Abstract
The authors validate the usefulness of the Diagnostic Interview Schedule (DIS) in prevalence studies of major depression (MD) in medically ill male veterans. The affective disorders portion of the DIS was administered by a trained interviewer to a representative sample of 408 medically-ill hospitalized patients (75% over age 70). Using a standardized handscoring method for the DIS, the patients were categorized as having 01 not having MD. Using a two-stage design, sixty-nine patients scoring above a cutoff score on a self-rated depression scale were referred for blind evaluation by a psychiatrist. The psychiatrist made a clinical diagnosis of MD using a standard checklist of DSM-III criteria. The DIS correctly diagnosed MD in 78 percent (14/18) of older and 69 percent (9/13) of younger patients, and correctly ruled out the disorder in 65 percent (15/23) of older and 67 percent (10/15) of younger patients; the kappa statistic of agreement was .40 in older and .36 in younger patients.