Widlöcher's Depressive Retardation Scale and Montgomery Asberg's Depression Rating Scale: an inter-rater study

Abstract

SummaryTo study the reliability of two depression assessment scales, the Montgomery-Asberg Depression Rating Scale and the Psychomotor Retardation Scale, 30 patients with depression according to Feighner's criteria were evaluated, by two groups of raters: 5 were trained psychiatrists or psychologists, and 6 were untrained raters. The experiment lasted three months. The patients were recorded on video-tape by two psychiatrists who did not take part in the experiment, during pre-designed interviews. Each scoring was discussed by the trained group but not by the untrained group. The Kappa test and the sign-test were used for the statistical analysis of global and itemized results, respectively. Intraclass coefficient Q3 was applied to global scores. With regard to global scores, there was no significant difference among the members of the trained group, who scored higher than those of the untrained group; a significant difference between some of the untrained raters, induced a significant difference with the trained group. With regard to each item of both scales, reliability appeared to be slight to moderate, and there was no significant difference between trained and inexpert raters. These results are discussed and modifications are proposed to improve the consistency of the two scales. The authors emphasize the necessity for therapeutic trials to be conducted with permanent groups of assessors who continue training even outside of trial procedures.