Secular trends in age at onset of major depressive disorder in a clinical sample of children.

Using a psychiatrically referred, depressed, school-age sample, we sought to cross-validate the clinically pertinent epidemiologic finding that the distribution of age at onset of first episode of major depressive disorder (MDD) is subject to birth-cohort and period effects. Demographic and historical variables also were considered in attempting to explain the variability in age-at-onset. The results indicated a birth-cohort effect, but no discernable period effect on age at onset of MDD; successive birth cohorts were younger when they first developed MDD even after stringent analyses were conducted that corrected for structural sampling biases in the sample. In view of the relatively small size and clinical nature of the sample and the restricted birth-year span that characterizes children, the results are presented to stimulate further discussion of this topical area.