Research on Social Work Practice http://rsw.sagepub.

com/

The Life Distress Inventory: A Psychometric Evaluation

Marianne R. Yoshioka and Tazuko Shibusawa Research on Social Work Practice 2002 12: 752 DOI: 10.1177/104973102237473 The online version of this article can be found at: http://rsw.sagepub.com/content/12/6/752

Published by:
http://www.sagepublications.com

Additional services and information for Research on Social Work Practice can be found at: Email Alerts: http://rsw.sagepub.com/cgi/alerts Subscriptions: http://rsw.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://rsw.sagepub.com/content/12/6/752.refs.html

>> Version of Record - Nov 1, 2002 What is This?

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

RESEARCH ON SOCIAL WORK PRACTICE 10.1177/104973102237473

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

The Life Distress Inventory: A Psychometric Evaluation

Marianne R. Yoshioka Tazuko Shibusawa
Columbia University

The Life Distress Inventory (LDI) is an 18-item rapid assessment tool measuring self-reported distress across areas of social life and functioning. Thomas, Yoshioka, and Ager (1994) originally developed this measure for clinical use with spouses of problem drinkers. The present report examines its utility with a sample of healthy, nonclinical adults. An assessment of the reliability and construct validity of the LDI was conducted based on a racially diverse sample of 176 men and women drawn from a medical study of body composition. A factor analysis resulted in the development of four sub-scales, each with adequate internal reliability. Subscale scores correlated with convergent factors in hypothesized directions and were unrelated to discriminant factors. Finally, LDI scores for these healthy adults were compared with those from a sample of HIV positive and obese patients also participating in the research. Recommendations for clinical use and future research are suggested.

The Life Distress Inventory (LDI) is an 18-item rapid assessment inventory (RAI) of subjective distress across various areas of social life and functioning. Thomas, Yoshioka, and Ager (1994) originally developed this measure for clinical use with spouses of problem drinkers. This current investigation extends the use of the LDI as a measure of quality of life with a broader population. Much research has broken down the broader construct of quality of life into medical, psychological, emotional, and social components (Haas, 1999; Smith, Avis, & Assmann, 1999). Quality of life is typically defined in terms of (a) physical functioning: mobility, range of movement, ability to manage grooming and tasks of daily living, and limitations in terms of ones ability to
Authors Note: The research reported here was supported by funds provided by an NIH award (Grant 2P01 DK42618-06A). We would like to acknowledge the contributions of Dr. Edwin J. Thomas of the University of Michigan School of Social Work and Dr. Richard Ager of Tulane University School of Social Work. Correspondence may be addressed to Marianne R. Yoshioka, Columbia University School of Social Work, 622 West 113th Street, New York, NY 100254600; e-mail: mry5@columbia.edu.
Research on Social Work Practice, Vol. 12 No. 6, November 2002 752-767 DOI: 10.1177/104973102237473 2002 Sage Publications

752

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

753

fulfill the behaviors related to the roles that we are responsible for in our lives, or (b) psychological distress or well-being: depression, anxiety, stress, hopelessness, hopefulness, satisfaction, and general happiness. Each of these terms is concerned with subjective evaluations of ones life as a whole or with specific domains of life on a global or facet level. The LDI defines quality of life in terms of psychological distress. Stress appraisal is concerned with an individuals psychological assessment of how events or encounters with the environment subjectively affect him or her (Lazarus & Folkman, 1984; Monroe & Kelley, 1995). Sources of stress include major life events and daily hassles (Pearlin, Lieberman, Menaghan, & Mullan, 1981). Events and demands of the environment that are perceived as stressful have been linked with negative mental health outcomes such as depression and anxiety (Lazarus & Folkman, 1984). Work, marital relationships, and conflict between family and work have been identified and examined as chronic stressor effects (Cohen, Evans, Stokols, & Krantz, 1986; Cohen, Kessler, & Gordon, 1995). An individuals response to stress is associated with his or her ability to cope and the resources that are available to him or her (Pearlin & Schooler, 1978). According to Cohen, Kessler, and Gordon (1995), there are three types of stress measurements. The first type focuses on the assessment of environmental events that require individuals to adapt to the environmental context. The second focuses on an individuals assessment of his or her ability to cope with specific events and other environmental demands. The third type focuses on the biological dimension of stress and on the physiological system that is activated by an event or environmental demands. Despite strong evidence that links stress and physical disease, relatively few self-report measures of stress have been developed (Monroe & Kelley, 1995). Quite often, distress is measured by instruments such as the Brief Symptom Inventory (BSI) (Derogatis, 1975), a 53-item measure assessing psychological symptom patterns as opposed to environmental stressors or ones coping abilities (DeGarmo & Kitson, 1996; Murphy, Gupta, Cain, & Johnson, 1999; Thompson, Kaslow, Kingree, & Rashid, 2000). The LDI, in contrast, is a rapid assessment measure of the level of distress within specific environmental contexts. Rapid assessment instruments (RAIs) have been used to assess the intensity, frequency, or duration of a client problem by social workers (Fischer & Corcoran, 1994; Levitt & Reid, 1981). Test scores can be compared with averages from samples of identified populations and can be used for monitoring progress and assessing treatment outcomes (Reid & Smith, 1989). The thrust for accountability in social work practice requires increasing use of

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

754

RESEARCH ON SOCIAL WORK PRACTICE

RAIs and other standardized tools in social work practice (Bloom, Fischer, & Orme, 1999; Blythe, Tripodi, & Briar, 1994; Jordan & Franklin, 1995). One of the weaknesses of RAIs is that there is an inherent limit to the amount and type of information that can be gathered. To make an RAI easy to understand and complete, the questions tend to be superficial (Reid & Smith, 1989). In addition, standardized response formats themselves may be culturally biased. Flaskerud (1988) has observed that degree of response variation that is captured in a Likert-type format often does not translate conceptually for many cultural groups. However, these weaknesses are not necessarily exclusive to RAIs but may be applicable to many quantitative measures. The limited number of self-report measures of distress makes the LDI a potentially useful tool for social workers.

DEVELOPMENT OF THE LDI

The LDI originally was developed as a clinical and research tool for use with the spouses of problem drinkers (Thomas, Yoshioka, & Ager, 1994). Drawing on clinical experience with alcoholic families and pilot testing with a sample of 24 spouses, 18 items that assess subjective distress were identified. Participants are asked to report how much distress they are currently experiencing in each of the 18 item areas. The introductory paragraph to the measure reads,
This scale is intended to estimate your current level of distress with each of the eighteen areas of your life listed below. Please circle one of the numbers (1-7) beside each area. Numbers toward the left ends of the seven-unit scale indicate higher levels of distress, while numbers toward the right end of the scale indicate lower levels of distress. Try to concentrate on how distressed you currently feel about each area. Please circle one number for each item.

Participants then indicate the severity of distress they are experiencing by circling one of the response categories that range from 1 = no distress to 7 = the most distressed. The items are a list of 18 categories including aspects of social life, relationships, and functioning (e.g., marriage, household management, expectations for future). Total scale scores can range from 18 to 126. Higher scores denote greater distress. Thomas et al. (1994) conducted preliminary psychometric analyses of the LDI based on the reports of 77 spouses of alcoholics recruited from the community to receive 6 months of treatment around issues related to their partners drinking. Their sample was made up predominantly of Caucasian women with a mean age of 46 years (SD = 10.8). All participants were married

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

755

at the time of the research with an average length of marriage of 22 years (SD = 11.2). A factor analysis of these data resulted in the development of five subscales (i.e., marriage concerns, career concerns, outside activities, self and family, and general life issues). Cronbach alphas for these subscales ranged from .55 to .84. The researchers reported satisfactory construct validity. LDI scores were correlated with convergent factors of psychological distress (r = .33 to .46) and marital happiness (.27 to .65) at statistically significant levels and were unrelated to discriminant indicators (e.g., demographic variables). By reexamining the performance of the LDI with a racially diverse nonclinical sample of both men and women, this report establishes its utility for a wider social work population.

METHOD Sample

The research subjects were participants in a medical study of body composition (i.e., the measurement of bone composition, water, and mineral content) conducted in a large city in the northeastern United States. A total of 354 participants was recruited to take part in the research. The primary group (N = 176) was made up of normal weight adults who self reported as free from major chronic diseases. Eligible participants were recruited through newspaper advertisements and flyers posted throughout the catchment area. As an incentive for participation, participants received a comprehensive description of their body composition and/or nutritional counseling and supplements. The medical study also included smaller samples of HIV-infected adults and obese adults (N = 81 and 97, respectively). The healthy, normal weight adults served as a comparison group for these two clinical samples in terms of body composition parameters. Both the HIV-infected and obese participants were recruited separately to take part in clinical interventions. The HIVinfected group was involved in clinical trials of medications and/or weight training interventions to address malnutrition. The obese group participated in a weight loss program. In the informed consent procedure, participants were told that they were being asked to provide both body composition and quality of life data. Although participants were informed that they had the right to decline participation in the quality of life assessment and continue their participation in the body composition study, no participants used this option.

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

756

RESEARCH ON SOCIAL WORK PRACTICE

Data Collection

The LDI was self-administered as a part of a measurement package assessing quality of life. At an initial intake, participants completed the LDI along with the depression and anxiety subscales of the BSI (Derogatis, 1975), the Medical Outcomes Study-36 Short Form (SF-36) (Ware & Sherbourne, 1992), and the Satisfaction With Life Scale (SWLS) (Diener, Emmons, Larsen, & Griffin, 1985). The depression and anxiety subscales of the BSI are widely used. Items are scored such that higher scores denote greater intensity of symptoms. Internal consistency estimate for the depression subscale is .85 and for the anxiety subscale is .81. The test-retest correlations are .84 and .79, respectively. Estimates of the construct validity of these subscales also are satisfactory. Correlations between subscale scores and those of the clinical scales of the MMPI and the Wiggins content scales of the MMPI are .46 to .72 for depression and .40 to .48 for anxiety. In an effort to reduce the length of the entire measurement package, only these two subscales of the BSI were administered. The SF-36 is a well-established measure of physical and social functioning with adequate psychometric integrity (McHorney, Ware, & Raczek, 1993; Ware & Sherbourne, 1992). The SF-36 has eight subscales measuring various aspects of physical and mental health. Five of the subscales assess aspects of physical functioning: (a) physical functioning: limitations in physical activity because of health problems, (b) role limitations: difficulty performing social roles due to physical health problems, (c) bodily pain, (d) vitality: energy and fatigue, and (e) general health. The remaining three subscales assess aspects of social and emotional functioning: (a) social functioning: limitations in social activities because of physical or emotional problems, (b) mental health: psychological distress and well-being, and (c) role limitations: difficulty performing social roles due to emotional problems. Subscales are standardized so that scores range from 0 to 100 with higher scores denoting better quality of life. Based on a sample of more than 2,000 participants, McHorney et al. (1993) reported that patients with minor and with serious medical conditions could be distinguished on aggregate based on their SF-36 scores. Patients with serious medical conditions scored significantly lower. The SWLS is a measure of overall happiness. Items are scored such that higher scores denote greater satisfaction. This short scale has been found to have a 2-month test-retest correlation coefficient of .82 and an alpha coefficient of .83 to .87. In addition, the SWLS has been found to be positively associated at statistically significant levels with other measures of subjective

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

757

well-being and negatively associated with measures of psychopathology (Diener et al., 1985).
Data Analysis

All psychometric analyses reported here are based on the responses of the healthy, normal weight adults. By focusing solely on this nonclinical group, the performance of the LDI under normative conditions is investigated. By describing the performance and norms of the LDI with a sample of healthy individuals, its applicability is expanded beyond specific groups defined by alcohol use or medical health status. Unlike the Thomas et al. (1994) study, it was anticipated that a sizable portion of this sample may not be in a relationship or have children. As a consequence, a not applicable response category was added to the instrument. In the psychometric analysis, not applicable responses were treated as missing data. However, to calculate LDI scores for the purposes of describing the sample, not applicable responses were recoded as 0. This is the recommended procedure for clinical use of this measure. By keeping in not applicable items but awarding them no points, both scoring and comparison are made easier. Prior to all data analyses, the scores were recoded to range from 0 to 6 as opposed to 1 to 7. By creating a zero point, the distress scores are easier to interpret. A principal components analysis with varimax rotation and listwise deletion was performed on the responses to identify the presence of underlying factors. Cronbach alpha coefficients were calculated for each of the identified factors. Construct validity was assessed by examining the bivariate correlations between factors and convergent and discriminant variables. Finally, LDI subscale and total scores were contrasted between the healthy, normal weight adults and the HIV and obese groups.

RESULTS Sample Characteristics

In total, there were 176 healthy, normal weight adults including equivalent percentages of men (48.9%) and women (51.1%) (see Table 1). This group was racially diverse. Asian, Black, and Hispanic participants each represented approximately one fifth of the group. White participants made up slightly more than one quarter. The remainder of the group was made up of individuals identifying as Other. The ages of these individuals ranged from

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

758

RESEARCH ON SOCIAL WORK PRACTICE

18 to 81, with a mean value of 39 years and a median value of 36 years. In terms of physical functioning, this group of adults scored slightly higher than the normative scores for the general U.S. population reported by Ware and Sherbourne (1992). Based on the responses of 2,474 American adults, the researchers report a normative score in terms of General Health as 72, Physical Functioning as 84, Bodily Pain as 81, Role Limitations Due to Physical Health as 81, and Vitality as 61. This sample of adults scored 78, 90, 78, 88, and 64, respectively, demonstrating comparable to slightly improved physical functioning.
Factor Analysis

The initial factor analysis yielded a 5-factor solution with eigenvalues equal to 1 or higher. In total, these five factors accounted for 77% of the variance and had an average interfactor correlation of .00. After careful inspection, a decision was made to reassign three items and effectively eliminate Factor 3. These three items were conceptually incongruent with each other. As a result, they were reassigned based both on an examination of their second highest factor loading and what made sense conceptually. Item 8, Education, was reassigned to Factor 2. Item 16, Role of Alcohol in the Home, was reassigned to Factor 1, and Item 6, Household Management, was reassigned to Factor 5. Ultimately, four subscales were created: Social Functioning, Life Satisfaction, Finances and Employment, and Marital Distress. The interscale correlations ranged from .274 to .655. The factor loading of each item within the subscale to which it was assigned can be found in Table 2.
Reliability of the LDI and Subscales

Cronbachs alpha was calculated for each of the subscales and the total inventory based on the responses of the healthy, normal weight adults. The reliability coefficients for the LDI subscales and total inventory were as follows: Social Functioning (8 items) .87, Life Satisfaction (5 items) .82, Finance and Employment (2 items) .77, Marital Distress (3 items) .80, and total LDI (18 items) .89. The mean of the interitem correlations was .47 with a range of .27 to .66.
Construct Validity of the LDI

Construct validity pertains to the way that a measure relates or does not relate to other variables on some theoretical basis (Babbie, 1986). Evidence

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

759

TABLE 1:

Sample Characteristics

Healthy, Normal Weight (n = 176)

% female*** Race*** Asian Black Hispanic White Other Age (mean years)** SD Age (median) SF-36 general health*** SD SF-36 physical functioning*** SD SF-36 bodily pain*** SD SF-36 role limitations due to physical health*** SD SF-36 vitality*** SD 51.1 18.8 22.2 18.8 25.6 14.8 38.72 15.47 36.0 77.94 17.03 89.97 17.76 77.74 16.42 87.78 26.72 64.10 19.42

HIV (n = 80)
55.6 0.0 40.7 13.6 44.4 1.2 41.57 8.80 40.0 48.86 24.00 72.47 24.41 63.85 23.78 56.17 43.58 48.40 20.92

Obese (n = 97)
100.0 0 56.7 0 41.2 2.1 43.78 10.62 42.0 69.64 21.04 75.00 22.86 68.46 23.84 70.88 37.10 49.54 22.71

NOTE: SF-36 = Medical Outcomes Study-36 Short Form. * p < .05. ** p < .01. *** p < .001.

of convergent validity would be found if the LDI subscale and total scores positively correlated with depression and anxiety scores and negatively with SWLS scores and the three emotional and social functioning subscales of the SF-36 (i.e., Mental Health, Social Functioning, and Role Limitations Due to Emotional Problems). More specifically, it was anticipated that individuals reporting higher levels of distress would be more likely to report higher levels of depression and anxiety. Appraisal of life distress has been identified as a factor in the development of depression and anxiety (Bedi, 1999; Ernst & Angst, 1992). Along this same vein, it was expected also that these individuals would report lower levels of life satisfaction. Past research has found a relationship between distress and life satisfaction (Headey, Kelley, & Wearing, 1993; Newcomb, 1986). Finally, we expected that individuals reporting higher levels of distress would report also a lower quality of life as assessed by social functioning. Past research has established a link between these variables when examining adaptation to illness and/or aging (Broers,

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

760

TABLE 2:

Principal Components Factor Analysis of Life Distress Inventory (LDI): Means, Standard Deviations, and Standard Error of Measurement of LDI Subscales and Item Factor Loading Based on Healthy, Normal Weight Adults (n = 147)

Factor Loading 1
Factor 1: Social functioning 12. Religion 5. Relationship to other relatives 2. Sex 11. Social life 4. Relationship to children 10. Recreation/leisure 16. Role of alcohol in the home 15. Personal independence Factor 2: Life satisfaction 17. Satisfaction with life 13. Management of time 18. Expectations for future 14. Physical health 8. Education Factor 3: Finances & employment 7. Financial situation 9. Employment Factor 4: Marital distress 3. Relationship to spouse 1. Marriage 6. Household management

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

M 5.89

SD 5.84

SEM
2.11

.87

.843 .790 .782 .774 .718 .671 .485 .437 .008 .005 .160 .614 .002 .133 .198 .354 .513 .145

.006 .105 .003 .007 .413 .186 .005 .212 .831 .778 .707 .685 .324 .143 .010 .140 .004 .005

.101 .197 .205 .002 .006 .377 .673 .396 .007 .374 .467 .008 .708 .003 .257 .126 .006 .632

.138 .262 .004 .307 .153 .113 .009 .386 .312 .184 .289 .126 .244 .874 .785 .002 .141 .253

.293 .197 .266 .247 .003 .226 .117 .424 7.43 .102 .133 .005 .001 .166 4.17 .147 .227 3.12 .803 .691 .562 3.40 1.52 .80 3.22 1.54 .77 5.79 2.46 .80

NOTE: n = 160-167. Bold type indicates largest factor loading.

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

761

TABLE 3:

Construct Validity of Life Distress Inventory (LDI): Pearson Correlation Coefficients Between LDI Subscales and Convergent and Discriminant Factors

Social Functioning
Convergent factors BSI depression BSI anxiety SWLS SF-36 mental health SF-36 social functioning SF-36 role limitations: emotional Discriminant factors Head circumference Spine length

Life Satisfaction

Finances and Employment

Marital Distress

.54* .49** .43** .49** .54** .39** .04 .19

.62** .57** .60** .61** .54** .36** .08 .08

.40** .30** .54** .38** .32** .23** .14 .15

.31** .35** .28** .34** .40** .26** .08 .03

NOTE: n = 172-176. BSI = Brief Symptom Inventory; SWLS = Satisfaction With Life Scale; SF-36 = Medical Outcomes Study-36 Short Form. * p < .05. ** p < .01. *** p < .001.

Kaptein, Le Cessie, Fibbe, & Hengeveld, 2000; Farber, Schwartz, Schaper, Moonen, & McDaniel, 2000; Lawton, 1999). Given that this was a medical study of body composition, measurements of head circumference and spine length were available and were selected as the discriminant variables. It was hypothesized that LDI scores would be unrelated to these two variables. As Table 3 shows, the LDI subscales and total scores were correlated with the convergent factors in the hypothesized direction at statistically significant levels. The strength of these coefficients ranged from .23 to .61 with a mean of .43. LDI subscale and total scores were unrelated to the discriminant factors as hypothesized.
Gender and Group Differences in LDI Scores

Mean item scores for each of the LDI subscales were compared between healthy, normal weight men and women. The analysis found that men and women scored similarly on all subscales and total score. No statistically significant gender differences were found. Subscale mean item scores also were compared between the healthy, normal weight adults and the two smaller clinical samples participating in the body composition research: the HIV-positive adults and obese adults. The HIV group (n = 81) was made up of slightly more women than men (56.6%

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

762

RESEARCH ON SOCIAL WORK PRACTICE

versus 44.4%) (see Table 1). The men within this group differed from the women only in terms of age (F = 4.143, (1,79), p < .05), with men being older by an average of 4 years. All the participants in this group were experiencing health difficulties associated with HIV infection. Consequently, as a group they scored below normative levels on the SF-36. Based on the responses of the HIV group, Cronbach alpha coefficients were calculated for the total scale and subscales. The coefficients were in the moderate to low range (i.e., Social Functioning = .71, Life Satisfaction = .76, Finance and Employment = .50, Marital Distress = .70, total scale = .87). Conceivably, the small numbers of participants in this group may have attenuated the results. The obese group (n = 97) was made up exclusively of women who met the National Institute of Health criteria for obesity. These women were an average of 5 years older than the members of the healthy, normal weight group (Tukey HSD p < .01, 95% C.I. = 8.9 to 1.2). There was no difference in age between the obese and HIV groups. As a group, these women scored below the U.S. norms for the SF-36, indicating a moderately compromised physical functioning. Cronbach alpha coefficients were calculated for the total scale and subscales based on the responses of the obese group. Similar to the results for the HIV group, the coefficients were moderate to low (i.e., Social Functioning = .71, Life Satisfaction = .80, Finance and Employment = .41, Marital Distress = .39, total scale = .86). Statistically significant differences were found across the groups in terms of each of the five physical functioning subscales of the SF-36. Tukey HSD post-hoc tests showed that the healthy, normal weight adults reported higher quality of life in terms of physical functioning in comparison to both the HIVinfected and obese adults. The healthy participants scored an average of 17.5 points higher than HIV-infected adults (p < .001, 95% C.I. = 10.9 to 24.1) and 15.0 points higher than obese adults (p < .001, 95% C.I. = 8.8 to 21.2). In terms of bodily pain, they scored an average of 13.9 points higher than the HIV group (p < .001, 95% C.I. = 7.4 to 20.3) and 9.3 points higher than the obese group (p < .001, 95% C.I. = 3.2 to 15.3). The healthy group scored an average of 15.7 points higher than the HIV group in terms of vitality (p < .001, 95% C.I. = 9.2 to 22.2) and 14.6 points higher than the obese group (p < .001, 95% C.I. = 8.4 to 20.7). In terms of general health and role limitations due to physical health, each group scored differently from the other. The healthy group scored an average of 21.1 points higher than the HIV group (p < .001, 95% C.I. = 22.8 to 35.3) and an average of 8.3 points higher than the obese group (p < .01, 95% C.I. = 2.4 to 14.2). Finally, in terms of general

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

763

TABLE 4:

Life Distress Inventory (LDI) Subscale Mean Item Scores and Standard Deviations for Healthy, Normal Weight (n = 176), HIV-Infected (n = 81), and Obese Adults (n = 97)

Healthy, Normal Weight

Social functioning*** SD Life satisfaction*** SD Finance and employment SD Marital distress** SD Total LDI*** SD * p < .05. ** p < .01. *** p < .001. 1.56 .82 2.47 1.21 3.05 1.68 1.79 1.31 2.02 .88

HIV-Infected
2.25 .94 3.06 1.24 3.46 1.65 2.28 1.47 2.62 1.00

Obese
1.88 .85 2.88 1.25 3.26 1.51 2.19 1.24 2.37 .91

health, the obese adults scored an average of 20.8 points higher than the HIV group (p < .001, 95% C.I. = 13.7 to 27.8). Based on these physical functioning indicators of quality of life and past research, it was expected that the HIV group would report relatively higher levels of distress. Limitations in social functioning have been found to be associated with compromised health (Leiberich et al., 1997; Vogl et al., 1999). Further, obesity is associated with increased health difficulties and aspects of social functioning (Fontaine, Cheskin, & Barofsky, 1996; Le Pen, Levy, Loos, Banzet, & Basdevant, 1998; Pagan & Davila, 1997). For these reasons, it was anticipated that the healthy, normal weight group would score lower or less distressed relative to the other two groups. These hypotheses were supported in the data (see Table 4). A statistically significant difference in terms of social functioning scores was found between the three groups (F (2,351) = 17.85, p < .001). Post-hoc testing found that the healthy, normal weight participants scored substantially lower than both other groups, reflecting lower levels of distress. In comparison to the HIV-infected participants, they scored an average of .68 points lower estimated by Tukey HSD (p < .001, 95% C.I. = .95 to .41), and in comparison to the obese participants, they scored an average of .30 points lower (p < .05, 95% C.I. = .55 to .04). On this scale, the obese participants scored an average of .38 points lower than the HIV participants (p < .05, 95% C.I. = .67 to .07). On both the Life Satisfaction and Marital Distress subscales, the healthy, normal weight participants scored lower than both other groups, although the scores of the HIV-positive and obese participants did not differ statistically from one another. In terms of the Life Satisfaction subscale, the healthy,

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

764

RESEARCH ON SOCIAL WORK PRACTICE

normal weight participants scored an average of .41 points lower than obese participants (Tukey HSD, p < .05, 95% C.I. = .77 to .04) and .58 points lower than HIV-positive participants (p < .001, 95% C.I. = .97 to .20). On the Marital Distress subscale, they scored an average of .5 points lower than obese participants (Tukey HSD, p < .05, 95% C.I. = .80 to .007) and .4 points lower than HIV-positive participants (p < .05, 95% C.I. = .92 to .07). Betweengroup differences on the Finances and Employment subscale were not statistically significant.

DISCUSSION AND APPLICATIONS TO SOCIAL WORK PRACTICE

This study examined the utility of LDI as a self-report measure of subjective distress. The results of the study suggest that the LDI has adequate psychometric properties including internal consistency and construct, concurrent validity. LDI scores differed in anticipated ways at statistically significant levels between the healthy, normal weight and clinical samples, providing preliminary evidence for known-groups validity. It should be noted that although group differences were statistically significant, the clinical significance of these differences is not known. Most self-report measures of subjective distress enumerate specific stressful events or evaluate affective states associated with being stressed. The LDI offers a unique alternative because it measures areas of stress and associated affective states. Social workers are called on to conduct multidimensional assessments that are often grounded in the daily functioning of the client. The LDI can be used to assess clients both in terms of affective state and areas of stressful social functioning. As with other RAIs, the LDI can improve efficiency by allowing social workers to screen more comprehensively and rapidly. It may facilitate the identification of problem areas that a client does not verbally self-report. These areas of distress that are identified through the LDI may then be explored more fully through interview methods. As stated previously, the LDI was originally developed to measure subjective distress among spouses of heavy drinkers. It is important to note that there were two notable demographic differences between the Thomas sample and the current one. The first is marital status. Approximately 41% of this sample were not married at the time of their participation. In contrast, 100% of Thomass sample were in conjugal relationships. Second, the current sample represents an urban and mobile group of adults with diversity in both race

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

765

and lifestyle. The Thomas sample drew from a more homogeneous, suburban population. These differences likely affected the results of the factor analysis. For example, in the original study, the item distressed about sex was a marital concern. In the current sample, it is a social functioning concern along with items such as social life and recreation. Therefore, to expand the clinical utility of the LDI, the following item wording changes are recommended: (a) The item marriage should be changed to intimate relationships (item 1), (b) the item relationship to spouse should be changed to relationship to partner (item 3), and (c) the item alcohol in the home should be changed to alcohol and drugs in daily life (item 16). There were some limitations of the research that should be noted. First, none of the samples were randomly selected, reducing the accuracy with which they represent the larger populations from which they were drawn. Also, although the healthy, normal weight group excluded those adults with chronic illness, no information was gathered concerning past psychiatric history. However, this group of healthy, normal weight adults scored similarly to the nonclinical norms of the SF-36 reported by Ware and Sherbourne (1992), suggesting that they may approximate characteristics of the U.S. general population. In the current climate of the need for accountability especially under managed care programs (Van Hook, Berkman, & Dunkle, 1996; Winegar, 1992), the LDI is a valid and reliable tool that can be used easily by social work practitioners. Although this investigation expanded the utility of the LDI, further research is required. Future investigations should focus on establishing psychometric parameters with larger clinical samples. In this study, the reliability estimates for the Finance and Employment and Marital Distress subscales were notably low, especially for the obese group. Although this may have been affected by the small sample sizes, it also suggests that the LDI may have a different factor structure for this population. To be useful as a clinical tool, the sensitivity of the LDI to evaluate changes in distress as a result of intervention requires further attention. Finally, the feasibility and utility of the LDI within practice settings should be examined.

REFERENCES
Babbie, E. R. (1986). The practice of social research (4th ed.). Belmont, CA: Wadsworth Publishing. Bedi, R. P. (1999). Depression: An inability to adapt to ones perceived life distress? Journal of Affective Disorders, 54, 225-234. Bloom, M., Fischer, J., & Orme, J. G. (1999). Evaluating practice: Guidelines for the accountable professional (3rd ed.). Needham Heights, MA: Allyn & Bacon.

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

766

RESEARCH ON SOCIAL WORK PRACTICE

Blythe, B., Tripodi, T., & Briar, S. (1994). Direct practice research in human service agencies. New York: Columbia University Press. Broers, S., Kaptein, A. A., Le Cessie, S., Fibbe, W., & Hengeveld, M. W. (2000). Psychological functioning and quality of life following bone marrow transplantation: A 3 year follow-up study. Journal of Psychosomatic Research, 48, 11-21. Cohen, S., Evans, G. W., Stokols, D., & Krantz, D. S. (1986). Behavior, health and environmental stress. New York: Plenum Press. Cohen, S., Kessler, R. C., & Gordon, L. U. (1995). Strategies for measuring stress in studies of psychiatric disorders. In S. Cohen, R. C. Kessler, & L. U. Gordon (Eds.), Measuring stress: A guide for health and social scientists (pp. 3-26). New York: Oxford University Press. DeGarmo, D. S., & Kitson, G. C. (1996). Identity relevance and disruption as predictors of psychological distress for widowed and divorced women. Journal of Marriage and the Family, 58, 983-997. Derogatis, L. R. (1975). Brief symptom inventory. Baltimore: Clinical Psychometric Research. Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The satisfaction with life scale. Journal of Personality Assessment, 49, 71-75. Ernst, C., & Angst, J. (1992). Panic disorder: Old wine in new bottles?: Commentary. Integrative Psychiatry, 8, 157-159. Farber, E. W., Schwartz, J. A. J., Schaper, P. E., Moonen, D. J., & McDaniel, J. S. (2000). Resilience factors associated with adaptation to HIV disease. Psychosomatics, 41, 140-146. Fischer, J., & Corcoran, K. (1994). Measures for clinical practice: A sourcebook (2nd ed.). New York: Free Press. Flaskerud, J. H. (1988). Is the Likert scale format culturally biased? Nursing Research, 37, 185186. Fontaine, K. R., Cheskin, L. J., & Barofsky, I. (1996). Health-related quality of life in obese persons seeking treatment. Journal of Family Practice, 43, 265-270. Haas, B. K. (1999). A multidisciplinary concept analysis of quality of life. Western Journal of Nursing Research, 21, 728-742. Headey, B. W., Kelley, J., & Wearing, A. J. (1993). Dimensions of mental health: Life satisfaction, positive affect, anxiety, and depression. Social Indicators Research, 29, 63-82. Jordan, C., & Franklin, C. (1995). Clinical assessment for social workers. Chicago: Lyceum Books. Lawton, M. P. (1999). Quality of life in chronic illness. Gerontology, 45, 181-183. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer. Le Pen, C., Levy, E., Loos, F., Banzet, M. N., & Basdevant, A. (1998). Specific scale compared with generic scale: A double measurement of the quality of life in a French community sample of obese subjects. Journal of Epidemiology & Community Health, 52, 445-450. Leiberich, P., Engeter, M., Olbrich, E., Rubbert, A., Schumacher, K., Brieger, M., Kalden, J. R., & Joraschky, P. (1997). Longitudinal development of distress, coping, and quality of life in HIV positive patients. Psychotherapy & Psychosomatics, 66, 237-247. Levitt, J. L., & Reid, W. J. (1981). Rapid assessment instruments for practice. Social Work Research and Abstracts, 17, 13-19. McHorney, C. A., Ware, J. E., & Raczek, A. E. (1993). The MOS 36-item short-form health survey (SF-36). Psychometric and clinical tests of validity in measuring physical and mental health constructs. Medical Care, 31, 247-263. Monroe, S. M., & Kelley, J. M. (1995). Measurement of stress appraisal. In S. Cohen, R. C. Kessler, & L. U. Gordon (Eds.), Measuring stress: A guide for health and social scientists (pp. 122-147). New York: Oxford University Press.

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014

Yoshioka, Shibusawa / THE LIFE DISTRESS INVENTORY

767

Murphy, S. A., Gupta, A. D., Cain, K. C., & Johnson, L. C. (1999). Changes in parents mental distress after the violent death of an adolescent or young adult child: A longitudinal prospective analysis, Death Studies, 23(2), 129-159. Newcomb, M. D. (1986). Nuclear attitudes and reactions: Associations with depression, drug use, and quality of life. Journal of Personality & Social Psychology, 50, 906-920. Pagan, J. A., & Davila, A. (1997). Obesity, occupational attainment, and earnings. Social Science Quarterly, 78, 756-770. Pearlin, L. I., & Schooler, C. (1978). The structure of coping. Journal of Health and Social Behavior, 19, 2-21. Pearlin, L. I., Lieberman, M. A., Menaghan, E. G., & Mullan, J. T. (1981). The stress process. Journal of Health and Social Behavior, 22, 337-356. Reid, W. J., & Smith, A. D. (1989). Research in social work. New York: Columbia. Smith, K. W., Avis, N. E., & Assmann, S. F. (1999). Distinguishing between quality of life and health status in quality of life research: A meta-analysis. Quality of Life Research, 8(5), 447459. Thomas, E. J., Yoshioka, M. R., & Ager, R. D. (1994). Life distress inventory. In J. Fischer & K. Corcoran (Eds.), Measures of clinical practice: A sourcebook (2nd ed., p. 267). New York: Free Press. Thompson, M. P., Kaslow, J. N., Kingree, J. B., & Rashid, A. (2000). Partner violence, social support, and distress among inner-city African American women. American Journal of Community Psychology, 28, 127-143. Van Hook, M. P., Berkman, B., & Dunkle, R. (1996). Assessment tools for general health care settings: PRIME-MD, OARS, and SF-36. Health and Social Work, 21, 230-234. Vogl, D., Rosenfeld, B., Breitbart, W., Thaler, H., Passik, S., McDonald, M., & Portenoy, R. K. (1999). Symptom prevalence, characteristics, and distress in AIDS outpatients. Journal of Pain & Symptom Management, 18, 253-262. Ware, J. E., & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey (SF-36). Medical Care, 30, 473-483. Winegar, N. (1992). The clinicians guide to managed mental health care. New York: Haworth Press.

Downloaded from rsw.sagepub.com at Alexandru Ioan Cuza on March 9, 2014