Adolescent Psychopathology: II. Psychosocial Risk Factors for ...

10 downloads 69 Views 2MB Size Report
stressful life events (e.g., Brown & Harris, 1989), and social and coping skills (e.g., Billings & Moos, 1984). Understanding the psychosocial problems and the ...
Journal of Abnormal Psychology 1994, Vol. 103, No. 2, 302-315

Copyright 1994 by the American Psychological Association, Inc. 0021-843X/94/S3.00

Adolescent Psychopathology: II. Psychosocial Risk Factors for Depression

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Peter M. Lewinsohn, Robert E. Roberts, John R. Seeley, Paul Rohde, Ian H. Gotlib, and Hyman Hops In a prospective study of adolescent depression, adolescents (N = 1,508) were assessed at Time 1 and after 1 year (Time 2) on psychosocial variables hypothesized to be associated with depression. Most psychosocial variables were associated with current (n = 45) depression. Formerly depressed adolescents (n = 217) continued to differ from never depressed controls on many of the psychosocial variables. Many of the depression-related measures also acted as risk factors for future depression (« = 112), especially past depression, current other mental disorders, past suicide attempt, internalizing behavior problems, and physical symptoms. Young women were more likely to be, to become, and to have been depressed. Controlling for the psychosocial variables eliminated the gender difference for current and future but not for past depression.

This article is one in a series reporting findings from the Oregon Adolescent Depression Project (OADP). The OADP consists of a large, randomly selected cohort of high school students (aged 14-18 years) who were assessed at two time points over a period of 1 year (Time 1 [Ti] and Time 2 [T2]) using rigorous diagnostic criteria and a wide array of psychosocial measures. Previous researchers have presented data regarding the comorbidity of major depressive disorder and dysthymia (Lewinsohn, Rohde, Seeley, & Hops, 1991), the comorbidity of depression with other mental disorders (Rohde, Lewinsohn, & Seeley, 1991), and basic epidemiological characteristics of the sample, including the point and lifetime prevalence, incidence, and relapse of depression and other mental disorders (Lewinsohn, Hops, Roberts, Seeley, & Andrews, 1993). In this article we present findings of a study of the psychosocial characteristics associated with current, past, and future episodes of depression during adolescence. Clinical and empirical descriptions of adults in a depressive

episode indicate that concurrent depression negatively affects virtually all spheres of psychosocial functioning, including cognitive style (e.g., Beck, 1976), occupational or academic functioning (e.g., Weissman & Paykel, 1974), interpersonal roles and intimate relationships (e.g., Gotlib & Hammen, 1992), levels of stressful life events (e.g., Brown & Harris, 1989), and social and coping skills (e.g., Billings & Moos, 1984). Understanding the psychosocial problems and the psychopathological conditions associated with depression has important theoretical implications; any viable theory of depression needs to be able to account for the multitude of associated deficits. Clinically, depression-related problematic behaviors often become the focus of treatment intervention. Although the psychosocial problems associated with being depressed are relatively well-known for adults (e.g., Barnett & Gotlib, 1988), the extent to which these characteristics are also associated with depression in adolescents has received much less attention. In pilot research for the OADP, Allgood-Merten, Lewinsohn, & Hops (1990) and Hops, Lewinsohn, Andrews, and Roberts (1990) found that many of the psychosocial characteristics associated with adult depression also characterize depressed adolescents, including other psychopathology, stressful life events, depressotypic cognitive style, low self-esteem, increased self-consciousness, reduced social support, and impaired coping skills. Although it is important to elucidate the psychosocial characteristics of currently depressed individuals, cross-sectional studies do not address the extent to which the depression-related psychosocial characteristics antedate the disorder and function as risk factors for the future occurrence of depression. Prospective information concerning the risk factors for future depression is of unquestioned importance (e.g., Barnett & Gotlib, 1988). Information regarding risk factors is essential for the development and evaluation of etiological theories of depression, in addition to providing an opportunity to identify adolescents who are at elevated risk for depression and perhaps influencing the design of more effective preventive intervention programs for the disorder. Knowledge regarding the risk factors for depression in ado-

Peter M. Lewinsohn, John R. Seeley, Paul Rohde, and Hyman Hops, Oregon Research Institute, Eugene, Oregon; Robert E. Roberts, University of Texas Health Science Center; Ian H. Gotlib, Department of Psychology, Northwestern University. Preliminary findings were presented at the 99th Annual Convention of the American Psychological Association held in San Francisco in August 1991. This research was partially supported by National Institute of Mental Health Grant MH40501. We wish to thank Judy Andrews (project coordinator), Beth Blackshaw, Debbie Blanchard, Jackie Bianconi, Barb Eisenhardt, Sheila Foulkes, Tina Hazzard, Jeanette Healey, Trish Hoerth, Wendy Hogan, Candace Holcomb, Sue Jordan, Patrice Kangas, Mara Levin, Renee Marcy, Katie Newman, Laila Pettersen, Julie Redner, Susan Reno, Diane Rosen, Mary Schnebly, Pat Stamps, Keith Smolkowski, Kevin Smolkowski, Nancy Stevens, Heidi Walker, and Jean Wilson for their contribution to this project. Correspondence concerning this article should be addressed to Peter M. Lewinsohn, Oregon Research Institute, 1715 Franklin Boulevard, Eugene, Oregon 97403-1983.

302

303

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

ADOLESCENT DEPRESSION

lescence would seem to be especially important given that early onset depression appears to represent a more serious form of the disorder (Asarnow et al., 1988; Bland, Newman, & Orn, 1986). Specifically, adolescents with mood disorder have been found to be at elevated risk for relapse (Kovacs et al., 1984; Lewinsohn et al., 1993). Adolescent depression has also been shown to predict a variety of negative outcomes, including academic problems, marital difficulties, delinquency, unemployment, drug involvement, medical hospitalization, car accidents, arrest, and criminal conviction (e.g., Carlson & Strober, 1979; Chiles, Miller, & Cox, 1980; Kandel & Davies, 1986). Although previous studies vary greatly in their methodology and have focused on adults more than on adolescents, sufficient convergence has emerged to implicate the following variables as potential risk factors for depression: (a) previous history of depression (e.g., Hammen, Adrian, & Hiroto, 1988; Lewinsohn, Hoberman, & Rosenbaum, 1988); (b) being female (e.g., Block, Gjerde, & Block, 1991; Lewinsohn et al., 1988); (c) living in a dysfunctional family (e.g., Kandel & Davies, 1982; Kashani, 1988); (d) low parental education (e.g., Velez, Johnson, & Cohen, 1989); (e) stressful life events and low social support (e.g., Hammen et al., 1988; Monroe, Bromet, Connell, & Steiner, 1986); (f) subclinical depression level (e.g., Garrison, Jackson, Marsteller, McKeown, & Addy, 1990; Lewinsohn et al., 1988); (g) anxiety (e.g., Reinherz et al., 1989); (h) suicidal behavior (e.g., Johnson, Weissman, & Klerman, 1990); (i) cigarette smoking (Kandel & Davies, 1986); (j) low self-esteem and body image (e.g., Allgood-Merten et al., 1990; Kaslow, Rehm, & Siegal, 1984); (k) high self-consciousness (e.g., Hops et al., 1990); (1) depression-related cognitions (e.g., Hops et al., 1990); (m) school problems and reduced intellectual competence and coping skills (e.g., Block etal., 1991;Velezetal., 1989); (n) physical disability and poor physical health (Reinherz et al., 1989); (o) excessive interpersonal dependence (Hirschfeld et al., 1989); (p) problematic interpersonal behaviors (e.g., Gotlib, 1982; Youngren & Lewinsohn, 1980), including conflict with parents and interpersonal attractiveness; (q) early death of a parent (Reinherz et al., 1989); and (r) early or late pubertal maturation (Petersen & Crockett, 1985). It should be noted that although many risk factors have been proposed and studied, only a relatively small number have been examined together in any given study. Therefore, knowledge of the contribution of the risk factors in interaction with each other is limited. Another shortcoming of previous research has been an almost exclusive reliance on self-report depression measures, such as the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977). The use of self-report measures is problematic for several reasons. First, the degree of concordance between self-report classification and diagnoses by interviewers using operational diagnostic criteria is moderate (e.g., Lewinsohn & Teri, 1982; Roberts & Vernon, 1983), primarily because of a high number of false positives on the selfreport questionnaire. Second, self-report measures of depression have poor discriminant validity, especially with respect to anxiety (Gotlib & Cane, 1989). Third, self-report measures provide no information concerning the temporal course of depression or about the occurrence and course of other comorbid psychiatric disorders. Finally, the results of several studies (e.g.,

Lewinsohn et al., 1988; O'Hara, Schlechte, Lewis, & Varner, 1991) suggest that the risk factors for depression as measured with self-report questionnaires differ somewhat from the risk factors for developing an episode as per diagnosis. Examining the characteristics of adolescents who have recovered from an episode of depression is important because a history of depression represents a strong risk factor for future depression (Lewinsohn et al., 1993). Determining the distinguishing characteristics of formerly depressed adolescents may add to the understanding of the factors contributing to their vulnerability. In a previous study (Rohde, Lewinsohn, & Seeley, 1990), the main distinguishing characteristic of formerly depressed adults was a continuing level of mild but significant depressive symptoms. On the assumption that early onset depression represents a more pernicious form of the disorder, we expected that formerly depressed adolescents, unlike adults, would continue to exhibit depression-related psychosocial characteristics even after they had recovered from the depressive episode. Finally, the examination of gender differences in adolescent depression has received insufficient attention. Female adolescents are more likely to have been, to be, and to become depressed (e.g., Lewinsohn et al., 1993). In view of the preceding discussion, our objectives in this study were to (a) identify the distinguishing psychosocial characteristics of depressed adolescents; (b) ascertain whether and to what extent depression-related psychosocial characteristics persist after remission; (c) assess the extent to which these psychosocial characteristics constitute risk factors for future episodes of depression; and (d) examine gender differences in the variables associated with current, past, and future depression.

Method Subjects and Procedure Participants were randomly selected in three cohorts from nine senior high schools representative of urban and rural districts in western Oregon. Sampling was proportional to size of the school, grade within school, and gender within grade. A total of 1,710 adolescents completed the initial (TO assessments (interview and questionnaires) between 1987 and 1989, with an overall participation rate of 61% (additional details are provided in Lewinsohn et al., 1993). Half of the TI sample (52.9%) were female, with an average age of 16.6 years (SD = 1.2). The representativeness of the TI sample was assessed using several approaches; differences between the sample and the larger population, and between participants and those who declined to participate, were small. No differences were found between the sample and 1980 census data on gender, ethnic status, or parental education, although our sample had a slightly higher proportion of two-parent families. Although decliners had a lower mean socioeconomic status (SES) level than did participants, both represented the middle class. Participants and decliners did not differ on gender of head of household, family size, and number of parents in the household. Adolescents were paid for their participation, and written informed consent was obtained from both the participants and their legal guardians. At the second assessment (T2), 1,508 participants (88.1%) returned for a readmmistration of the interview and questionnaire (mean T!-TZ interval =13.8 months, SD = 2.3). Biases that might have emerged due to attrition in the Ti-T2 panel sample were examined by comparing the adolescents who did not participate at T2 (« = 202) with the 1,508 subjects on demographic characteristics and measures of psychopathology.

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

304

LEWINSOHN ET AL.

Small but statistically significant differences were found. Attrition was associated with lower parental SES, F( 1, 1431) = 11.6, p < .001; smaller household number, F( 1,1683) = 4.0, p < .05; being male (60% of dropouts vs. 46%), x2( 1, N = 1,710) = 13.2, p < .001; and lower educational level of the parents, F(\, 1544) = 14.1, p < .001. Significantly higher attrition rates also were noted for subjects who had a history of disruptive behavior disorders (16.8% vs. 10.8%), *2(1, N= 1,710) = 30.1,p < .01; and for young men with a history of substance use disorders (26.1 % vs. 13.7%), X2( 1, A' = 819) = 7.7, p < .01. However, the nonparticipants did not differ from the Ti-T2 panel sample on other measures of psychopathology such as number of suicide attempts, number of episodes of current and past disorder including depression, the self-report depression measures, or on race or grade level. Approximately half of the T,-T2 panel sample was female (54%), with an average T, age of 16.5 years (SD = 1.2). A total of 8.9% were non-White; 71.3% were living with two parents, and 53% were living with two biological parents. Parental education (maximum value for mother or father) consisted of the following categories: 1.9% did not complete high school, 16.1% completed high school, 35.1% had some college, and 46.9% had an academic or professional degree.

Diagnostic Interview Participants were interviewed with a version of the Schedule for Affective Disorders and Schizophrenia for School-Age Children (KSADS) that combined features of the Epidemiologic version (K-SADSE; Orvaschel, Puig-Antich, Chambers, Tabrizi, & Johnson, 1982) and the Present Episode version (K-SADS-P), and included additional items to derive diagnoses of past and current psychiatric disorders as outlined in the revised third edition of the Diagnostic and Statistical Manual of Menial Disorders (DSM-III-R; American Psychiatric Association, 1987). Interviewers also completed a 14-item version of the Hamilton Rating Scale for Depression (HRSD; Hamilton, 1960) for any current or past depressive episode. At T2, subjects were diagnostically interviewed using the Longitudinal Interval Follow-Up Evaluation (LIFE; Shapiro & Keller, 1979). The LIFE procedure provides detailed information about the course of psychiatric symptoms and disorders since the initial K.-SADS interview, with rigorous criteria for recovery from a disorder (i.e., symptom free for 8 or more weeks). Therefore, diagnostic information was available regarding the occurrence, onset age, and duration of all disorders prior to, and during the course of, the study. At both the T, and T2 interviews, information was gathered regarding current and past suicidal ideation and past suicide attempts, with ratings made regarding the intentionality and medical lethality of any past attempt. At the T2 interview, the interviewers elicited information and rated the presence and severity of depression symptoms and completed the 14-item HRSD for current symptoms and worst past symptoms in the T,-T2 interval. The interviewers also made DSM-III-R Axis V Global Assessment of Functioning (GAF; American Psychiatric Association, 1987) for (a) the current level of functioning (i.e., at the T2 evaluation) and (b) the highest level of functioning during the period between T, and T2. Diagnoses at T, were divided into current and past (i.e., prior to Ti but not present at TO. The following diagnostic categories were created (current and past prevalence rates are provided in parentheses): depression, consisting of major depressive disorder (MDD) and dysthymia (current = 2.9%, past = 17.4%); anxiety, consisting of panic disorder, agoraphobia, social phobia, simple phobia, obsessive-compulsive disorder, separation anxiety, and overanxious disorder (current = 3.2%, past = 5.6%); disruptive behavior, consisting of attention-deficit hyperactivity disorder, conduct disorder, and oppositional disorder (current = 1.8%, past = 5.5%); and substance use, consisting of substance abuse disorders and substance dependence disorders (current = 2.3%, past =

6.0%). A miscellaneous category, "other," consisted of all other assessed disorders (current = 1.2%, past = 6.6%). Diagnostic interviewers were carefully selected, trained, and supervised and all interviews were audio- or videotaped. For reliability purposes, a second interviewer reviewed the recordings of approximately 12% of the interviews and made diagnoses. Interrater reliability was evaluated by the kappa statistic (Cohen, 1960). With four exceptions (kappas for diagnoses for lifetime dysthymia, lifetime eating disorders, and current and lifetime anxiety disorders were .58, .66, .60, and .53, respectively), all TI kappas were equal to or greater than .80. We computed kappas across all disorders at T2 because of the low frequency of many disorders. The kappas for any disorder versus no disorder at T2 and between T, and T2 were .87 and .72, respectively.

Diagnostic Groups For the purposes of this study, four diagnostic groups were formed: (a) depressed at TI (n = 45); (b) future cases (n = 112), adolescents who were not depressed at TI who became depressed between T, and T2 (63 of the future cases had had a depressive episode before TI; it was a first depressive episode for the remaining 49 adolescents); (c) formerly depressed (n = 217), adolescents with a history of depression who were not depressed at any time during the study; and (d) never depressed (« = 1,134), adolescents with no history of depression at the time of the T2 assessment. Inclusion in the depressed at TI group took precedence over the remaining diagnostic groups. The percentage of subjects in each of the diagnostic groups with the lifetime occurrence of a disorder other than depression at T2 were 73%, 46%, 44%, and 24%, respectively.

Assessment Instruments Demographics. Adolescents and their parents reported gender, age, race (six categories), grade in school, number of siblings (natural and step), birth order, composition of the household, as well as parental age, race, marital status, occupational status, and education level. Psychosocial constructs. An extensive battery of psychosocial measures was administered to all participants at TI and T2, with the intent of assessing all psychosocial variables known or hypothesized to be related to depression. On the basis of several extensive pilot studies involving four high schools, many of the measures had been abbreviated (Andrews, Lewinsohn, Hops, & Roberts, 1993). Because a large number of interrelated variables had been administered at TI and T2, measures were reduced to a smaller number of composite variables. Factorial structures reported by the original authors of a measure that could be replicated using confirmatory factor analyses were retained. The remaining variables were rationally categorized into general clusters, which were submitted to principal-components factor analysis with varimax rotation. Measures in each factor with factor loadings greater than .40 were standardized and summed using unit weighting to create a composite score. If composite scores were found to be strongly correlated (i.e., r > .50) and conceptually similar, the factors were combined into a single construct. Using these procedures, the psychosocial measures were categorized into 22 constructs. Brief descriptions of each construct are presented in turn. Internal consistency of the constructs (i.e., a coefficient alpha based on standardized measure scores at T]), the T,-T2 correlation, and the correlation with TI CES-D score are shown in Table 1. All variables were scored such that higher values indicate more problematic functioning. 1. Stress: Daily hassles assessed the frequency of occurrence of unpleasant social and nonsocial events in the past month (20 items; Unpleasant Events Schedule, Lewinsohn, Mermelstein, Alexander, & MacPhillamy, 1985).

ADOLESCENT DEPRESSION

Table 1 Psychometric Properties of the Psychosocial Variables CES-D r

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Variable Stress: Daily hassles Stress: Major life events Current depression Other psychopathology: Internalizing behavior problems Other psychopathology: Externalizing behavior problems Depressotypic cognitions: Pessimism Depressotypic cognitions: Attributions Self-consciousness Self-esteem Self-rated social competence Emotional reliance Future goals: Academic Future goals: Family Future goals: Occupational Coping skills Social support: Family Social support: Friends Interpersonal: Conflict with parents Interpersonal: Attractiveness Physical health and illness Maturational level Female adolescents Male adolescents Death of parent before age 12 Low social desirability Hypomanic personality style Low vocabulary level Dissatisfaction with grades Perceived parental dissatisfaction with grades Days missed of school Late for school Failure to complete homework Repeated a grade Poor self-rated health Quetelet obesity index Infrequent exercise Use of medications Overnight hospital stays Difficulty with feeding or dressing Difficulty climbing stairs Difficulty at work or school Difficulty with activities Number of lifetime physical symptoms Ever smoked cigarettes Current rate of cigarette smoking

.79 .78 .81

.55 .52 .40

.48 .23 .88

.72

.55

.66

.68 .61 .63 .74 .59 .81 .83 .77 .61 .63 .76 .77 .72 .81 .94 .51

.42 .61 .55 .54 .62 .64 .54 .74 .58 .48 .55 .64 .60 .51 .22 .46

.31 .58 .43 .30 .38 .31 .43 .23 ns .15 .51 .37 .31 .28 .13 .21

.64 .74 — .53 .68 .72 — —

.58 .74 N/A .54 .55 .71 .24

ns ns ns .20 .19 .15 .13

— — — — — — — — — — — — —

.30 .34 .39 .46 N/A .49 .75 .47 .29 .05 .24 .27 .23 .21

.12 .17 .13 .17 .10 .22 ns .13 ns ns .10 .11 ns ns

.90 — —

.48 N/A .28

.16 .10 .15

Note. Unless otherwise noted, correlations are significant at the .01 level. T, = Time 1; CES-D = Center for Epidemiologic Studies Depression Scale; N/A = not applicable.

2. Stress: Major life events assessed the occurrence of 14 negative life events to self or significant others (i.e., parent, sibling, other relative, or close friend) during the past year (selected from the Schedule of Recent Experiences, Holmes & Rahe, 1967; Life Events Schedule, Sandier & Block, 1979). Because they represented symptoms of psychopathology, we did not include 3 of the 14 items (i.e., "got in a lot of arguments or fights," "had problems with drugs or alcohol," "tried to commit suicide") as stressful events to self. 3. Current depression assessed depression as per the CES-D (20

305

items; Radloff, 1977), 21 items from the Beck Depression Inventory (BDI, Beck, Ward, Mendelson, Mock, & Erbaugh, 1961), a single item assessing depression level during the past week, and the interviewerrated HRDS. 4. Other psychopathology: Internalizing behavior problems assessed the tendency to worry (5 items; e.g., Maudsley Obsessional Compulsive Inventory, Hodgson & Rachman, 1977), frequently recurrent hypomaniclike behavioral fluctuations (12 items; General Behavior Inventory, Depue & Klein, 1988), state anxiety (10 items; State-Trait Anxiety Inventory, Spielberger, Gorsuch, & Lushene, 1970), quantity and nature of sleep (8 items), and hypochondriasis (8 items; Pilowsky, 1967). 5. Other psychopathology: Externalizing behavior problems assessed externalizing problems as per the number of K-SADS symptoms at TI for conduct disorder, oppositional disorder, and attention-deficit hyperactivity disorder; an unpublished scale assessing conduct problems during the past week (6 items); and a current diagnosis of disruptive behavior disorder, substance use disorder, or eating disorder. 6. Depressotypic cognitions: Pessimism assessed attitudes regarding self-reinforcement (10 items; Frequency of Self-Reinforcement Attitude Questionnaire, Heiby, 1982), likelihood of the occurrence of future positive events (5 items; Subjective Probability Questionnaire, Munoz & Lewinsohn, 1976), endorsement of dysfunctional attitudes (9 items; Dysfunctional Attitude Scale, Weissman & Beck, 1978), and perceived control over one's life (3 items; Pearlin & Schooler, 1978). 7. Depressotypic cognitions: Attributions assessed attributional style along the internal-external, stable-unstable, and global-specific dimensions in which a negative events scale and a positive events scale were derived (48 items; Kastan Attributional Style Questionnaire for Children, Kaslow, Tanenbaum, & Seligman, 1981). 8. Self-consciousness assessed private and public self-consciousness (9 items; Self-Consciousness Scale, Fenigstein, Scheier, & Buss, 1975). 9. Self-esteem assessed satisfaction with specific body parts (3 items; Body Parts Satisfaction Scale, Berscheid, Walster, & Bohrnstedt, 1973), general satisfaction with physical appearance (3 items; Physical Appearance Evaluation subscale, Winstead & Cash, 1984), and self-esteem (3 items; Self-Esteem Scale, Rosenberg, 1965). 10. Self-rated social competence assessed self-perceived social competence (5 items; Social subscale of the Perceived Competence Scale for Children, Harter, 1982; 7 items from Lewinsohn, Mischel, Chaplin, & Barton, 1980). 11. Emotional reliance assessed the extent to which individuals desire more support and approval from others, are anxious about being alone or abandoned, and are interpersonally sensitive (10 items; Emotional Reliance Scale, Hirschfeld, KJerman, ChodofF, Korchin, & Barrett, 1976). 12. Future goals: Academic assessed estimated future education (1 item), self-reported grade average for the last term, perceived adequacy of scholastic performance (1 item), perceived ability to complete college (1 item), and the importance of future academic achievements (5 items; adapted from the Importance Placed on Life Goals Scale, Bachman, Johnston, & O'Malley, 1985). 13. Future goals: Family assessed the importance of future goals related to marriage and family; e.g., "Finding the right person to marry," "Having children" (5 items, adapted from Bachman et al., 1985). 14. Future goals: Occupational assessed the importance of future income level and steady employment; e.g., "Having lots of money," "Being able to find steady work" (3 items; adapted from Bachman et al., 1985). 15. Coping skills assessed the ways in which individuals cope with stressful situations, using 17 items originally selected from the Self-Control Scale (Rosenbaum, 1980), the Antidepressive Activity Questionnaire (Rippere, 1977), modified by Parker and Brown (1979), and the Ways of Coping Questionnaire (Folkman & Lazarus, 1980).

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

306

LEWINSOHN ET AL.

16. Social support: Family assessed enjoyable and aversive interactions with family members on the basis of items from the Appraisal of Parents subscale of the Conflict Behavior Questionnaire (11 items; Prinz, Foster, Kent, & O'Leary, 1979), the Parent Attitude Research Instrument (6 items; Schaefer, 1965), the Cohesion subscale of the Family Environment Scale (3 items; Moos, 1974), the Arizona Social Support Interview Schedule (Barrera, 1986), and the Social Competence Scale of the Youth Self-Report (2 items; Achenbach & Edelbrock, 1987). 17. Social support: Friends assessed the number of friends, frequency of interaction, and relationship quality on the basis of items from the Social Competence Scale (2 items; Harter, 1982), the Social Competence Scales of the Youth Self-Report (3 items; Achenbach & Edelbrock, 1987), the UCLA Loneliness Scale (8 items; Russell, Peplau, & Cutrona, 1980), and the number of friends listed as providing social support (Barrera, 1986). 18. Interpersonal: Conflict with parents assessed the number of parent-child conflictual issues during the past 2 weeks and the average intensity of discussions regarding these issues (45 items; Issues Checklist Scale; Robin & Weiss, 1980). 19. Interpersonal: Attractiveness assessed physical attractiveness and attractiveness as a potential friend and as a co-worker as per interviewers' evaluation (17 items; Interpersonal Attraction Measure, McCroskey & McCain, 1974). 20. Physical health and illness assessed the number of days spent in bed as a result of illness in the past year (1 item), the number of visits to a physician in the past year (1 item), and the occurrence of 88 physical symptoms (e.g., broken bones, ulcers, double vision) during the past 12 months. 21. Maluralional level assessed pubertal development (8 items for female subjects, 11 items for male subjects; adapted from Petersen, Crockett, Richards, & Boxer, 1988). Items assessed the amount of hair on various body parts and changes in body shape (female subjects only) and changes in voice (male subjects only). 22. Miscellaneous measures. Additional variables were assessed but were not included in the constructs because either they did not rationally fit into a general cluster or they did not load significantly on a factor (i.e.,