Adolescent depression and subsequent earnings across early to middle adulthood: A 25-year longitudinal cohort study

Journal article, 2020

Aims: The few available studies on early-onset depression and future earnings offer ambiguous findings, and potential sources of heterogeneity are poorly understood. We examined the differences in adult earnings of males and females with and without a history of depressive disorder in adolescence, with specific focuses on (1) future earnings in clinical subtypes of adolescent depression; (2) the growth and distribution of earnings over time within these subgroups and (3) the mediating role of subsequent depressive episodes occurring in early adulthood. Methods: Data were drawn from the Uppsala Longitudinal Adolescent Depression Study, a community-based cohort study initiated in Uppsala, Sweden, in the early 1990s. Comprehensive diagnostic assessments were conducted at age 16-17 and in follow-up interviews 15 years later, while consecutive data on earnings for the years 1996 to 2016 (ages 20-40) were drawn from population-based registries. The current study included participants with a history of persistent depressive disorder (PDD) (n = 175), episodic major depressive disorder (MDD) (n = 82), subthreshold depression (n = 64) or no depression (n = 218) in adolescence. The association of adolescent depression with earnings in adulthood was analysed using generalised estimating equations. Estimates were adjusted for major child and adolescent psychiatric comorbidities and parental socioeconomic status. The indirect (mediated) effect of depression in early adulthood (ages 19-30) on earnings in mid-adulthood (31-40) was estimated in mediation analysis. The study followed the 'STrengthening the Reporting of OBservational studies in Epidemiology' (STROBE) guidelines. Results: Earnings across early to middle adulthood were lower for participants with a history of a PDD in adolescence than for their non-depressed peers, with an adjusted ratio of mean earnings of 0.85 (0.77-0.95) for females and 0.76 (0.60-0.95) for males. The differences were consistent over time, and more pronounced in the lower percentiles of the earnings distributions. The association was partially mediated by recurrent depression in early adulthood (48% in total; 61% for females, 29% for males). No reduction in earnings was observed among participants with episodic MDD in adolescence, while results for subthreshold depression were inconclusive. Conclusions: Our findings suggest that future earnings of adolescents with depressive disorders are contingent on the duration and natural long-term course of early-onset depression, emphasising the need for timely and effective interventions to avoid loss of human capital.