Rating depression in normals and depressives: observer versus self-rating scales.

Different methods of assessing depression and anxiety were tested in 20 patients suffering from a major depressive disorder with melancholia and 20 matched control subjects. Depressives were assessed before and after treatment with amitriptyline and normals were retested at the same interval. The scales used were: Paykel's Clinical Interview for Depression--which is an expanded version of the Hamilton Depression Rating Scale; the Brief Depression Rating Scale; and Symptom Questionnaire (SQ). All scales discriminated sensitively between patients and normals and the scores changed substantially with treatment. Except for the well-being subscales of the SQ, the scales showed an adequate test-retest reliability in normals. Although all scales were suitable for the measurement of depression, they differed in psychometric properties. For example, the Depression subscale of the SQ showed an unusually high test-retest reliability in normals, whereas the Contentment subscale was unreliable. Yet, the latter has been found to be highly sensitive in detecting differences between the effects of psychotropic drugs and placebo in drug trials, so it appears to measure sensitively a fleeting mood. The combined use of all three scales in patients with affective disorders yields information that might not be revealed if only one scale is used