Objective: Estimates of incidence of late-life depression vary greatly with few studies excluding demented cases through in-depth evaluation and most studies failing to control for the effect of mortality and interval treatment. In a large population-based study, the authors examined the effect on incidence of first-onset depressive syndrome to determine whether any gender or age differences in incidence are attenuated with inclusion of these additional measures.

Method: Incidence rates of depressive syndrome per 1,000 person-years are presented for 2,877 nondemented elderly (ages 65 to 100 years) residents of Cache County, Utah. Cases are identified by direct interview methods, by inference from prescription antidepressant medicine use, and by postmortem informant interview for decedents.

Results: In-person interviews yielded incidence rates of first-onset depressive disorder (any type) of 13.09 for men and 19.44 for women. Inclusion of antidepressant users increased these figures to 15.55 for men and 23.30 for women. Addition of postmortem interview data yielded rates of 20.66 for men and 26.29 for women. Individuals with no history of depression had rates for major depression of 7.88 for men and 8.75 for women; minor depression rates were 19.23 for men and 24.46 for women (p=0.691; effect for minor depression p

Conclusions: Incidence of first-onset major depression varies with data source and prior lifetime history of depression. Gender effects apparent in interview data are attenuated when postmortem information and pharmacotherapy were considered.