Natural history can be characterized by incidence, recurrence, and duration of episodes. Research on the incidence of major depression is rare; studies of recurrence and duration are limited to clinical samples.
The Baltimore, Md, site of the Epidemiologic Catchment Area Program followed up its 1981 baseline cohort of 3481 respondents with an additional assessment in 1993 to 1996. Interviews were obtained from 1920 respondents (73% of the survivors). The Diagnostic Interview Schedule and the same survey procedures as in 1981 were used, augmented with a Life Chart Interview for dating the onset and duration of syndromes.
There were 71 new cases of Diagnostic Interview Schedule/DSM-IVmajor depression and 23 698 person-years of exposure, generating an estimated incidence of 3.0 per 1000 per year. Incidence peaked while subjects were in their 30s, with a smaller peak when they were in their 50s. Prodromal symptoms often occurred many years before the full criteria for diagnosis were met. Women were at higher risk for becoming new cases but had neither higher risk for recurrence nor longer episodes than men. Episodes of depression lasted for 12 weeks. The duration of an episode, and time to an episodefree year, was longer in the first episode than in recurrent episodes.
The incidence estimated in this study is consistent with that found in the few other similar studies performed. The bimodality of onset suggests the value of further exploring the heterogeneity of depression via its natural history. Reported differences in prevalence between men and women seem to be due to differences in incidence, not chronicity.