Search

This study explored the age trajectories of depressive symptoms across multiple cohort groups who were in middle and late adulthood; examined sociodemographic differences in these trajectories; and investigated how relevant factors contributed to depressive symptoms trends of different cohorts.Drawing on data from the 1994-2016 Health and Retirement Study (HRS), we used growth curve models to examine the age patterns of depressive symptoms, changes in sociodemographic gaps in depressive symptoms trajectories, and predictors of changes in depressive symptoms.In general, adults' depressive symptoms started high in middle-adulthood, declined in young-old life, increased moderately in mid-old life, and peaked in old-old life; In detail, more nuanced cohort-specific age trajectories of depressive symptoms were observed, challenging the prevailing assumption of a common age trajectory of depressive symptoms. Later-born cohorts displayed higher levels of depressive symptoms than earlier-born cohorts at observed ages. Second, we found intra-cohort sociodemographic differences in levels of depressive symptoms, but these differences' growth rates varied by specific factors. Regardless of the cohort group, as people age, the gender gap in depressive symptoms persisted but the partnership gap reduced. A widening educational gap across cohorts was observed, but it declined with age in some cohorts.Results suggest more evidence for the persistent inequality and age-as-leveler hypotheses rather than the cumulative (dis-)advantage hypothesis.Supplemental data for this article can be accessed online at https://doi.org/10.1080/13607863.2021.2010182 .