Journal of Pediatric Psychology, Vol. 27, No. 6, 2002, pp. 509-517
© 2002 Society of Pediatric Psychology
The Role of Self-Assessed Health in the Relationship Between Gender and Depressive Symptoms Among Adolescents
1 Washington State University, 2 State University of New York at Buffalo, 3 Loyola University Chicago
All correspondence should be sent to Paula G. Williams, Department of Psychology, P.O. Box 644820, Washington State University, Pullman, Washington 99164-4820. E-mail: pwillms{at}wsu.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Objective: To examine the concurrent and longitudinal relations between gender, self-assessed health (SAH), and depressive symptoms among adolescents.
Method: Two measures of SAH (physical symptom reports and global health ratings) and a measure of depression were completed on two occasions over two years by 232 adolescent boys and girls.
Results: Physical symptom reports were related to depressive symptoms both concurrently and longitudinally. Longitudinal path analysis revealed a significant path from gender to physical symptom reports (Wave 1) to depressive symptoms (Wave 2). Although global health ratings were related to depressive symptoms concurrently, the prospective relation was not significant in the cross-lagged path model.
Conclusions: These findings suggest that the development of poorer SAH, particularly the perception of physical symptoms, may place adolescent girls at risk for subsequent depressive symptoms. Potential mechanisms for the SAH-depression relationship are discussed.
Key words: adolescence; gender; physical symptoms; self-assessed health; depression.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Adolescence is typically considered a period of development characterized by excellent health. Hence, we often assume that physical illness and its sequelae do not pose significant problems for adolescents. Perhaps because of this assumption, there has been a relative lack of research on the effects of physical illness in this age group (Williams, Holmbeck, & Greenley, in press
) or on the mechanisms by which individual differences (e.g.,
gender; Sweeting, 1995)
influence self-assessed health (SAH) among adolescents. Recently, however,
Lewinsohn and colleagues (Lewinsohn et
al., 1994; Lewinsohn, Seeley,
Hibbard, Rohde, & Sack, 1996) identified self-reported
physical illness and functional disability due to illness as risk factors for
depression in adolescents. Because gender differences in both depression and
SAH emerge in adolescence, an examination of the interrelations among these
variables seems warranted. This study examined the relations among gender,
depressive symptoms, and SAH in the context of a longitudinal study of normal
adolescent development.
Depression and Health
Depression has been related to higher reports of both chronic and acute
illness (e.g., Katon & Sullivan,
1990; Levenson, Hamer,
Silverman, & Rossiter, 1987). Moreover, self-reported physical
illness is a factor that distinguishes individuals with and without major
depressive disorder in both adults (Rohde,
Lewinsohn, & Seeley, 1990) and adolescents
(Lewinsohn, Gotlib, & Seeley,
1995; Rohde, Lewinsohn, &
Seeley, 1994). Among adults, depression and perceived illness
appear to be reciprocally related: Self-reported physical morbidity is a
strong predictor of later depression; depression in turn has a weaker, but
statistically significant lagged effect on subsequent physical illness reports
(Aneshensel, Frerichs, & Huba,
1984). Among adolescents, both reports of physical disease and
functional impairment due to disease are risk factors for future major
depressive disorder, and major depressive disorder is a risk factor for future
disease and impairment due to disease
(Lewinsohn et al., 1996). The
mechanism for the interrelationship between depression and illness is not
fully understood. Current evidence suggests that biological (e.g., immune
system disruption), behavioral (e.g., health practices), cognitive (e.g.,
perception and interpretation of symptoms), and social (e.g., alteration of
social roles) factors may play a role
(Cohen & Rodriguez,
1995).
Gender Differences in Depression and Self-Assessed Health
The well-documented gender difference in depression among adults is thought
to first emerge in adolescence
(Nolen-Hoeksema, 1990). By the
age of 15, girls and women are about twice as likely to be depressed as boys
and men (Nolen-Hoeksema & Girgus,
1994). A recent 10-year study of depression in adolescents
reported that gender differences first emerged between the ages of 13 and 15
and peaked between the ages of 15 and 18
(Hankin et al., 1998). Similar
to depression, there are significant gender differences in SAH and related
self-regulatory behavior among adults. Women rate their health more poorly,
report more physical symptoms, take more sick days from work, make more
physician visits, and take higher levels of prescription and nonprescription
medication compared to men (National
Center for Health Statistics [NCHS], 1994;
Verbrugge, 1985). It appears
that gender differences in SAH are first evident in early adolescence
(Lewinsohn, Gotlib, & Seeley,
1997; Sweeting,
1995).
A variety of hypotheses have attempted to explain gender differences in
physical morbidity. However, there has been a relative dearth of empirical
research investigating potential mechanisms and no current hypothesis appears
to be definitive (Gijsbers van Wijk &
Kolk, 1997). For example, when biological differences (e.g.,
reproduction-related health problems and physician visits) are controlled,
women still exhibit more morbidity
(Verbrugge, 1985), indicating
that psychosocial variables likely play a role. Unfortunately, research
examining the emergence of gender differences in SAH in adolescence has been
scant, and the role of psychological factors (e.g., mood, self-concept,
willingness to report illness/seek help) has been virtually ignored
(Sweeting, 1995).
Prospective Relations Among Gender, Health, and Depression
Although the emergence of gender differences in both SAH and depressive
symptoms around the time of puberty has been demonstrated, little is known
about the direction of causal relationships among these variables. For
example, it may be that girls report more depression because of their poorer
perceived health, or it may be that their higher rates of depression lead to
greater reports of physical symptoms, or both. Reinherz et al.
(1993) reported that the onset
of health problems between ages 10 and 15 was a significant antecedent risk
for later depression among adolescent girls; however, mediation was not
formally tested. Nevertheless, the pattern of findings suggests that gender
differences in perceived physical health may precede the gender shift in
depression. Thus, one pathway to depression for women and girls may be via
poorer SAH. To test this hypothesis, this study examined the reciprocal
relations between depressive symptoms and SAH over two time points in a
longitudinal sample of adolescents (see
Figure 1).
|
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Sample and Procedures
The sample consisted of 103 boys and 129 girls who participated in a longitudinal study of the experiences of adolescents (Larson & Richards, 1989
).
Participants were randomly selected from four elementary and junior high
schools; two of the schools were located in a working class, blue-collar
suburb of Chicago, whereas the other two schools were located in a
middle-class community. Stratified random sample selection was utilized to
ensure equivalent representation of gender, grade, community, and season of
data collection in the original sample. All participants were Caucasian,
reflecting the ethnic composition of the communities from which they were
sampled. For this research, the variables of interest were assessed at the
second and third assessments of the larger longitudinal study (henceforth
referred to as Waves 1 and 2). Participants were in grades 7 through 10 for
Wave 1 of this report (mean age in years = 13.7, SD = 1.2, range =
11-16) and in grades 9 through 12 at Wave 2 (mean age in years = 15.6,
SD = 1.3, range = 13-18). At each wave, participants completed
questionnaires in a group format in schools, with examiners present to clarify
questions if necessary. Informed consent was obtained from both the
adolescents and their parents.
There were missing data because of absenteeism when surveys were
administered or attrition. Reasons for attrition included refusal or failure
to obtain parental permission to participate, family moves, and two cases of
death of the child (Larsen, Richards, Monetta, Holmbeck, & Duckett, 1996).
Of the 232 participants included in the study, 82 (35.3%) were not surveyed at
Wave 2. The effect of missing data on study variables and demographics was
examined by comparing participants with complete data with those with missing
data. Compared to participants with complete data, those with missing data
rated their global health more poorly at Wave 1 (t[230] = 3.102,
p < .01). Sex of the adolescent and depression were marginally
related to missingness (
2[1, N=232] = 2.85,
p = .09, t[230] = 1.75, p = .08, respectively).
There were no differences between the groups on mean level of physical
symptoms at Wave 1 (t[230] = 1.19,p=.24) or on school grade
(t[230]=-.81, p=.42). The data analytic strategy used in the
study permitted the inclusion of all possible cases (N=232) (vs. a
list-wise deletion procedure). This helped minimize the potential for biased
results that can occur in the presence of missing data (see Results section
for a more detailed discussion of the missing data strategy).
Measures
Depressive Symptoms. Depression is typically viewed in terms of
three constructs: depressive mood, depressive syndromes, and depressive
disorders (Compas, Ey, & Grant,
1993). Because this was a sample of normal adolescents, this study
assessed the continuum of depressive symptoms and not clinical depression.
Participants completed a 14-item version of the Children's Depression
Inventory (CDI; Kovacs, 1985)
modeled after the short form of the Beck Depression Inventory (BDI-SF;
Beck & Beck, 1972) at Waves
1 and 2. The CDI is a self-report scale that quantifies a variety of
depressive symptoms (e.g., perceived mood, vegetative functions,
self-evaluation). It has demonstrated a strong relationship to clinical
interview measures of children's depression
(Garber, 1984). Coefficient
alphas were .81 and .71 for Waves 1 and 2, respectively. The correlation
between the 14 items of the short form and the full 27-item version was .93 in
Wave 1 of this data set (only the 14-item version was given at Wave 2). Items
were averaged to form scale scores. Items are presented as a choice of three
sentences that receive scores of 0, 1, or 2, with higher scores indicating
more severe depressive symptoms (e.g., 0 pts: "I have fun in many
things"; 1 pt.: "I have fun in some things"; 2 pts.:
"Nothing is fun at all").
Global Health Ratings. Global health ratings were assessed at each
time point using the standard single-item format: "How would you rate
your physical health" with a 5-point response scale from
"excellent" to "poor." Lorig et al.
(1996) report the test-retest
reliability on this item over a 10-day interval to be .92 in a sample of
adults. This single item has been shown to be a strong predictor of physical
morbidity and mortality (Idler &
Benyamini, 1997). This item exhibited a modest, significant
correlation with physical symptom reports at both waves of this study (see
Table I), indicating that these
are related, though independent measures of SAH.
|
Physical Symptom Reports. At each time point, participants rated
the extent to which they had experienced 15 commonly occurring physical
symptoms (headache, coughs, sore throat, muscular aches and pains, stomach
ache, earache, congested or stuffy nose, constipation, vomiting, bad acne,
skin rashes, diarrhea, weight change, allergies, and breathing trouble) over
the previous 3 months. Each symptom was rated on a four-point scale: never,
once or twice, occasionally, once a week or more. These symptom items are
consistent with those used in previous studies of adolescent health
(Mechanic & Hansell, 1987;
Wickrama, Lorenz, & Conger,
1997) and did not overlap with the CDI items. Items were averaged
to form a scale score such that higher scores indicate more frequent
experience of physical symptoms.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Descriptive Analyses
Item means (average rating per item), standard deviations, and correlations among study measures appear in Table I. On average, participants endorsed relatively low levels of depressive symptoms (scale means = 4.6 and 5.5 for each wave, possible range = 0 to 28) and physical symptoms (scale means = 12.3 and 12.6, possible range = 0 to 45), consistent with the nonclinical nature of the sample. On average, participants rated their overall health between "good" and "very good." Cross-sectional correlations suggested that high levels of depressive symptoms were associated with more frequent physical symptoms and with poorer ratings of global health. These relationships were also evident from the prospective correlations. Finally, depressive symptoms, frequency of physical symptoms, and global health ratings were all moderately stable.
Path Analyses
Cross-lagged path models were used to evaluate the prospective
relationships between the health variables and depressive symptoms and our
mediational hypotheses. Separate path models were run for frequency of
physical symptoms and global health ratings because of the significant
correlation between these two variables (r = .46 at Wave 1 and
r = .28 at Wave 2, both ps < .01). The hypothesized model
is presented in Figure 1. When
the pattern of findings was suggestive of one of our hypothesized mediational
pathways, we tested whether the path was significantly different from zero
using the delta method (Sobel,
1988).
As with most prospective studies, data for some participants were not
available at both waves of assessment. Analyzing only cases with complete data
has the potential to produce biased results
(Muthén, Kaplan, & Hollis,
1987). Therefore, we wanted to reduce the potential bias in our
findings by analyzing the total sample (N = 232). To do this, we
estimated our path model using full-information maximum likelihood estimation.
This estimation procedure does not impute data; rather, it breaks the
likelihood function down into components based on patterns of missing data,
which allows estimation to proceed using all available data (for a more in
depth discussion of this approach to missing data, see
Arbuckle,
1996).1
The path models were estimated in Mplus 1.4
(Muthén & Muthén,
1998). To implement full-information maximum likelihood
estimation, intercepts (for endogenous variables) and means (for the exogenous
gender variable) were estimated. Because of the relatively low levels of
depressive symptoms and physical symptoms in this nonclinical sample, square
root transformation (Neter, Wasserman,
& Kutner, 1990) was used to reduce skew in the study
variables.
There was significant age variability in our sample. Therefore, preliminary analyses tested whether age moderated relationships between the health variables and depressive symptoms and the effects of gender on depressive symptoms and the health variables. To do this, two age cohorts (7th/8th graders and 9th/10th graders at Wave 1) were created and the path models were compared across these two groups. Nested model comparisons suggested that only 1 of 16 path coefficients differed across the age cohorts. Although significant in both cohorts, the stability of the frequency of health symptoms was stronger for the younger than older cohort (path coefficients=.68 and .49, respectively, ps < .01). All other paths and covariances were equivalent in both the frequency of health symptoms and the health ratings models. In subsequent analyses, the younger and older cohorts were combined.
In addition, we tested whether our path models were different for boys and
girls. To do this, nested model comparisons were used to compare our
crosslagged model across gender. These nested comparisons suggested that the
path models were equivalent for boys and girls, which is consistent with other
findings (Anhesensel et al., 1984;
Lewinsohn et al., 1994).
Physical Symptoms. First, we tested the prospective relationships between frequency of physical symptoms and depressive symptoms and the effect of gender on these variables. The path coefficients for significant paths in this model are presented in Figure 2. Both depressive and physical symptoms were stable over time. Gender predicted physical symptoms at both time points and marginally predicted depressive symptoms at Wave 1. Girls reported a higher frequency of physical symptoms and marginally higher levels of depressive symptoms. Higher reported frequency of physical symptoms predicted higher levels of depressive symptoms 2 years later. However, depressive symptoms did not prospectively predict physical symptoms. These findings suggest that physical symptoms at Wave 1 might mediate the relationship between gender and depressive symptoms at Wave 2. This mediational path was significant (Z= 2.13, p < .05). Thus, girls experienced higher levels of physical symptoms, which in turn predicted higher levels of depressive symptoms 2 years later.2
|
Global Health Ratings. Second, we tested the prospective relationships between global health ratings and depressive symptoms and the effect of gender on these variables. Gender predicted global health ratings (path coefficient = .17, p < .001) and, to a lesser extent, depressive symptoms (path coefficient = .11, p < .10) at Wave 1. Girls reported somewhat higher levels of depressive symptoms and rated themselves in poorer health than boys. Neither cross-lagged path was significant (path coefficients = .05 and .06, ps > .10). Thus, our mediational hypotheses were not supported for global health ratings and depressive symptoms. Although the cross-sectional correlations (Table I) indicate a strong concurrent relation between depressive symptoms and global health ratings, the direction of causality cannot be determined with these data.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
This study is the first to directly examine reciprocal relations between gender, SAH, and depressive symptoms in a sample of adolescents. The 2-year lag in assessments of these variables and controlling for previous levels of depressive symptoms made this an extremely conservative test of causal relationships. Results suggest that adolescent girls' perception of greater physical symptoms places them at risk for increases in subsequent depression. Consistent with previous research (Lewinsohn et al., 1996
),
reports of physical symptoms were related to depressive symptoms both
concurrently and longitudinally. Moreover, the path from gender to physical
symptom reports (Wave 1) to depressive symptoms (Wave 2) was significant.
Thus, greater symptom reports among adolescent girls partially accounted for
subsequent increases in depressive symptoms. The direction of these
relationships is consistent with previous findings in adults over a shorter
time frame (Aneshensel et al.,
1984). Global health ratings, on the other hand, did not
prospectively predict depressive symptoms but did bear strong concurrent
relations to depression. It may be that global health perception is part of
the generally negative self-perceptions that often accompany depressed
mood.
Most previous accounts of the emergence of gender differences in depression
in adolescence have not considered SAH among key factors, so it is unclear
where health perceptions may fit into larger theoretical models of gender
differences in adolescent depression. Lewinsohn and colleagues (Lewinsohn et
al., 1994,
1996,
1997) suggest that a key
mechanism for the relation between SAH and depression may be functional
impairment. Thus, it may be that adolescent girls' higher rates of perceived
physical symptoms lead to restriction of important social or academic
activities, which in turn leads to increases in depressed mood. Consistent
with this notion, adult women exhibit greater levels of disability due to
illness (e.g., sick days from work/school)
(Verbrugge, 1985). Of
relevance is research suggesting that, among adolescents, perception of
competence across a variety of domains (e.g., social, scholastic, athletic) is
related to depression (Harter &
Whitesell, 1996). It is possible that poor SAH influences
perceived competency in these important domains. One point of uncertainty,
however, is whether women's greater restriction of activities is related to
greater actual illness compared to men or to gender differences in the
propensity to assume the sick role. Empirical examination of gender
differences in illness socialization in children has been scant; however, it
has been suggested that girls may be socialized to more readily adopt sick
role behavior than boys (Walker &
Zeman, 1992).
Another possible mechanism for the physical symptom—depression
relationship is that increases or fluctuations in physical symptoms among
adolescent girls may lead to greater self-focused attention, which may make
them vulnerable to depressed mood. Consistent with this notion, previous
research has found a significant relationship between symptom reports and
introspectiveness among adolescents
(Hansell & Mechanic,
1985). Moreover, self-consciousness is a psychosocial variable
linked to depression in adolescents (Chen,
Mechanic, & Hansell, 1998;
Lewinsohn et al., 1997) and
adults (Ingram, 1990).
Additionally, coping strategies characterized by focusing inward on negative
mood states (i.e., rumination), as opposed to engaging in distraction
strategies, are thought to enhance and perpetuate depressed mood states
(Nolen-Hoeksema, Morrow, &
Fredrickson, 1993). Although gender differences in trait measures
of self-focused attention are not consistently found, the relation between
self-focused attention and symptom reports appears to be stronger for girls
and women (Williams & Wiebe,
2000). Future research should attempt to clarify the role of
self-focused attention in the relationship between gender, SAH, and depressed
mood.
The potential relations between SAH and other hypothesized risk factors for
depressed mood deserve mention. For example, early puberty relative to peers
has been suggested as a risk factor for depressed mood among adolescent girls
(e.g., Petersen, Sarigiani, & Kennedy,
1991). Recently, however, Wichstrom
(1999) reported data
suggesting that pubertal timing in and of itself does not appear to mediate
the gender-depression relationship. Rather, dissatisfaction with appearance
(which may co-occur with early pubertal timing) may be an important mechanism
for the development of gender differences in depression. Given that both
dissatisfaction with appearance and poorer SAH have now been linked to
depressed mood in adolescent girls, the manner in which these two risk factors
are related warrants empirical investigation.
In a review of the literature on the emergence of gender differences in
depression during adolescence, Nolen-Hoeksema and Girgus
(1994) conclude that an
interactive model is best supported by the available data. Such an interactive
model posits that girls possess characteristics or risk factors for depression
prior to adolescence, but it is only in the face of the unique challenges of
early adolescence that their higher rates of depression emerge. It is not
clear how SAH fits into such a model. Current data do not indicate that girls
have poorer SAH prior to adolescence
(Sweeting, 1995). However,
similar to the case of depression, girls may be predisposed to focus on
symptoms or assume the sick role (i.e., are socialized differently than boys
with respect to illness behavior) and it is only in conjunction with shifts in
their physical symptoms during puberty that they begin to assess their health
more poorly. That is, bodily changes (e.g., menarche, increases in body fat)
that occur in early adolescence may interact with preexisting tendencies to
produce a shift in SAH. These predisposing tendencies may be similar to those
thought to influence the development of depression (e.g., ruminative coping,
low sense of mastery; Nolen-Hoeksema,
Larson, & Grayson, 1999), or it may be that gender differences
in SAH emerge independently, but then serve to make girls vulnerable to
concurrent and subsequent depression.
There are several important limitations to this study. Whereas depressive
symptoms are a risk factor for the development of depressive disorders and
have been linked to many of the same psychosocial problems (e.g., decreased
quality of life) as clinical depression
(Gotlib, Lewinsohn, & Seeley,
1995), generalization of these findings to clinical depression
disorders should be made cautiously. Future research should replicate and
extend these findings by including more comprehensive and objective measures
of both health and psychopathology.
Conclusions based upon these findings are also limited by the relatively long time lag between study variables and by participant attrition. Although this provided a very conservative test of causal relations and statistical techniques were used to compensate for attrition, it is possible that stronger relationships between SAH and depressed mood would be found using a shorter time frame and participants with greater variability in the constructs of interest. Furthermore, our data suggest that gender differences in both SAH and depressed mood are already evident by the seventh grade. Future research attempting to capture the shift in both of these variables should target children before age 13. Finally, the sample was composed of Caucasian, middle-class adolescents. Study findings should be replicated with adolescents of varying ethnicity and socioeconomic status.
Despite these limitations, this study represents the first test of reciprocal relations between gender, SAH, and depressive symptoms. Findings of this study suggest that the emergence of gender differences in perceived physical symptoms in adolescence may have important ramifications for the development of higher levels of depressed mood among girls. Empirical examination of the consequences of poor SAH among adolescents has been relatively neglected. This study suggests that adolescence is a critical time point for studying the development of gender differences in both health perceptions and depression.
|
Acknowledgments |
|---|
This research was partially supported by National Institute of Mental Health Grant MH38324. We thank Jennifer Dugdale for assistance with data management and preliminary analyses and Grayson Holmbeck for his comments on an earlier draft of this article.
|
Notes |
|---|
1 In maximum likelihood estimation, the logarithm of a likelihood function is taken, and it is differentiated with respect to unknown parameters (e.g., path coefficients) and set equal to zero. In the complete data case for a sample i = 1,..., n observations, the likelihood function would be
,
where xi is an observed data vector for case
i, µ is a mean vector, and 
is a covariance matrix.
Structure is imposed on µ and by letting them be functions of a
smaller set of parameters, such as structural paths in a structural equation
model or factor loadings in a factor analysis. If some data were missing, then
the likelihood function would be
,
where µi is the subvector of µ that deletes the corresponding
elements that are missing from xi, and
i is a submatrix of formed by deleting the rows and
columns corresponding to missing values of x. In this way, all possible
data contribute to the likelihood function, and thus to the estimation of
unknown parameters. 
2 The path model was also run using only participants who had complete data
at both time points (N = 148). Results were comparable: sex was
related to symptoms at Wave 1, path coefficient = .23, p < .05,
which in turn were related to depressive symptoms at Wave 2, path coefficient
= .20, p < .05. This indirect path was significantly different
from zero, Z = 1.963, p < .05. Depressive symptoms were
not prospectively related to physical symptoms, p > .10.
Received September 1, 2000; revision received February 6, 2001; accepted August 1, 2001
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Aneshensel, C. S., Frerichs, R. R., & Huba, G. J. (1984). Depression and physical illness: A multiwave, nonrecursive causal model. Journal of Health and Social Behavior, 25, 350-371.[ISI][Medline]
Arbuckle, J. L. (1996). Full information estimation in the presence of incomplete data. In G. A. Marcoulides & R. E. Schumaker (Eds.), Advanced structural equation modeling: Issues and techniques (pp. 243-277). Mahwah, NJ: Lawrence Erlbaum.
Beck, A. T., & Beck, R. W. (1972). Screening depressed patients in family practice: A rapid technique. Postgraduate Medicine, 52, 81-85.[Medline]
Chen, H., Mechanic, D., & Hansell, S. (1998). A longitudinal study of self-awareness and depressed mood in adolescence. Journal of Youth and Adolescence, 27, 719-734.
Cohen, S., & Rodriguez, M. S. (1995). Pathways linking affective disturbance and physical disorders. Health Psychology, 14, 374-380.[ISI][Medline]
Compas, B. E., Ey, S., & Grant, K. E. (1993). Taxonomy, assessment, and diagnosis of depression during adolescence. Psychological Bulletin, 114, 323-344.[ISI][Medline]
Garber, J. (1984). The developmental progression of depression in female children. In D. Cicchetti, & K. Schneider-Rosen (Eds.), Childhood depression (pp. 29-58). San Francisco: Jossey-Bass.
Gijsbers van Wijk, C. M. T., & Kolk, A. M. (1997). Sex differences in physical symptoms: The contribution of symptom perception theory. Social Science and Medicine, 45, 231-246.
Gotlib, I. H., Lewinsohn, P. M., & Seeley, J. R. (1995). Symptoms versus a diagnosis of depression: Differences in psychosocial functioning. Journal of Consulting and Clinical Psychology, 63, 90-100.[ISI][Medline]
Hankin, B. L., Abramson, L. Y., Moffitt, T. W., Silva, P. A., McGee, R., & Angell, K. E. (1998). Development of depression from preadolescence to young adulthood: Emerging gender differences in a 10-year longitudinal study. Journal of Abnormal Psychology, 107, 128-140.[ISI][Medline]
Hansell, S., & Mechanic, D. (1985). Introspectiveness and adolescent symptom reporting. Journal of Human Stress, 11, 165-176.[ISI][Medline]
Harter, S., & Whitesell, N. R. (1996). Multiple pathways to self-reported depression and psychological adjustment among adolescents. Development & Psychopathology, 8, 761-777.
Idler, E. L., & Benyamini, Y. (1997). Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior, 38, 21-37.[ISI][Medline]
Ingram, R. W. (1990). Self-focused attention in clinical disorders: Review and a conceptual model. Psychological Bulletin, 107, 156-176.[ISI][Medline]
Katon, W., & Sullivan, M. D. (1990). Depression and chronic medical illness. Journal of Clinical Psychiatry, 51, 3-11.
Kovacs, M. (1985). The Children's Depression Inventory. Psychopharmacology Bulletin, 21, 995-998.[ISI][Medline]
Larsen, R. W., Richards, M. H., Moneta, G., Holmbeck, G., & Duckett, E. (1996). Changes in adolescents' daily interactions with their families from ages 10 to 18: Disengagement and transformation. Developmental Psychology, 32, 744-754.
Larson, R., & Richards, M. H. (1989). Introduction: The changing life space of early adolescence (Special issue). Journal of Youth and Adolescence, 18, 501-509.
Levenson, J. L., Hamer, R., Silverman, J. J., & Rossiter, L. F. (1987). Psychopathology in medical inpatients and its relationship to length of hospital stay. International Journal of Psychiatry in Medicine, 16, 231-237.
Lewinsohn, P. M., Gotlib, I. H., & Seeley, J. R. (1995). Adolescent psychopathology: IV. Specificity of psychosocial risk factors for depression and substance abuse in older adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 34, 1221-1229.[ISI][Medline]
Lewinsohn, P. M., Gotlib, I. H., & Seeley, J. R. (1997). Depression-related psychosocial variables: Are they specific to depression in adolescents? Journal of Abnormal Psychology, 106, 365-375.[ISI][Medline]
Lewinsohn, P. M., Roberts, R. E., Seeley, J. R., Rohde, P., Gotlib, I. H., & Hops, H. (1994). Adolescent psychopathology: II. Psychosocial risk factors for depression. Journal of Abnormal Psychology, 103, 302-315.[ISI][Medline]
Lewinsohn, P. M., Seeley, J. R., Hibbard, J., Rohde, P., & Sack, W. H. (1996). Cross-sectional and prospective relationships between physical morbidity and depression in older adolescents. Journal of the Academy of Child and Adolescent Psychiatry, 35, 1120-1129.
Lorig, K., Stewart, A., Ritter, P., Gonzalez, V., Laurent, D., & Lynch, J. (1996). Outcome measures for health education and other health care interventions. Thousand Oaks, CA: Sage Publications.
Mechanic, D., & Hansell, S. (1987). Adolescent competence, psychological well-being, and self-assessed health. Journal of Health and Social Behavior, 28, 364-374.[ISI][Medline]
Muthén, B. O., Kaplan, D., & Hollis, M. (1987). On structural equation modeling with data that are not missing completely at random. Psychometrika, 52, 431-462.
Muthén, L. K., & Muthén, B. O. (1998). Mplus: The comprehensive modeling program for applied users. Los Angeles, CA: Muthén & Muthén.
National Center for Health Statistics. (1994). Vital and health statistics: Current estimates from the National Health Interview Survey, 1992. Series 10, Number 189. Washington, DC: U.S. Government Printing Office.
Neter, J., Wasserman, W., & Kutner, M. H. (1990). Applied linear statistical models: Regression, analysis of variance, and experimental designs (3rd ed.). Homewood, IL: Irwin.
Nolen-Hoeksema, S. (1990). Sex differences in depression. Stanford, CA: Stanford University Press.
Nolen-Hoeksema, S., & Girgus, J. S. (1994). The emergence of gender differences in depression during adolescence. Psychological Bulletin, 115, 424-443.[ISI][Medline]
Nolen-Hoeksema, S., Larson, J., & Grayson, C. (1999). Explaining the gender difference in depressive symptoms. Journal of Personality and Social Psychology, 77, 1061-1072.[ISI][Medline]
Nolen-Hoeksema, S., Morrow, J., & Fredrickson, B. L. (1993). Response styles and the duration of episodes of depressed mood. Journal of Abnormal Psychology, 102, 20-28.[ISI][Medline]
Petersen, A. C., Sarigiani, P. A., & Kennedy, R. E. (1991). Adolescent depression: Why more girls? Journal of Youth and Adolescence, 20, 247-271.
Reinherz, H. Z., Giaconia, R. M., Pakiz, B., Silverman, A. B., Frost, A. K., & Lefkowitz, E. S. (1993). Psychosocial risks for major depression in late adolescence: A longitudinal community study. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 1155-1163.[ISI][Medline]
Rohde, P., Lewinsohn, P. M., & Seeley, J. R. (1990). Are people changed by the experience of having an episode of depression? A further test of the Scar hypothesis. Journal of Abnormal Psychology, 99, 264-271.[ISI][Medline]
Rohde, P., Lewinsohn, P. M., & Seeley, J. R. (1994). Are adolescents changed by an episode of major depression? Journal of the American Academy of Child and Adolescent Psychiatry, 33, 1289-1298.[ISI][Medline]
Sobel, M. E. (1988). Direct and indirect effects in linear structural equation models. In J. S. Long (Ed.), Common problems/proper solutions (pp. 46-64). Beverly Hills, CA: Sage.
Sweeting, H. (1995). Reversals of fortune? Sex differences in health in childhood and adolescence. Social Science and Medicine, 40, 77-90.
Verbrugge, L. M. (1985). Gender and health: An update on hypotheses and evidence. Journal of Health and Social Behavior, 26, 156-182.[ISI][Medline]
Walker, L. S., & Zeman, J. L. (1992). Parental
response to child illness behavior. Journal of Pediatric
Psychology, 17,
49-71.
Wichstrom, L. (1999). The emergence of gender difference in depressed mood during adolescence: The role of intensified gender socialization. Developmental Psychology, 35, 232-245.[ISI][Medline]
Wickrama, K. A. S., Lorenz, F. O., & Conger, R. D. (1997). Parental support and adolescent physical health status: A latent growth-curve analysis. Journal of Health and Social Behavior, 38, 149-163.[ISI][Medline]
Williams, P. G., Holmbeck, G. N., & Greenley, R. N. (in press). Adolescent health psychology. Journal of Consulting and Clinical Psychology.
Williams, P. G., & Wiebe, D. J. (2000). Individual differences in self-assessed health: Gender, neuroticism, and physical symptom reports. Personality and Individual Differences, 28, 823-835.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||