Patients age 60 years and older (N = 3,767) were screened for depression using the Center for Epidemiological Studies Depression Scale (CES-D). A random sample of patients scoring ≤ 16 on the CES-D (n = 44) and of patients scoring < 16 (n = 81) were recruited to complete a structured psychiatric interview. The estimated prevalence of major depressive disorder (MDD), using a clinically derived diagnosis with the “etiological” method, was 1.78%. When a computer-generated algorithm for MDD using the “inclusive” method for counting somatic symptoms was applied, the estimated prevalence for MDD rose to 5.84%. All subjects with MDD diagnosed by any method had significantly higher scores on the Sickness Impact Profile (SIP), compared with nondepressed subjects.