Interest in self-reported measures of depression in clinical trials has grown in recent years. This study compared the reliability and validity of the clinician-administered Montgomery-Asberg Depression Rating Scale (MADRS) to a computer-administered version administered over the telephone using Interactive Voice Response (IVR) technology. Sixty subjects were administered both the clinician- and computer-administered versions of the MADRS in a counter-balanced order. A subsample of 20 patients was reassessed 24h later by both methods. Mean score differences between IVR and clinician were not statistically significant (<1 point) and a high correlation was found between forms (r=.815, p<.001). Reliability measures (Cronbach's Alpha and 24-h test-retest) were comparable. Clinicians rated the severity of subjects' sadness and pessimistic thoughts lower than subjects self-report. The data obtained in this pilot study provide support for the equivalence between the clinician and IVR versions of the MADRS.