Measurement of depression in alcoholics.

Estimates of the prevalence of depression among alcoholics vary widely, partly because of different methods used to define depression. To assess the specificity, sensitivity, change over time and agreement of several common methods, 52 men alcoholics were studied. Using DSM-III diagnosis by clinical interview as the standard, the Hamilton Depression Scale (Ham-D) showed high sensitivity, high specificity and good agreement with DSM-III. The Beck Depression Inventory and the Depression Scale of the Minnesota Multiphasic Personality Inventory were not sensitive enough for screening purposes. Both self-report measures correlated more highly with the Ham-D and clinical interview than with each other. The dexamethasone suppression test had both low sensitivity and low specificity. All measures showed significant improvement over three weeks. More attention should be paid to using DSM-III criteria or the Ham-D instead of self-report scales in screening for depression among alcoholics.