Abstract
Objective:
Depression is common, impairing, and the leading cause of disease burden in youth. This study aimed to identify the effects of childhood/adolescent depression on a broad range of longer-term outcomes.
Methods:
The analysis is based on the prospective, representative Great Smoky Mountains Study of 1,420 participants. Participants were assessed with the structured Child and Adolescent Psychiatric Assessment interview up to 8 times in childhood (ages 9 to 16; 6,674 observations; 1993 to 2000) for DSM-based depressive disorders, associated psychiatric comorbidities and childhood adversities. Participants were followed up 4 times in adulthood (ages 19, 21, 25, and 30; 4,556 observations of 1,336 participants; 1999 to 2015) with the structured Young Adult Psychiatric Assessment Interview for psychiatric outcomes and functional outcomes.
Results:
7.7% of participants met criteria for a depressive disorder in childhood/adolescence. Any childhood/adolescent depression was associated with higher levels of adult anxiety and illicit drug disorders and also with worse health, criminal, and social functioning; these associations persisted when childhood psychiatric comorbidities and adversities were accounted for. No sex-specific patterns were identified. However, timing of depression mattered: Individuals with adolescent-onset depression had worse outcomes than those with child-onset. Average depressive symptoms throughout childhood and adolescence was associated with more adverse outcomes. Finally, specialty mental health service use was protective against adult diagnostic outcomes.
Conclusion:
Early depression and especially persistent childhood/adolescent depressive symptoms have robust, lasting associations with adult functioning. Some of these effects may be attenuated by service use.
Funding:
National Institutes of Mental Health, Drug Abuse, and Child Health and Development.
Keywords: Childhood, depression, longitudinal, psychiatric disorders, adult outcomes, epidemiologic
Lay Summary
This study used data from the 25+ year Great Smoky Mountains Study to identify which aspects of adult functioning are affected by childhood/adolescent depression. Childhood depression was associated with increased risk for adult anxiety and illicit drug problems and compromised adult health and social functioning. Children with persistent depressed mood had the worst adult outcomes. At the same time, mental health services use within childhood was protective against adult anxiety.
Introduction
Among youth aged 10 to 24 years, the leading cause of disease burden worldwide is unipolar depressive disorders, with an even greater burden in high-income countries.1 Two recent systematic reviews and meta-analyses by Colman and colleagues found that 1) there is a sufficient literature on adult associations of adolescent depression to support meta-analyses, 2) adolescent depression is associated with adult depression and anxiety, and 3) depression is also significantly associated with reduced educational attainment, unemployment and early parenthood.2,3 Meta-analytic work has the benefit of sampling from multiple populations around the world, having distinct assessment procedures and follow-up periods, and providing a robust synthesis of multiple decades of prospective-longitudinal work. Results from such work suggest that the public health impact of adolescent depression may be greater than estimated by the Global Burden of Diseases study, because depression is both a disease in itself and a possible risk factor for future disease and impairment.
The previous systematic reviews/meta-analyses provide a strong foundation, but key gaps in knowledge about the long-term effects of early depression remain. Many of the studies included in the meta-analyses/reviews did not account for a broad range of potential confounds of adolescent depression-adult outcomes associations. Indeed, a third of studies did not account for any covariates; more than half did not adjust for sex or socioeconomic status. Studies accounting for a range of covariates, often reported mixed results. Additional work is needed to clarify which effects of childhood/adolescent depression on adult outcomes are independent of other childhood experiences—including adversity and other psychiatric disorders.
Next, the meta-analyses/reviews were unable to identify subgroups at particular risk for poor adult outcomes after suffering from childhood/adolescent depression Sex-specific effects are particularly of interest given that both the prevalence and risk factors for depression in early life vary for male and female individuals.4–7 Adult outcomes may also differ by the ‘dose’ of depression that the individual was exposed to. Multiple exposures to depression or depressive symptoms may be associated with worse outcomes than those with one-time or time-limited exposures.8,9 Few studies have followed children across multiple childhood and adolescent assessments to capture cumulative exposure to depressive symptoms. Finally, many studied included in the meta-analyses had measured depression in mid- or late adolescence (mean age - 15.6), but depression often develops prior to or during early/mid puberty.2 Indeed, pre-pubertal depression has been associated with distinct biomarkers and potentially higher levels of familiarity and early risk factors in some work, suggesting a potential developmental subtype,10,11 which could be associated with distinct adult outcomes.
The current study uses the 2+ decade prospective-longitudinal Great Smoky Mountains Study (GSMS) to address challenges identified by the two reviews, including 1) potential confounding, 2) sex-specific associations, 3) cumulative exposure to depression across childhood, and 4) the outcomes of pre/postpubertal depression onset. Together, this analysis builds on the replicated main effect findings to identify clinically relevant predictors of a broad range of longer-term outcomes.
Perhaps the most important question about early depression is whether potentially poor outcomes can be averted by early intervention. Here, the literature is not promising. Using a benchmarking strategy, Weersing and Weisz showed that outcomes of depressed children receiving community care more closely resembled the control condition than of those treated with an active intervention.12 While effective interventions exist for childhood and adolescent depression,13 the effects tend to be modest, there is little evidence of sustained effects beyond a year and, and, like in many areas, translation of empirically-supported treatment to community practice is lacking.14 This study will test whether community treatment for childhood and adolescent depression improves adult outcomes.
Methods
Participants
This report follows the STROBE reporting guidelines for cohort studies.15 The Great Smoky Mountains Study is a longitudinal, representative study of children in 11 predominantly rural counties of North Carolina (see16). Three cohorts of children, ages 9, 11, and 13 years, were recruited from a pool of some 12,000 children using a two-stage sampling design, resulting in N = 1,420 participants (49% female participants; see also17). First, potential participants were randomly selected from the population using a household equal probability design. Next, participants were screened for risk of psychopathology and participants screening high were oversampled in addition to a random sample of the rest. In addition, American Indians were oversampled to constitute 25% of the sample. Sampling weights were applied to adjust for differential probability of selection and to allow results to generalize to the broader population. See ascertainment figure S1 and16–18 for additional detail.
Annual assessments were completed on the 1,420 children until age 16 (6,674 observations of 1,420 individuals; 1993 to 2000) and then again at ages 19, 21, 25, and 30 (4,556 observations of 1,336 participants; 1999 to 2015) for a total of 11,230 total assessments. Interviews were completed separately with a parent figure and the participant until age 16, and with the participant only thereafter. The study protocol and consent consent/assent forms were approved by the Duke University Medical Center Institutional Review Board.
Childhood variables:
All childhood constructs were assessed using information collected from participants and parents using the structured Child and Adolescent Psychiatric Assessment (CAPA).19,20
Depression status.
Participants and parents were interviewed with the depressive disorders module of the CAPA. Responses from this interview were then used to generate DSM diagnoses from the symptoms assessed by computer algorithms. If either parent or child reported a symptom as present in the past three months, it was counted as present. A three-month “primary period” was selected because longer recall periods are associated with forgetting and recall bias.21–23 Three depression diagnoses were considered: DSM-IV major depressive episode, dysthymia, and Depression Not Otherwise Specified (NOS). Diagnostic one-week test-retest reliabilities for child self-reports were k=.90 for major depression, k=.85 for dysthymia, k=.63 for depression-NOS.
Childhood psychiatric disorders, service use other adversities/hardships.
Psychiatric disorders assessed included anxiety disorders, conduct disorder (CD), oppositional defiant disorder (ODD), attention-deficit/hyperactivity disorder (ADHD), and substance use disorders. Twenty-one types of mental health service use were identified by using the Child and Adolescent Services Assessment (CASA)24. Test-retest reliability and validity of the CAPA diagnoses are similar to other psychiatric interviews.20,25 The following categories of family hardships or childhood adversities were assessed at each observation: a) low socioeconomic status (SES); b) unstable family structure (e.g., single parent family, divorce, presence of step-parent); c) family dysfunction (e.g., inadequate parental supervision, domestic violence, maternal depression); d) family maltreatment including physical abuse, sexual abuse and neglect; and e) peer victimization. Additional detail is available in supplemental 1, available online.
Adult variables:
All outcomes except where noted (e.g., official criminal records) were assessed using the Young Adult Psychiatric Assessment (YAPA),26 an upward extension of the CAPA interview administered to the participants. The assessment of adult psychiatric disorders resembled that of childhood disorders, but relied upon self-report only. Disorders included any DSM anxiety disorder, depressive disorder, nicotine use disorder, alcohol use disorder, cannabis use disorder, illicit drug disorders, and antisocial personality disorder (ASPD). Psychosis and bipolar disorder were not included in analyses due to very low prevalence (< 1 %) in the community. The participant was positive for diagnosis if criteria were met at any adult observation. Standardized scales were derived to provide a broad profile of adult functioning across four domains: health, risky/illegal behaviors, wealth (financial/educational), and social function. These scales were summed from dichotomous indicators in each domain (e.g., college completion for wealth, smoking status for health). In some cases, indicators were coded positive if reported at any point in adulthood; in other cases (e.g., educational attainment) the last observation was used to determine status. Standardized scores were obtained by subtracting the individual score from the group mean and dividing the resultant score by the standard deviation. A full description of all indicators used to construct these scales is available in supplemental 1, available online.
Statistical Analysis
All analyses included sampling weights inversely proportional to each participant’s probability of selection. Models used the generalized estimating equations option within SAS PROC GENMOD to derive robust variance (sandwich type) estimates to adjust standard errors for the stratified design. Weighted logistic (for binary outcomes such as psychiatric status) and linear (for continuous variables such as the z-scores for the adult function scales) regression models were used to look at differences in adult outcomes by childhood/adolescent depression status. All percentages provided in the results are weighted and Ns/sample sizes are unweighted. Findings are considered statistically significant at p < 0.05.
Missing Data
Across all assessments, 83% of possible interviews were completed (table S1). All 1,420 participants were interviewed at least once in childhood (ages 9 to 16); 1,260 participants (88.7%) had 3+ childhood observations. Of the total sample, 1,336 (94.0%) were followed up at least once in adulthood at ages 19, 21, 25, or 30. Childhood/adolescent depression was not associated with lower levels of participation in adulthood (p=0.32) suggesting no differential dropout.
Results
Depression in Childhood and Adolescence
The point-prevalence of depression at any single childhood observation was 2.2% (95% CI 1.5–2.9). When aggregating across ages 9 to 16 (i.e., across up to 7 assessments), 7.7% (n=140) of participants had met full criteria for a depressive disorder by age 16, with 2.2% (n=31) meeting criteria at >1 observation. Childhood/adolescent depression status was strongly associated with all types of other childhood psychiatric disorders and with most types of childhood adversities (see table 1). Childhood comorbidities and adversities were included as covariates in subsequent analyses.
Table 1.
Prevalence and unadjusted associations between childhood/adolescent depression status and childhood psychiatric problems and adversities.
| No depression | Depression | Unadjusted associations | |||
|---|---|---|---|---|---|
| % (n) | % (n) | OR | CI | p | |
| Total | 92.3% (1280) | 7.7% (140) | |||
| % Female | 48.2% (556) | 56.9% (74) | 1.4 | 0.8–2.6 | 0.26 |
| White | 89.1% (878) | 93.4% (105) | ref | ref | ref |
| African-American | 7.3% (81) | 2.8% (7) | 0.4 | 0.2–0.9 | 0.02 |
| American Indian | 3.7% (321) | 3.8% (28) | 1.0 | 0.6–1.6 | 0.97 |
| Psychiatric disorders | |||||
| Any anxiety dx. | 7.5% (129) | 46.0% (65) | 10.6 | 5.5–20.4 | <0.001 |
| ADHD | 2.7% (58) | 10.9% (20) | 4.4 | 1.8–11.0 | 0.001 |
| ODD | 6.7% (163) | 46.3% (75) | 12.1 | 6.4–22.8 | <0.001 |
| CD | 6.9% (158) | 28.5% (45) | 5.4 | 2.7–10.6 | <0.001 |
| Substance dx. | 4.7% (78) | 25.5% (30) | 6.9 | 3.1–15.4 | <0.001 |
| Adversities | |||||
| Low family SES | 33.7% (584) | 38.1% (65) | 1.2 | 0.7–2.3 | 0.54 |
| Family instability | 25.2% (405) | 38.2% (65) | 1.8 | 1.0–3.4 | 0.05 |
| Family dysfunction | 24.7% (383) | 53.4% (93) | 3.5 | 1.9–6.5 | <0.001 |
| Maltreatment | 18.0% (343) | 51.0% (86) | 4.8 | 2.6–8.8 | <0.001 |
| Peer victimization | 24.9% (357) | 41.0% (64) | 2.1 | 1.1–3.9 | 0.02 |
Note: All percentages are weighted and ns are unweighted. ADHD = attention-deficit/hyperactivity disorder; CD = conduct disorder; CI = 95% confidence intervals; Dx = disorder; ODD = oppositional defiant disorders; OR = odds ratio.
Bolded odds ratios are significant at p <0.05.
Associations with adult psychiatric status and functioning
Childhood/adolescent depression was associated with elevated rates of adult anxiety and depression and substance abuse disorders (table 2). In models adjusted for childhood disorders and adversities, associations between childhood/adolescent depression and adult depression and substance disorders were no longer statistically significant. Childhood/adolescent depression status was also associated with worse adult functional outcomes: physical health, risky/criminal behavior, financial/educational, and social function. Associations with health, risky/criminal, and social functioning continued to be significant after adjusting for childhood covariates, although the sizes of the associations were modestly attenuated.
Table 2.
Prevalence and associations between childhood/adolescent depression status and adult (ages 19 to 30) psychiatric status and functional outcomes
| No depression | Depression | Associations adjusted for sex, race | Associations adjusted for sex, race, psych. dx. and adversities | Sex inter.? | |||||
|---|---|---|---|---|---|---|---|---|---|
| % (n) | % (n) | OR | CI | p | OR | CI | p | ||
| Adult Psychiatric problems | |||||||||
| Any disorder | 43.4% (505) | 76.1% (87) | 4.6 | 2.5–8.7 | <0.001 | 2.5 | 1.2–5.4 | 0.02 | 0.50 |
| Any anxiety dx. | 14.3% (170) | 48.7% (47) | 5.9 | 3.0–11.6 | <0.001 | 3.3 | 1.6–6.8 | 0.001 | 0.73 |
| Any depressive dx. | 8.9% (128) | 26.7% (29) | 3.9 | 1.8–8.4 | <0.001 | 1.8 | 0.8–4.1 | 0.17 | 0.63 |
| Any substance dx. | 32.3% (370) | 45.2% (63) | 2.0 | 1.1–3.7 | 0.02 | 1.1 | 0.4–2.6 | 0.83 | 0.15 |
| Adult Functional outcomes | M (SD) | M (SD) | β | CI | p | β | CI | p | Sex diff? |
| Physical Health | 0.07 (0.97) | −0.83 (0.95) | 0.89 | 0.55–1.22 | <0.001 | 0.48 | 0.13–0.84 | 0.008 | 0.74 |
| Risky/criminal behavior | 0.04 (0.99) | −0.52 (0.92) | 0.65 | 0.40–0.90 | <0.001 | 0.29 | 0.02–0.56 | 0.04 | 0.94 |
| Financial/educational | 0.05 (0.99) | −0.58 (0.94) | 0.68 | 0.38–0.99 | <0.001 | 0.31 | −0.03–0.65 | 0.08 | 0.36 |
| Social | 0.06 (0.99) | −0.61 (0.94) | 0.69 | 0.37–1.01 | <0.001 | 0.38 | 0.05–0.71 | 0.02 | 0.14 |
All percentages are weighted and ns are unweighted. All models adjusted for sex and race/ethnicity. Psychiatric diagnoses include childhood anxiety, ADHD, conduct, oppositional defiant and substance use disorders. Childhood adversities include low SES, familial instability, family dysfunction, maltreatment, and peer victimization. CI = 95% confidence intervals; Dx = disorder; OR = odds ratio.
Bolded odds ratios are significant at p <0.05.
All analyses were rerun looking only at the latter observations (ages 25 and 30) to see if associations attenuated with time (table S2). Early depression continued to be significantly associated with diagnostic, substance, health, and social outcomes, but not with risky/criminal behavior or financial/educational outcomes, in these later observations.
Groups at increased risk for adult outcomes
Moderation by sex was tested with an interaction term between sex and childhood/adolescent depression status. There was little evidence that associations between depression and adult psychiatric or functional outcomes varied by sex (see last column in tables 2 and 3). Analyses were also rerun for males and female participants separately to facilitate future reviews/meta-analyses (tables S3 and S4).
Table 3.
Associations between average depressive symptoms during childhood/adolescence and adult psychiatric problems.
| Associations adjusted for sex, race | Associations adjusted for sex, race, psych. dx. and adversities | |||||
|---|---|---|---|---|---|---|
| β | 95% CI | p | β | 95% CI | p | |
| Psychiatric problems | ||||||
| Any disorder | 0.70 | 0.41–0.99 | <0.001 | 0.36 | 0.02–0.70 | 0.04 |
| Any anxiety dx. | 1.22 | 0.88–1.56 | <0.001 | 0.96 | 0.57–1.35 | <0.001 |
| Any depressive dx. | 0.86 | 0.51–1.21 | <0.001 | 0.50 | 0.07–0.93 | 0.02 |
| Any substance dx. | 0.27 | 0.02–0.52 | 0.04 | −0.05 | −0.38–0.28 | 0.78 |
| Functional outcomes | β | 95% CI | p | β | 95% CI | p |
| Physical health | 0.44 | 0.29–0.59 | <0.001 | 0.26 | 0.08–0.44 | 0.004 |
| Risky/criminal behavior | 0.23 | 0.13–0.33 | <0.001 | 0.05 | −0.06–0.16 | 0.42 |
| Financial/educational | 0.35 | 0.22–0.48 | <0.001 | 0.19 | 0.05–0.32 | 0.008 |
| Social | 0.35 | 0.22–0.48 | <0.001 | 0.22 | 0.08–0.35 | 0.002 |
All models adjusted for sex and race/ethnicity. Psychiatric diagnoses include childhood anxiety, ADHD, conduct, oppositional defiant and substance use disorders. Childhood adversities include low SES, familial instability, family dysfunction, maltreatment, and peer victimization. CI = 95% confidence intervals; Dx = disorder.
Bolded values are significant at p <0.05.
Depression is often a persistent cluster of symptoms that recurs over time. Associations were retested with two definitions to test the cumulative effect of depression exposure: the number of childhood/adolescent observations at which participants met criteria for a depressive disorder (count distribution) and the average symptom score across all available childhood/adolescent observations (normal distribution). Both of these definitions resulted in similar psychiatric and functional associations that were observed with the dichotomous diagnostic variable (table 3) for mean depressive symptoms and table S5 for cumulative depressive episodes). Average depressive symptoms also showed associations with adult depression and functional outcomes not observed with the dichotomous diagnostic variable. Average depressive symptoms continued to show associations with diagnostic and functional variables even among individuals that did not meet criteria for an early depression diagnosis (results not shown).
With its repeated observations in childhood and adolescence, this study could examine whether associations with adult outcomes differed depending on when depression was first diagnosed (childhood ≤ 12 years old; adolescent ≥ 13 years old). For most outcomes, there were no differences in risk based on age-of-onset. When differences were identified (e.g., anxiety disorders and social problems), adolescent-onset depression was more strongly associated with adult outcomes compared to childhood-onset depression (table 4).
Table 4.
Test of whether child and adolescent onset depression were differentially associated with adult psychiatric and functional outcomes.
| No depression | Child-onset | Adolescent-onset | Unadjusted associations b/t child and adol. onset | Adjusted associations b/t child and adol. onset | |
|---|---|---|---|---|---|
| % (n) | % (n) | %(n) | p | p | |
| Psychiatric dx. | 1280 | 59 | 80 | ||
| Any disorder | 43.4% (505) | 73.2% (36) | 78.2% (51) | 0.17 | 0.02 |
| Any anxiety dx. | 14.3% (170) | 38.0% (19) | 55.3% (28) | 0.26 | 0.01 |
| Any depressive dx. | 8.9% (128) | 23.6% (14) | 28.7% (15) | 0.39 | 0.06 |
| Any substance dx. | 32.3% (370) | 43.9% (25) | 46.3% (38) | 0.06 | 0.13 |
| Functional outcomes | M (SD) | M (SD) | M (SD) | p | p |
| Health | 0.07 (0.97) | −0.76 (0.91) | −0.90 (1.00) | 0.11 | 0.06 |
| Risky/criminal behavior | 0.04 (0.99) | −0.35 (0.85) | −0.69 (0.98) | 0.70 | 0.94 |
| Financial/educational | 0.05 (0.99) | −0.33 (0.85) | −0.82 (1.00) | 0.42 | 0.12 |
| Social | 0.06 (0.99) | −0.43 (0.92) | −0.79 (0.93) | 0.008 | 0.01 |
All models adjusted for sex and race/ethnicity, childhood psychiatric diagnoses, and childhood adversities. Dx = disorder.
Bolded values are significant at p <0.05.
Protective effect of childhood mental health service use
Sixty-three point one percent of participants with childhood/adolescent depression reported receiving some type of service in the community; 34.5% reported receiving specialty mental health services (e.g., private professional, community mental health center). Despite the limited sample size (i.e., only those who met criteria for depression), we identified a positive effect of treatment of depression during childhood on adult diagnostic status (table 5). Use of specialty mental health services among children/adolescents with depression was associated with lower risk for any adult psychiatric disorder and anxiety disorder specifically. There was little evidence that service use moderated the effect of early depression on functional outcomes.
Table 5.
Associations of childhood mental health service use with adult outcomes in participants with childhood depression.
| Any Service Use | Specialty Mental Health Service Use | |||||
|---|---|---|---|---|---|---|
| None | Some | None | Some | |||
| Psychiatric dx. | % (n) | % (n) | p | % (n) | % (n) | p |
| Any disorder | 88.7 (27) | 68.7 (60) | 0.11 | 85.7 (51) | 57.6 (36) | 0.003 |
| Any anxiety dx. | 58.1 (15) | 43.2 (32) | 0.34 | 57.5 (26) | 31.8 (21) | 0.002 |
| Any depressive dx. | 26.9 (8) | 26.6 (21) | 0.08 | 27.3 (14) | 25.5 (15) | 0.06 |
| Any substance dx. | 54.6 (18) | 39.8 (45) | 0.49 | 48.8 (36) | 38.4 (27) | 0.78 |
| Functional outcomes | M (SD) | M (SD) | M (SD) | M (SD) | ||
| Health | −0.75 (1.22) | −0.87 (0.81) | 0.76 | −0.74 (0.97) | −1.00 (0.91) | 0.44 |
| Risky/criminal behavior | −0.72 (1.09) | −0.41 (0.83) | 0.31 | −0.55 (0.94) | −0.48 (0.90) | 0.77 |
| Financial/educational | −0.72 (1.15) | −0.49 (0.84) | 0.23 | −0.64 (0.99) | −0.45 (0.88) | 0.34 |
| Social | −0.61 (1.20) | −0.61 (0.81) | 0.99 | −0.60 (1.00) | −0.63 (0.86) | 0.93 |
All models adjusted for sex and race/ethnicity, childhood psychiatric diagnoses, and childhood adversities. Bolded values are significant at p <0.05.
Discussion
This study examined longer-term outcomes of early depression based on a well-characterized community sample of children who were prospectively followed up to 11 times over 20+ years. Several findings are particularly noteworthy. Childhood and adolescent depression was associated with broad effects on psychiatric, substance and functional outcomes in adulthood. These associations generally persisted when accounting for childhood psychiatric comorbidities (e.g., anxiety and conduct disorder) and adversities (e.g., maltreatment and peer victimization). The depression-adult outcome associations did not show a clear sex-specific pattern. Risk for poor adult outcomes was elevated for those with higher levels of cumulative depression exposure and for those with an adolescent (as compared to a child) onset. Finally, use of childhood specialty mental health services was associated with reduced risk of adult diagnoses, particularly anxiety but not with improved functional outcomes.
This study had the opportunity to add to recent work suggesting that the effects of adolescent depression persist into adulthood and extent to various functional outcomes. In this study, early depression – whether measured by diagnostic levels or average symptoms – was associated with many psychiatric, substance use, and functional outcomes, and these associations were generally robust to adjustment for other markers of early risk. These findings provide additional evidence that 1) the effects of early depression may be independent of other adversities, 2) the magnitude of these effects is moderate, and 3) the cumulative exposure to depression and depressive symptoms is as important as whether a child/adolescent ever met criteria for depression. It is not merely a diagnosis of depression that is associated with long-term outcomes, but the persistence of elevated levels of depressive symptoms.
Much less is known about who is most affected by childhood depression. There was no clear evidence for sex differences in long-term outcomes of depression in our study. This might be surprising given that female individuals are twice as likely to be depressed in adolescence and some risk factors have been sex-specific.6,27,28 A ‘gender paradox’ might be anticipated wherein the gender group less likely to be disordered (i.e., males) has a more severe form or presentation of the disorder.29,30 Other longitudinal studies of early-onset depression have failed to identify sex-specific associations.9,31,32 Although the mechanisms by which depression emerges differ by sex, the mechanisms by which depression affects later outcomes may be shared.
Adolescent-onset depression was more likely to be associated with increased levels of a few adult outcomes (e.g., anxiety disorder and social problems) than childhood-onset depression. This runs counter to previous notions of child-onset depression as a distinct subtype of depression with the poorest prognosis.11,33,34 Contrariwise, we found that the recency of the diagnosis was a stronger predictor of later functioning. Overall, our interaction analyses showed a dearth of group differences limiting the ability of clinicians to provide patients with individualized risk predictions.
Our findings were more promising (and surprising) with respect to the potentially protective effects of service use. An extensive literature has reported the limited effects of community-based treatments for childhood psychiatric disorders including depression even within childhood.12–14 Yet, in our study, children/adolescents who had met criteria for depression and had also received specialty mental health services were almost half as likely to receive an anxiety diagnosis as adults compared to their depressed peers who did not receive specialty mental health services (i.e., 31.8% versus 57.5%). Adult anxiety rates were still higher than in the non-depressed group, but significantly decreased compared to untreated youth. The beneficial effects of service use did not generalize to functional outcomes, however. Future research in community samples should examine possible mediators of the effects of childhood service use on adult outcomes.
This study is representative of a rural area in the Southeast US, but not of the US population. Although children were assessed repeatedly for depression, some participants may have met depression criteria prior to the initial assessment or between assessments. Additional unmeasured childhood variables including genetic vulnerability may account for observed associations. Thus, these results are not to be interpreted as suggesting a causal effect of early depression on these adult outcomes. Onset of some adverse adult outcomes effects may not be evident until later in life (e.g., chronic diseases). Most adult outcomes were assessed with self-report information only. Finally, detecting significant interactions is difficult when parsing modest cell counts further for subgroup analyses.
Our results affirm the public health burden of early depression and depressive symptoms, but, despite our efforts, make only a limited contribution to our understanding of individual risk. Depression – like other adverse childhood experiences – may increase future risk in a nonspecific way that belies hope for individualized prognoses and treatment planning. Some potential effects may be attenuated by early service use, but this profile (long-term plus limited identification of vulnerable subgroups) suggests that efforts to reduce the public health burden of depression should focus on reducing children’s cumulative exposure to depression and depressive symptoms. The optimal strategy will likely involve public policies that target psychosocial risk factors associated with depression symptoms (e.g., caregiver instability, maltreatment).4,35 Here, there is some reason for optimism as the risk factors for depression are also shared by a variety of other childhood problems allowing for potentially pleiotropic effects on well-being.
Supplementary Material
References
- 1.Gore FM, Bloem PJN, Patton GC, et al. Global burden of disease in young people aged 10–24 years: a systematic analysis. The Lancet. 377(9783):2093–2102. [DOI] [PubMed] [Google Scholar]
- 2.Clayborne ZM, Varin M, Colman I. Systematic Review and Meta-Analysis: Adolescent Depression and Long-Term Psychosocial Outcomes. Journal of the American Academy of Child & Adolescent Psychiatry. 2019;58(1):72–79. [DOI] [PubMed] [Google Scholar]
- 3.Johnson D, Dupuis G, Piche J, Clayborne Z, Colman I. Adult mental health outcomes of adolescent depression: A systematic review. Depression and Anxiety. 2018. [DOI] [PubMed] [Google Scholar]
- 4.Shanahan L, Copeland WE, Costello EJ, Angold A. Child-, adolescent- and young adult-onset depressions: differential risk factors in development? Psychol Med. 2011;41(11). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Angold A, Costello EJ, Worthman CM. Puberty and depression: The roles of age, pubertal status, and pubertal timing. Psychological Medicine. 1998;28:51–61. [DOI] [PubMed] [Google Scholar]
- 6.Hankin BL, Abramson LY. Development of gender differences in depression: An elaborated cognitive vulnerability-transactional stress theory. Psychological Bulletin. 2001;127(6):773–796. [DOI] [PubMed] [Google Scholar]
- 7.Copeland WE, Worthman C, Shanahan L, Costello EJ, Angold A. Early pubertal timing and testosterone associated with higher levels of adolescent depression in girls. Journal of the American Academy of Child & Adolescent Psychiatry. 2019;58(12):1197–1206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Colman I, Wadsworth ME, Croudace TJ, Jones PB. Forty-year psychiatric outcomes following assessment for internalizing disorder in adolescence. American Journal of Psychiatry. 2007;164(1):126–133. [DOI] [PubMed] [Google Scholar]
- 9.Jonsson U, Bohman H, von Knorring L, Olsson G, Paaren A, von Knorring AL. Mental health outcome of long-term and episodic adolescent depression: 15-year follow-up of a community sample. Journal of Affective Disorders. 2011;130(3):395–404. [DOI] [PubMed] [Google Scholar]
- 10.Weissman MM, Wolk S, Wickramaratne P, et al. Children with prepubertal-onset major depressive disorder and anxiety grown up. Archives of General Psychiatry. 1999;56:794–801. [DOI] [PubMed] [Google Scholar]
- 11.Kaufman J, Martin A, King R, Charney D. Are child-, adolescent-, and adult-onset depression one and the same disorder? Biological Psychiatry. 2001;49:980–1001. [DOI] [PubMed] [Google Scholar]
- 12.Weersing VR, Weisz JR. Community clinic treatment of depressed youth: benchmarking usual care against CBT clinical trials. Journal of consulting and clinical psychology. 2002;70(2):299. [DOI] [PubMed] [Google Scholar]
- 13.Weisz JR, McCarty CA, Valeri SM. Effects of psychotherapy for depression in children and adolescents: a meta-analysis. Psychological bulletin. 2006;132(1):132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Weisz JR, Krumholz LS, Santucci L, Thomassin K, Ng MY. Shrinking the Gap Between Research and Practice: Tailoring and Testing Youth Psychotherapies in Clinical Care Contexts. Annual Review of Clinical Psychology. 2015;11(1):139–163. [DOI] [PubMed] [Google Scholar]
- 15.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: Guidelines for Reporting Observational Studies. Annals of Internal Medicine. 2007;147(8):573–577. [DOI] [PubMed] [Google Scholar]
- 16.Copeland WE, Angold A, Shanahan L, Costello EJ. Longitudinal Patterns of Anxiety From Childhood to Adulthood: The Great Smoky Mountains Study. Journal of the American Academy of Child and Adolescent Psychiatry. 2014;53(1):21–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Costello EJ, Mustillo S, Erkanli A, Keeler G, Angold A. Prevalence and development of psychiatric disorders in childhood and adolescence. Archives of General Psychiatry. 2003;60:837–844. [DOI] [PubMed] [Google Scholar]
- 18.Costello EJ, Angold A, Burns B, et al. The Great Smoky Mountains Study of Youth: Goals, designs, methods, and the prevalence of DSM-III-R disorders. Archives of General Psychiatry. 1996;53:1129–1136. [DOI] [PubMed] [Google Scholar]
- 19.Angold A, Costello E. The Child and Adolescent Psychiatric Assessment (CAPA). Journal of the American Academy of Child and Adolescent Psychiatry. 2000;39:39–48. [DOI] [PubMed] [Google Scholar]
- 20.Angold A, Costello EJ. A test-retest reliability study of child-reported psychiatric symptoms and diagnoses using the Child and Adolescent Psychiatric Assessment (CAPA-C). Psychological Medicine. 1995;25:755–762. [DOI] [PubMed] [Google Scholar]
- 21.Hardt J, Rutter M. Validity of adult retrospective reports of adverse childhood experiences: review of the evidence. Journal of Child Psychology and Psychiatry. 2004;45(2):260–273. [DOI] [PubMed] [Google Scholar]
- 22.Patten SB. Recall bias and major depression lifetime prevalence. Social Psychiatry & Psychiatric Epidemiology. 2003;38(6):290–296. [DOI] [PubMed] [Google Scholar]
- 23.Coughlin SS. Recall bias in epidemiologic studies. Journal of Clinical Epidemiology. 1990;43:87–91. [DOI] [PubMed] [Google Scholar]
- 24.Ascher BH, Farmer EMZ, Burns BJ, Angold A. The Child and Adolescent Services Assessment (CASA): Description and psychometrics. Journal of Emotional and Behavioral Disorders. 1996;4:12–20. [Google Scholar]
- 25.Angold A, Costello EJ. The Child and Adolescent Psychiatric Assessment (CAPA). Journal of the American Academy of Child and Adolescent Psychiatry. 2000;39:39–48. [DOI] [PubMed] [Google Scholar]
- 26.Angold A, Cox A, Prendergast M, et al. The Young Adult Psychiatric Assessment (YAPA). Durham, NC: Duke University Medical Center;1999. [Google Scholar]
- 27.Kendler KS, Gardner CO, Prescott CA. Toward a Comprehensive Developmental Model for Major Depression in Men. Am J Psychiatry. 2006;163(1):115–124. [DOI] [PubMed] [Google Scholar]
- 28.Kendler KS, Gardner CO, Prescott CA. Toward a comprehensive developmental model for major depression in women. American Journal of Psychiatry. 2002;159(7):1133–1145. [DOI] [PubMed] [Google Scholar]
- 29.Wasserman GA, McReynolds LS, Ko SJ, Katz LM, Carpenter JR. Gender Differences in Psychiatric Disorders at Juvenile Probation Intake. American Journal of Public Health. 2005;95(1):131–137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Tiet QQ, Wasserman GA, Loeber R, McReynolds LS, Miller LS. Developmental and sex differences in types of conduct problems. Journal of Child and Family Studies. 2001;10(2):181–197. [Google Scholar]
- 31.McLeod G, Horwood L, Fergusson D. Adolescent depression, adult mental health and psychosocial outcomes at 30 and 35 years. Psychological medicine. 2016;46(7):1401–1412. [DOI] [PubMed] [Google Scholar]
- 32.Fergusson DM, Horwood LJ, Ridder EM, Beautrais AL. Subthreshold depression in adolescence and mental health outcomes in adulthood. Archives of general psychiatry. 2005;62(1):66–72. [DOI] [PubMed] [Google Scholar]
- 33.Kovacs M, Devlin B, Pollock M, Richards C, Mukerji P. A controlled family history study of childhood-onset depressive disorder. Archives of General Psychiatry. 1997;54:613–623. [DOI] [PubMed] [Google Scholar]
- 34.Wickramaratne PJ, Greenwald S, Weissman MM. Psychiatric disorders in the relatives of probands with prepubertal-onset or adolescent-onset major depression. Journal of the American Academy of Child and Adolescent Psychiatry. 2000;39:1396–1405. [DOI] [PubMed] [Google Scholar]
- 35.Jaffee SR, Moffitt TE, Caspi A, Fombonne E, Poulton R, Martin J. Differences in early childhood risk factors for juvenile-onset and adult-onset depression. Archives of General Psychiatry. 2002;59:215–222. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.