Age-period-cohort analysis of depression trends: are depressive symptoms increasing across generations in Germany?

Abstract

Several studies have examined trends in depression, but only few have explicitly considered possible generational differences. I examined changes in the burden of depressive symptoms between 2002 and 2017 according to age, time period and birth cohort in Germany. I used population-based data drawn from the German Aging Survey (N = 33,723, 54% female, ages 40 +) from 2002, 2008, 2011, 2014, and 2017. Depressive symptoms were measured via the CES-D 15. Hierarchical age-period-cohort models were used to examine trends in depression. I found that depressive symptoms changed across age, time period and birth cohorts. While there was a general decrease across time periods, strong evidence for a U-shaped cohort effect was also found: Younger generations, beginning with cohorts born after the World War II, increasingly report more depressive symptoms than older generations. This U-shaped cohort trend appeared most pronounced for the somatic symptoms subscale. Contrarily, only minimal cohort differences were found regarding the positive affect subscale. Therefore, depressive symptoms, and especially somatic symptoms, seem to increase in more recent birth cohorts in Germany, who might thus be at risk to experience more mental health problems in the future. Potential reasons for these trends and the generalizability of the results to other countries should be investigated by future studies.