The feasibility of the use of multiple informant reports in clinical practice was examined in a sample of 105 psychiatric outpatients who provided self-ratings and (2-3) informants' reports on the Five Factor Personality Inventory. The response rate was 97%. The patients assessed themselves as less extraverted and more emotionally stable than their proxies did. In addition, the significance of self-other disagreement was investigated. Our first hypothesis, stating that self-other disagreement would correlate with (personality) pathology, was confirmed: self-other disagreement predominantly occurred in introverted, shy, hostile and depressed persons who tended to have more personality problems and co-morbidity. We found no support for our second hypothesis, stating that self-other disagreement would predict a diminished therapy effect. An important finding, however, was that self-other disagreement proved to be a strong predictor of dropout. Furthermore, a decrease in depression, hostility and shyness was positively correlated with a decrease in self-other disagreement.