3% of the sample. Four factors account for 34.2% of method of contraception among those who indicated having multiple sexual partners. Method of contraception can be accounted for by social class (lower class, Odds ratio (OR) = 0.32, 95% CI = 0.03 0.60); age at first sexual intercourse (OR = 0.90, 95% CI = 0.68 1.21); employment status (employed, OR = 5.07, 95% CI = 1.06 24.36); and marital status (married or common-law, OR = 0.09, 95% CI = 0.02 0.38).
492

Limitations of study
This study examines females with multiple sexual partners and their reproductive health matters, and was extracted from a cross-sectional survey. Using a nationally representative crosssectional data denotes that the work can be used to generalize about the population of females who indicated having multiple sexual partners (2+); however, it cannot be used to make predictions, forecast, establish trends and causality.

Discussion
This study revealed that sexual promiscuity is associated with the risk of sexual violence as 29 out of every 100 females who had multiple sexual partners were sexually assaulted and the figure is as high as 1 in every 2 among those who are commercial sex workers. Almost 12% of sexually promiscuous female Jamaicans were commercial sex workers. There is a high level of inconsistent condom usage among the sample with a non-steady partner compared to their steady sexual partners. Furthermore, female sexual promiscuity is higher in rural areas, among those with post-secondary education, middle class, and those in visiting unions. Three out of every 4 women in the sample were middle-to-upper class respondents; middle class women were 1.7 times more likely than lower class women to have multiple sexual partners; rural women were 1.9 times more likely than semi-urban women to have multiple sexual partners and this was 3.7 times more than urban women compared to rural women. Of the 54% of respondents who indicated being currently sexually active (in last 30 days), 8.5% were in urban zones, 25.8% were in semi-urban areas and 65.8% were in rural areas. Continuing, 48.3% of the currently sexually active females used a condom the last time. Method of contraception can be accounted

493

for by social class; age at first sexual intercourse; employment status; and marital status (married or common-law). The percentage of the current paper population represents 3.2% of the females aged 1549 years in Jamaica [37], suggesting that the percentage of females with multiple sexual partners in this research is less than that observed in female undergraduate students in China (5.31%) [17]. However, female Jamaicans were having first coitus almost 3 years earlier than Chinese undergraduate women [17]. Another difference which emerged in the comparison was inconsistent condom usage. There were clear dissimilarities between the survey of female undergraduate students in China and the current paper. Inconsistent condom usage among female undergraduate students with multiple sexual partners was 38.6% compared to 56% among females with steady partners and 46% among those with non-steady partners. Hence, this justifies the present work which found an inverse statistical association between age of sexual debut and method of contraception. This means that women who become engaged in later sexual activities are more likely to involved less risky sexual practices, and that encouraging later sexual debut will reduce fertility, STIs, and cost of caring for high school dropouts. The risk behaviour among females in this research highlights the fact that policy-makers need to use the empirical evidence provided herein to formulate preventative strategies targeted at this group. The high percentage of inconsistent condom usage among promiscuous females demonstrates that need for urgent public health intervention to address the pending public health problems which may surface from the current realities. A study by Eversley and Newstetter [23] found that females who are exposed to multiple partners do have a significantly higher

494

chance of encountering a male in a high risk category... which speaks to a justification for the high HIV/AIDS prevalence in the developing nations, particularly Jamaica [10]. Clearly embedded in the findings of the current research is the high rate of sexual violence against women in commercial sex work, and to a lesser extent, females with multiple sexual partners. Sexual violence against women commenced with forced sexual encounters during their adolescent years by older men. In this study, males who initiate sexual intercourse with adolescent females were about 10 years older. So they are not only older and sexually more experienced, but they have more material resources compared to the adolescent females, which opens these vulnerable females to various sexually transmitted infections, in particular HIV/AIDS, and the human papillomavirus (HPV) [9]. Within the context of this study, which found an inverse association between age at first sexual encounter and method of contraception use, it follows that many adolescent females would be 10% less likely to use a method of protection against sexually transmitted infection. The socio-economic challenges faced by many Jamaica, in particular those who are poor; include how to avoid the temptations of being lured by sexual predators, who offer material resources to these vulnerable individuals. The costs of sexually transmitted infections are enormous, but not limited to the individual, and this extends to the wider society [20]. It is this reality which public health practitioners must evaluate when deciding not to immediately address future public health problems associated with multiple sexual partners, early sexual initiation, sexual violence, sexually transmitted infections, and sexual promiscuity. Thus, early sexual initiation among Jamaican females accounts for the multiple sexual relationships, and this within the context of the increasing HIV/AIDs means that adolescent promiscuity must be lowered,
495

altered, and discouraged. Beside the evident enjoyment from the sexual coitus, during the encounter for adolescents, they must be made aware of the risk factors associated with multiple sexual partners and promiscuity. One of the findings which emerged from this work is the percentage of females who commence commercial sex work during school. Based on the findings, 3 in every 10 commercial sex workers in Jamaica admitted that this began during school years, which speaks to (1) the social pressure of peer groups in regard to sexual initiation, (2) sex traders recruiting school girls, (3) social decay in the general society, and (4) economic challenges faced by many families. Early sexual initiation, multiple sexual partners and sexual promiscuity are, therefore, byproducts of socio-economic ills of the Jamaican society. With the high percentage of commercial sex workers who commence their craft during school, this speaks to the economic cost of survivability and how it plays a role in influencing adolescents in such activities. With sexual initiation beginning during adolescent years, multiple sexual partners are the outcome in our day. [20]. Whether it is early sexual initiation, illicit drug use or any other factors [20], the reality in Jamaica is that 33 out of every 100 females who indicated having multiple sexual partners were between 15 and 24 years (16 out of every 100 aged 15-19 years), and most of the individuals resided in rural area, with post-secondary education. Previous studies have established a direct association between education and health status, and other factors and health [24-35], and even though education opens possible opportunities for the recipient, the economic hardship of females in Jamaica is eroding this reality. Poverty hampers economic freedom and choice, and so despite ones willingness, many realities are circumvented. The poor are held in a vicious cycle
496

of continuous poverty, and on the onset of health conditions poverty could extend to the family. With the reality that the prevalence rate of poverty, since 1990, is at least twice as high in rural as in urban areas, [36], the economic difficulty is accounting for rural females having many sexual partners, as this is in keeping with the needed assistance. Still, some of these females, although they receive gifts, money, material items and other articles, for being with males in a sexual relationship, it is not construed as commercial sex work. Economic deprivation is, therefore, creating sexual promiscuity as can be demonstrated by the current findings. This research found that employed females with multiple sexual partners were 5.1 times more likely to use a method of contraception. Thus, unemployed females give males vetoing power over reproductive health, because they do not want the males to withdraw the needed material and other forms of support. Statistics showed that unemployment among females is greater than that for males [37], and with the context that poverty is greater for the former group than the latter and among rural residents [36], a part of the survivability strategy of females is to rely on males for financial support. It should not be surprising that poverty, which along with unemployment is higher in rural Jamaica, sees more multiple sex partners in those zones. Furthermore, males who have economic power, are more likely to be employed and receive greater emoluments, are still able to wield this power, even over their spouses. In this work, married women or females in common-law unions were 91% less likely to use a method of contraception as against leaving it up to the male to make the decision. It is the opportunity cost of his economic provision for the household. Previous studies have established that there is a statistical association between poverty and illness [38-43]. Poverty does not only have an impact on illness, it causes pre-mature deaths,
497

lower quality of life, lower life (and unhealthy life) expectancy, low development, high rates of pregnancy and social degradation of the community [44]. The WHO [44] opined that 80% of chronic illnesses were in low and middle income countries, suggesting that illness interfaces with poverty and other socio-economic challenges. The WHO captures this aptly ...People who are already poor are the most likely to suffer financially from chronic diseases, which often deepen poverty and damage long term economic prospects [44]. Another by-product of poverty is multiple partner relationships, as females use these to reduce the cost of unemployment, poverty and material deprivation. While multiple sexual relations increase the risk of sexually transmitted infections, the price of poverty is not the same as contracting some sexually transmitted infections such as gonorrhea, Chlamydia, and syphilis, Although the individual does not compute the cost of promiscuity in this manner, the burden of escaping poverty is such that the consequences of certain actions are sometimes not considered, and this is more the case among adolescents. When poverty is coalesced with unemployment, the consequences can be devastating for the individual as well as the society. According to Bourne [45], a moderate and direct correlation existed between the prevalence of poverty (in %) and unemployment (R2 = 0.48); not seeking medical care (in %) and prevalence of poverty (in %, R2 = 0.58); and the prevalence of poverty and mortality (R2 = 0.51). People have a desire to live, and with the reality that poverty affects mortality, and the direct association between poverty and unemployment, the alleviation from poverty for females, in particular rural dwellers, is having multiple sex partners. While some sexually transmitted infections can be fatal, the use of a condom reduces this probability, which points to a high condom usage among the study population.

498

Outside of the aforementioned issues on women aged 15-49 years who indicated having multiple sexual partners, the current paper highlights some variables which account for their contraceptive use. Again, 18 out of every 25 women aged 15-49 years old who had multiple sexual partners used a method of contraception, and this study examined factors which account for this choice. This research showed that four social factors account for 34.2% of choice of using a method of contraception. Like a study done by Degni, Ojanlatva and Essen [46], the present findings highlight that changes in women conditions to life (employment, marital and subjective social statuses) influence contraceptive use (or non-use). In this study, women in the lower socioeconomic status were 68% less likely to use a method of contraception in reference to those in the wealthy stratum, and that those who were employed were 5.1 times more likely to use a method of contraception, which concurs with previous studies [46-49]. Embedded in those finding is the incapacitating power of economic dependency over women sexual freedom and autonomy, and how money and economic opportunities influence choices over their reproductive health matters. While the current work concurs with Degni, Ojanlatva and Essens research that social factors are related with contraceptive use, mainly from changes to their life status [46], it showed that only employment status positively influences increased contraceptive use. Married women, lower subjective social status and age at first sexual debut negatively associated with reduced method of contraception. And that marital status (ie. married respondents) contributed the most to the reduction in contraceptive use. Thus, a stable union, particularly marriage, changes the dynamics of contraceptive use and opens an avenue of risky sexual practices as some of those who are engaged in sexual promiscuity are married women.

499

Previous studies have established that the education of the wife [47] or the womens education level and their husbands' level of education [50] as having an effect on contraceptive use. While married women who are engaged in multiple sexual partnerships were less likely to use a method of contraception in the current paper, educational level was not statistical associated with contraceptive use which is contrary to previous studies [47, 50]. However, another research using data from the Pakistan Demographic and Health Survey of 1990-91, which examined the effect of selected socio-cultural and supply factors on contraceptive use as reported by married women of reproductive ages found that (1) womans age, (2) number of living children, (3) education, and (4) place of residence positively affect their contraceptive use. None of those social factors were found to be associated with positive contraceptive use among the current sample, and different explanation for increased contraceptive use among women of the reproductive years who are engaged in multiple sexual partnerships. Using a sample of national probability sample of Jamaican women in the reproductive ages, Bourne et al. [49], women with multiple sexual partners were about two times more likely to use a method of contraception (OR = 1.85, 95% CI: 1.572.17), that those in the lower socioeconomic stratum were 17% less likely to do so (OR = 0.83, 95% CI: 0.730.95). While those in the present sample were more likely to use a method of contraception (20 out of every 25) compared with women of the reproductive ages (17 out of every 25), disaggregating the former group provides invaluable information on those in promiscuous relationships. Clearly, poverty is a retarding factor influencing contraceptive use of women aged 15-49 years who indicated having multiple partnerships compared to general population of women aged 15-49 years. According to Bourne et al. [49], those in the poor income class were 17% less likely to use a method of contraception, while this work found that women in the same socioeconomic
500

stratum were 68% less likely to use a method of contraception. It can be extrapolated from the findings that poverty will cripple good sexual practices, and justifying women involvement in multiple sexual partnerships and inconsistent contraceptive uses. A previous study found that inconsistent pill or condom use was associated mainly with partnership [51], and this work provides further clarity to that finding. It was revealed that inconsistent condom usage was lesser among women with their steady partners (11 out of every 25) than with non-steady partners (14 out of every 25). By using condom usage (ie. 17 out of every 25 women of the reproductive ages who indicated their involvement in multiple relationships) for family planning interventions, measures would be overstating needed approaches for among women in steady sexual unions. Likewise the moderate consistency of condom use with non-steady sexual partners denotes that women would be exposed to HIV/AIDS and other sexually transmitted infections because of their sexual expression and involvement with older men. The current work showed that women begin their sexual debut with men that were about 10 year