Journal of Tropical Pediatrics Advance Access published February 18, 2010

JOURNAL OF TROPICAL PEDIATRICS, VOL. 0, NO. 0, 2010

Socio-demographic Factors and Appropriate Health

Care-seeking Behavior for Childhood Illnesses
by Tinuade A. Ogunlesi, and Durotoye M. Olanrewaju
Department of Paediatrics, Olabisi Onabanjo University Teaching Hospital, Sagamu, Nigeria

Correspondence: Dr T. A Ogunlesi, P. O. Box 652, Sagamu-121001NG, Sagamu, Nigeria.

E-mail: <tinuade_ogunlesi@yahoo.co.uk>.

Summary
The objective of the study was to determine the influence of socio-demographic factors on healthcare-
seeking behaviors for childhood illnesses. This cross-sectional survey was conducted among consecu-
tively admitted acutely ill children in the Children Emergency Room of a Nigerian tertiary care hospital.
A total of 168 respondents were surveyed out of which only 12 (7.1%) performed well with regard to all
the four indicators of appropriate healthcare-seeking behaviors. Bivariate analysis showed significant
association between high maternal education and early care-seeking, utilization of orthodox health
facilities, and drug use at home (p < 0.001 in each case). Similarly, age of child <1 year was associated
with early care-seeking, care-seeking outside home, and utilization of orthodox health services
(p < 0.001 in each case). Logistic regression showed that high maternal education and high family
socioeconomic status were strong predictors of early care-seeking and care-seeking outside the home.
It is concluded that maternal age, maternal education, and family socioeconomic status are predictors of
appropriate healthcare-seeking behaviors for childhood illnesses.

Key words: care-seeking, childhood, demographic factors, Nigeria, socioeconomic status.

Introduction
Health-education models have demonstrated various
stages and characteristics of the perception and recip-
rocal actions for human illnesses. The knowledge
of the causation and characteristics of illness is dis-
tinct from the actions taken in respect of that knowl-
edge. The actions taken may reflect in the attitude
and practice of those who are ill or the care-giver.
The attitude and practice are highly dependent
on the degree of motivation to seek appropriate
health care. Furthermore, the healthcare-seeking
model describes factors which may influence the
effect of motivation on appropriate action. Some of
such factors include the ability to recognize symp-
toms and signs, perception of severity, social charac-
teristics, availability of treatment resources, and
competing needs [1].
Acknowledgements
The assistance of Dr (Mrs) Mojisola Ogundeyi, Dr
Alex Oyinlade, and all the resident doctors in the
Children Emergency Room of the Olabisi Onabanjo
University Teaching Hospital, Sagamu, is deeply
appreciated.
The Integrated Management of Childhood
Illnesses (IMCI) is an intervention specifically tar-
geted at improving the survival of children by
improving the efficacy of diagnostic and therapeutic
measures. Specifically, community involvement is a
vital component of IMCI and this is directed at rec-
ognition of danger signs in childhood illnesses and
encouraging early presentation in the hospital for
appropriate care [2]. Studies have highlighted the
problem of non-recognition of danger signs in child-
hood illnesses and its relationship, direct or indirect,
with childhood mortality [3]. Therefore, beyond the
recognition of danger signs, it is also important to
identify factors which may influence healthcare-seek-
ing behavior for childhood illnesses [4]. For instance,
low socioeconomic status has been linked with poor
utilization of prenatal and delivery services [5] and
poor home care for childhood malaria [6]. Indeed,
high cost of available orthodox health services may
inhibit appropriate use of such facilities [7]. Studies
have also related childhood morbidities and mortal-
ities to delay in seeking care [8, 9]. However, appro-
priate and efficient care could be provided in most
health facilities in the developing world if the children
are presented early. Thus, it is important to examine

ß The Author [2010]. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org 1 of 7
doi:10.1093/tropej/fmq009

factors which may be associated with poor
healthcare-seeking behaviors.
The objective of this study was to determine the
influence of socio-demographic factors on health-
care-seeking behaviors for childhood illnesses. Most
of the earlier studies were conducted in the commu-
nity and with regard to recent illnesses. The present
study differs in the fact that it was conducted with a
focus on on-going illnesses and using a Children
Emergency Room (CHER) setting in a tertiary care
hospital.
Method
This cross-sectional survey of mothers who brought
their ill children to the Children Emergency Room of
the Olabisi Onabanjo University Teaching Hospital,
Sagamu, Ogun State, Nigeria, was conducted between
October 2008 and February 2009. The hospital pro-
vides both emergency and specialized pediatric ser-
vices to children from within and outside the State of
Ogun.
Institutional ethical clearance was obtained and
the subjects were mothers presenting consecutively
who gave informed consent to inclusion in the
study. The major inclusion criterion for this study
was the presence of life-threatening events which
required emergency care, such as resuscitation, rehy-
dration, blood transfusion, respiratory supports, etc.
Children not requiring emergency care who were only
admitted through the Emergency Room into the
wards were excluded.
The research tool was a self-designed close-ended
questionnaire and the data obtained from the respon-
dents included age, sex, educational qualifications,
and occupation of the parents and details of care
given prior to presentation in our hospital. The socio-
economic classification of the children into the upper
classes I and II, middle class III, and lower classes IV
and V was done using parental education and occu-
pation as recommended by Ogunlesi [10]. For the
purpose of this study, orthodox health facilities
include private clinics, primary health centers, and
general hospitals. Mothers 30 years were classi-
fied as young while high educational qualification
for who referred to educational qualifications above
the secondary level.
Data were managed with the Statistical Package
for the Social Sciences (SPSS), version 15, soft-
ware using descriptive and inferential statistics.
Proportions were compared using chi-square test,
odds ratio (OR), and adjusted odds ratio (AOR).
The respondents were grouped according to age, edu-
cational qualification, family type, and socioeco-
nomic class. These groups were compared for the
four indicators of appropriate healthcare-seeking
behavior (care-seeking within 24 h of the onset,
care-seeking outside home, utilization of orthodox
health facilities, and drug use at home).
2 of 7
T. A. OGUNLESI AND D. M. OLANREWAJU
Socio-demographic factors which showed significant
relationship with each of the four indicators were
entered into a binary logistic regression model as
independent variables and analyzed against each of
the four indicators as dependent variables. Statistical
significance was established when p-values were
<0.05 or when 95% confidence intervals (CIs) did
not include unity.
Results
General characteristics
Out of 181 CHER admissions, 168 (92.8%) partici-
pated in the study. Seventy-four (44.0%) mothers
were 30 years while 71 (42.3%) children were
infants. The majority of children were males (90;
53.6%). Seventy-four (44.0%) were of the first birth
order and 48 (28.6%) did not have siblings. Sixty
(35.7%) mothers and 66 (39.3%) fathers were
highly educated. Sixty-four (38.1%) families belo-
nged to social classes I–III but only 28 (16.7%) of
the respondents belonged to a polygamous family
setting.
Healthcare-seeking behaviors
Seventy-two (42.9%) respondents sought care out-
side home while 64 (38.1%) sought care within 24 h
of the onset of illness. Similarly, 85 (50.6%) utilized
orthodox health services prior to presentation in our
hospital while 122 (72.6%) administered various
drugs to their children before presentation in our
hospital. Overall, only 12 (7.1%) respondents
met the criteria for appropriate healthcare-seeking
behaviors. There was no statistically significant
relationship between appropriate healthcare-seeking
behavior and young maternal age [3/12 (25.0%) vs.
71/156 (45.5%); 2 ¼ 1.90, p ¼ 0.16], high maternal
education [6/12 (50.0%) vs. 54/156 (34.6%);
2 ¼ 1.14, p ¼ 0.28], high paternal education [7/12
(58.3%) vs. 59/156 (37.8%); 2 ¼ 1.96, p ¼ 0.16] and
high family socioeconomic status [7/12 (58.3%) vs.
57/156 (36.5%); 2 ¼ 2.24, p ¼ 0.13].
Bivariate analysis
Table 1 shows that a significantly higher proportion
of the children for whom care was sought within 24 h
of the onset of illness were infants (p < 0.001), first
child (p < 0.001), and only child (p < 0.001). High
maternal education, high paternal education, and
high family socioeconomic status were also signifi-
cantly associated with care-seeking within 24 h of
the onset of illness (p < 0.001, p ¼ 0.002, and
p ¼ 0.013, respectively).
In Table 2, age of child <1 year (p < 0.001) and
first birth order (p ¼ 0.004) were significantly asso-
ciated with care-seeking outside the home. Young
mothers (p < 0.001) and high family socioeconomic
Journal of Tropical Pediatrics
T. A. OGUNLESI AND D. M. OLANREWAJU

TABLE 1
Socio-demographic factors associated with care-seeking within 24 h of onset of illness

Parameters Within 24 h (n ¼ 64) After 24 h (n ¼ 104) Statistics

N % N %

Age <1 year 43 67.2 28 26.9 AOR ¼ 5.43; p < 0.001

Male sex 33 51.5 63 60.6 AOR ¼ 1.25; p ¼ 0.57
First child 42 65.6 32 30.7 AOR ¼ 4.21; p < 0.001
Only child 32 50.0 16 15.4 AOR ¼ 5.36; p < 0.001
Maternal age <30 years 43 67.2 31 29.8 AOR ¼ 4.72; p < 0.001
Maternal educationa 49 76.5 11 10.6 AOR ¼ 8.19; p < 0.001
Paternal educationa 35 54.7 31 29.8 AOR ¼ 2.71; p ¼ 0.002
High social classb 32 50.0 32 30.7 AOR ¼ 2.23; p ¼ 0.013
Polygamous family 13 20.3 15 14.4 AOR ¼ 0.66; p ¼ 0.32
a
Educational qualification higher than Senior Secondary School Certificate.
b
Socio-economic classes I, II, and III.

TABLE 2
Socio-demographic factors associated with care-seeking outside the home

Parameters Care within home (n ¼ 96) Care outside home (n ¼ 72) Statistics
N % N %

Age <1 year 23 23.9 48 66.7 AOR ¼ 0.16; p < 0.001

Male sex 33 57.3 63 87.5 AOR ¼ 1.41; p ¼ 0.26
First child 33 34.4 41 56.9 AOR ¼ 0.4; p ¼ 0.004
Only child 35 36.4 13 18.1 AOR ¼ 2.54; p ¼ 0.009
Maternal age <30 years 25 26.0 49 68.1 AOR ¼ 0.17; p < 0.001
Maternal educationa 42 43.8 18 25.0 AOR ¼ 2.30; p ¼ 0.01
Paternal educationa 32 33.3 34 47.2 AOR ¼ 0.56; p ¼ 0.06
High social classb 30 31.3 34 47.2 AOR ¼ 0.51; p ¼ 0.035
Polygamous family 14 14.6 14 19.4 AOR ¼ 0.40; p ¼ 0.40
a
Educational qualification higher than Senior Secondary School Certificate.
b
Socioeconomic classes I, II, and III.

status (p ¼ 0.035) are also significantly associated
with care-seeking outside the home.
Table 3 shows that a significantly higher propor-
tion of the children for whom orthodox health ser-
vices were utilized were 1 year of age (p < 0.001).
High maternal education (p < 0.001), high paternal
education (p < 0.001), and high family socioeconomic
status (p < 0.001) were also significantly associated
with the utilization of orthodox health services.
In Table 4, drug use at home was significantly asso-
ciated with child’s age <1 year (p < 0.001), first child
(p ¼ 0.001), and only child (p ¼ 0.006). Similarly, high
maternal education (p < 0.001) and high paternal
education (p ¼ 0.004) were associated with drug use
at home.
Multivariable analysis
The results of binary logistic regression are shown in
Tables 5–8. High maternal education was the only
factor which predicted care seeking within 24 h
while care-seeking outside the home was predicted
by age of child being <1 year, first birth order, lack
of siblings, polygamous family type, young maternal
age, high maternal education, and high family socio-
economic status. In addition, only high maternal
education predicted utilization of orthodox health
services whereas the age of child being <1 year was
the only predictor of drug use at home.
Discussion
The study showed that majority of the mothers
(92.9%) did not seek healthcare for their ill children
appropriately and promptly in agreement with 88.7%
reported in a similar study in Nepal [11]. Although, a
remarkably high proportion of the mothers adminis-
tered drugs at home in accordance with previous
observations in Ilesa, Nigeria [6] and Nepal [11], it
is worrisome that most of them sought care after 24 h

Journal of Tropical Pediatrics 3 of 7

 T. A. OGUNLESI AND D. M. OLANREWAJU

TABLE 3
Socio-demographic factors associated with utilization of orthodox healthcare

Parameters Orthodox care (n ¼ 85) Non-orthodox care (n ¼ 83) Statistics

N % N %

Age <1 year 50 58.8 21 25.3 AOR ¼ 4.14; p < 0.001

Male sex 42 49.4 48 57.8 AOR ¼ 0.72; p ¼ 0.27
First child 42 49.4 32 38.6 AOR ¼ 1.55; p ¼ 0.15
Only child 58 68.2 62 74.7 AOR ¼ 0.73; p ¼ 0.35
Maternal age <30 years 33 38.8 41 49.4 AOR ¼ 0.65; p ¼ 0.16
Maternal educationa 47 55.3 13 15.6 AOR ¼ 6.44; p < 0.001
Paternal educationa 45 52.9 19 22.9 AOR ¼ 3.60; p < 0.001
High social classb 57 67.0 27 32.5 AOR ¼ 4.15; p < 0.001
Polygamous family 11 12.9 17 20.5 AOR ¼ 1.70; p ¼ 0.19
a
Educational qualification higher than Senior Secondary School Certificate.
b
Socio-economic classes I, I,I and III.

TABLE 4
Socio-demographic factors associated with drug use at home

Parameters Drug use (n ¼ 122) No drug use (n ¼ 46) Statistics

N % N %

Age <1 year 64 52.4 7 15.2 AOR ¼ 5.81; p < 0.001

Male sex 63 51.6 27 58.7 AOR ¼ 0.76; p ¼ 0.41
First child 63 51.6 11 23.9 AOR ¼ 3.29; p ¼ 0.001
Only child 42 34.4 6 13.0 AOR ¼ 3.29; p ¼ 0.006
Maternal age <30 years 51 41.8 23 50.0 AOR ¼ 0.72; p ¼ 0.34
Maternal educationa 54 44.3 6 13.0 AOR ¼ 4.96; p < 0.001
Paternal educationa 56 45.9 10 21.7 AOR ¼ 2.95; p ¼ 0.004
High social classb 42 34.4 22 47.8 AOR ¼ 0.57; p ¼ 0.11
Polygamous family 16 13.1 12 26.1 AOR ¼ 2.34; p ¼ 0.044
a
Educational qualification higher than Senior Secondary School Certificate.
b
Socio-economic classes I, II, and III.

TABLE 5
Logistic regression of factors associated with care-seeking within 24 h

Variables  Odds ratio p-values 95% CI

Lower Upper

Age <1 year 0.755 2.12 0.153 0.75 5.99

First child 0.684 1.98 0.224 0.65 5.97
Only child 0.002 1.00 0.998 0.27 3.63
Maternal agea 0.385 1.47 0.488 0.49 4.36
Paternal educationb 0.969 2.63 0.245 0.51 10.48
Maternal educationb 3.637 17.99 0.000 2.20 10.41
High social classc 1.148 0.11 0.032 0.01 0.83

: coefficient of regression; CI: confidence interval.

a
Maternal age 30 years.
b
Education higher than senior secondary school.
c
Social classes I, II, and III.

4 of 7 Journal of Tropical Pediatrics

T. A. OGUNLESI AND D. M. OLANREWAJU

TABLE 6
Logistic regression of factors associated with care-seeking outside the home

Variables  Odds ratio p-values 95% CI

Lower Upper

Age <1 year 2.564 12.98 0.000 4.44 14.56

First child 1.456 4.28 0.009 1.44 4.76
Only child 2.667 1.06 0.000 0.16 1.24
Maternal agea 3.471 11.41 0.000 9.70 16.78
Polygamous family 1.801 6.05 0.016 1.40 9.03
Maternal educationb 1.524 1.21 0.049 0.04 0.99
High social classc 1.804 6.05 0.016 1.40 10.03

: coefficient of regression; CI: confidence interval.

a
Maternal age 30 years.
b
Education higher than senior secondary school.
c
Social classes I, II, and III.

TABLE 7
Logistic regression of factors associated with orthodox healthcare

Variables  Odds ratio p-values 95% CI

Lower Upper

Age <1 year 0.724 0.47 0.062 0.21 1.04

Paternal educationa 0.042 0.95 0.931 0.36 2.50
Maternal educationa 1.458 1.22 0.004 0.08 0.61

: coefficient of regression; CI: confidence interval.

a
Education higher than senior secondary school.

TABLE 8
Logistic regression of factors associated with drug use at home

Variables  Odds ratio p-values 95% CI

Lower Upper

Age <1 year 1.337 3.80 0.006 1.46 9.92

First child 0.873 2.39 0.091 0.86 6.59
Only child 1.042 0.96 0.948 0.27 3.35
Paternal educationa 0.115 0.89 0.825 0.32 2.47
Maternal educationa 1.026 2.78 0.079 0.89 8.74

: coefficient of regression; CI: confidence interval.

a
Education higher than senior secondary school.

of the onset of illness, within their homes, and poorly
utilized the available orthodox health facilities.
Although, the high drug-use rate is commendable, the
overall poor care-seeking behaviors gave the impres-
sion that the respondents in the present study both-
ered less about appropriate professional evaluation
of ill children prior to drug administration. This
practice poses greater risk to child health.
It is also worrisome that the low rate of prompt
and appropriate healthcare-seeking behaviors cuts
across all classes and creeds as it lacked significant
relationship with maternal age, maternal and pater-
nal education, or family socioeconomic status. This is
in contrast to previous reports that highly educated
parents and socially advantaged parents tend to be
better caregivers [12].
Care-seeking behaviors were better for infants than
for older children in the present study. This suggests
that more attention is paid to the health of infants in
this population. This is commendable because the

Journal of Tropical Pediatrics 5 of 7


younger the age of a child, the higher the risk of mor-
tality but it may appear to be unfair to the older chil-
dren. Every child, irrespective of the age, has a right to
good health [13]. Surprisingly, there was no gender
difference in the care-seeking behaviors in the present
study despite the fact that the study was conducted in
a population that has cultural preference for the male
child and such preference usually extends into health-
care for children [14]. We speculate that this may be a
consequence of the efforts of various sociopolitical
groups who openly canvass for gender sensitivity in
the country.
Children who were of the first birth order and those
without siblings also had good care-seeking attention.
This may also be related to the cultural values to the
‘only-child’ and the ‘first-child’ in the studied popula-
tion. However, it is important to note that children of
other birth orders should equally enjoy the same priv-
ileges [13].
It is instructive that there is a general apathy to the
utilization of orthodox health services in the popula-
tion studied. This may suggest that the perceived
needs for such facilities by the populace may be in
question or that the facilities are inaccessible either in
terms of cost or distance.
Young and highly educated mothers also tended to
have good care-seeking behaviors. It is not certain
why young age would confer an advantage in this
regard, but it could be speculated that they are less
likely to be socially independent similar to the older
mothers [4, 15]. Highly educated mothers are known
to be better users of health information and services
[5, 6] and are thus expected to have better care-seek-
ing behaviors for their ill children. Similarly, they
are better equipped for initiating and controlling
decision making with regard to health [11, 16]. This
may be extrapolated to the overall socioeconomic
status of the family as observed in the present study
that families in the high socioeconomic groups had
better care-seeking behaviors than those in the lower
groups. This observation was similar to a previous
report from Peru [17].
Multivariable analysis in the present study showed
that high family socioeconomic status independently
predicted early care-seeking and care-seeking out-
side the homes while high maternal education also
predicted all the indicators of appropriate care-seek-
ing behaviors with the exception of drug use.
Interestingly, child’s age <1 year was the only factor
that independently predicted drug use at home. Thus,
mothers who are >30 years, those with low education,
families with low socioeconomic status, and children
>1 year of age should be regarded as high-risk groups
for inappropriate care-seeking for childhood illnesses.
Conclusion
It can be concluded from the findings in this study
that maternal age, maternal education, as well as
6 of 7
T. A. OGUNLESI AND D. M. OLANREWAJU
family socioeconomic status are independent
predictors of prompt and appropriate care-seeking
behaviors for childhood illnesses. Community-based
health-education programs are essential for improv-
ing the present care-seeking behaviors. This had been
shown to be effective in Lusaka, Zambia [18]. Such
programs may focus primarily at mothers aged 30
years, mothers with low education, and families in
the lower socioeconomic groups. Older mothers are
likely to be over-confident while mothers with low
education and lower socioeconomic status may not
appreciate the need to put the health of the child first.
Therefore, the health-education programs must be
designed to address the various sociocultural barriers
to appropriate healthcare-seeking behaviors, particu-
larly within 24 h of the onset of illness and preferably
outside the home. The health-education program
may need to be designed with the participation of
community, religious, and opinion leaders for better
impact. Overall, this has the potential to improve
child survival especially in under-resourced parts of
the world.
References
1. Downie RS, Fyfe C, Tannhill A. Health Promotion
Models and Values. Oxford: Oxford University Press,
1994.
2. World Health Organization. Counsel the Mother:
Management of Childhood Illnesses. Geneva: WHO,
1997.
3. Awasthi S, Verma T, Agarwal M. Danger signs of neo-
natal illness: perceptions of caregivers and health work-
ers in northern India. Bull World Health Org
2006;84:819–26.
4. Bazzano AN, Kirkwood BR, Tawiah-Agyemang C,
et al. Beyond symptom recognition: care-seeking for
ill newborns in rural Ghana. Trop Med Int Health
2008;13:123–8.
5. Ogunlesi TA. The pattern of utilization of prenatal and
delivery services in Ilesa, Nigeria. Internet J Epidemiol
2005;2. Available at http://www.ispub.com (12 July
2008, date last accessed).
6. Ogunlesi TA, Okeniyi JAO, Oyedeji GA, Oyedeji OA.
The influence of maternal socio-economic status on the
management of malaria in their children: implications
for the ‘Roll Back Malaria’ Initiative. Nig J Paediatr
2005;32:40–6.
7. Thind A, Andersen R. Respiratory illness in the
Dominican Republic: what are the predictors of
health services utilization of young children? Soc Sci
Med 2003;56:173–82.
8. Sutrisna B, Reingold A, Kresno S, et al. Care-seeking
for fatal illnesses in young children in Indramayu, west
Java, Indonesia. Lancet 1993;342:787–9.
9. D’Souza RM. Role of health-seeking behavior in child
mortality in the slums of Karachi, Pakistan. J Biosoc
Sci 2003;35:131–44.
10. Ogunlesi TA, Dedeke IOF, Kuponiyi OT. Socio-eco-
nomic classification of children attending Specialist
Pediatric Centers in Ogun State, Nigeria. Nig Med
Pract 2008;54:21–25.
Journal of Tropical Pediatrics
T. A. OGUNLESI AND D. M. OLANREWAJU
11. Sreeramareddy CT, Shankar RP, Sreekumaran BV,
et al. Care seeking behavior for childhood illness – a
questionnaire survey in western Nepal. BMC Int Health
Human Rights 2006;6:7.
12. Oyedeji GA, Ajibola AJ, Oyedeji OA. Are the mothers
of hospitalized socially-disadvantaged children worse
care givers than those of non-disadvantaged children?
Nig J Paediatr 2002;29:113–18.
13. Convention on the Rights of the Child. United Nations
General Assembly Resolution A/RES/44/25. New
York: United Nations, 1989.
14. Park K. Preventive medicine in obstetrics,
pediatrics and geriatrics. Park’s Textbook
of Preventive and Social Medicine. 17th edition.
India: M/s Banarsides Bhanot Publishers, 2002;
pp. 359–411.
Journal of Tropical Pediatrics
15. Thairu LN, Pelto GH, Rollins NC, et al. Sociocultural
influences on infant feeding decisions among
HIV-infected women in rural Kwazulu, South Africa.
Maternal Child Nutr 2005;1:2–10.
16. Saito K, Korzenik JR, Jekel JF, Bhattacharji S. A case-
control study of maternal knowledge of malnutrition
and health care-seeking attitudes in rural South India.
Yale J Biol Med 1997;70:149–60.
17. Kristiansson C, Reilly M, Gotuzzo E, et al. Antibiotic
use and health-seeking behavior in an underprivileged
area of Peru. Trop Med Int Health 2008;13:434–41.
18. Fujino Y, Sasaki S, Igarashi K, et al. Improvement in
mothers’ immediate care-seeking behaviors for chil-
dren’s danger signs through a community-based inter-
vention in Lusaka, Zambia. Tohoku J Exp Med
2009;21:73–85.
7 of 7