Assessment of treatment outcome in depressed patients: concordance of methods

Authors who compared the use of various depression scales for monitoring the effects of treatment have reported that the observed pattern of improvement depends on the scale used. The discrepancies have generally been attributed to interscale differences in content and administration mode. We investigated this phenomenon in a group of 52 out-patients with non-psychotic, non-melancholic DSM-III major depression disorders using three widely employed depression scales in both their usual formats and in formats for the alternative mode of administration. The six instruments were applied at admission and after 12 and 24 weeks' treatment. The results suggest that differences between the contents of different scales lead to differences in the global improvement reflected, but provide no evidence of significant discrepancy between the patterns of interviewer-rated and self-rated progress.