Objective: Using items from two existing depression scales, we have sought to develop a brief self-rated instrument for detecting major depressive disorder (M.D.D.) in medically ill, hospitalized patients.
Method: Forty-two items from the Geriatric Depression Scale (G.D.S.) and Carroll Depression Scale were administered to 559 men under age 40 or over age 70 consecutively admitted to the hospital. Eighty-two M.D.D.'s were diagnosed in this group by structured psychiatric interview. After eliminating 12 items confounded by medical illness, 11 items were selected using regression analysis, correlation with the total score, and factor analysis. The 11-item scale includes an assessment of the five DSM-III-R criteria for M.D.D. which are least confounded by medical illness (mood, suicidal intent, guilt or worthlessness, concentration, and psychomotor agitation). The scale was then tested in 78 medical inpatients who were later assessed for M.D.D. using a structured psychiatric interview.
Results: Ten out of twelve M.D.D.'s were identified (83% sensitivity) and depression excluded in 51 of 66 non-depressed subjects (77% specificity) (compared with 82% sensitivity and 76% specificity for the 30-item G.D.S.). Scores on the 11-item scale were also correlated with the G.D.S. (.92), the Zung Depression Scale (.58), and the C.E.S.-D (.67).
Conclusion: The 11-item scale is a practical tool for clinicians who screen patients for depression and for investigators who need a brief measure of depression in studies involving medical inpatients.