Validation of a severity threshold for the mania rating scale:: A receiver-operating characteristic analysis

Background: The aim of this study was to determine a cutoff score on the Mania Rating Scale ( MRS), which easily allows identification of severe mania in a population of manic patients to be included in clinical trials of antimanic drugs. Method: 1,090 hospitalized patients meeting DSM-IV criteria for a manic episode were subtyped according to the specifier for severity and assessed for demographic characteristics, illness course and clinical symptomatology. Using receiver-operating characteristic ( ROC) analysis, the optimal threshold for severity was determined on the MRS. Results: In a French national clinical sample (n = 1,090), 851 cases were specified as severe and 239 as nonsevere ( mild + moderate) mania according to DSM-IV criteria ( 307 without psychotic features, 544 with psychotic features). Patients with severe mania scored higher on the MRS but showed the same levels of scores on the Montgomery Asberg Depression Rating Scale compared to nonsevere cases. Many characteristics of the whole sample and of the psychotic group were found to be comparable, respectively, to those reported in recent epidemiological studies, which was particularly true for age, gender, age at onset, number of mood episodes and suicide attempts. The optimal ROC solution for separating severe from nonsevere mania was a cutoff score of 39 on the MRS. This cutoff score displayed a positive predictive value of 0.91. Conclusion: A cutoff score of 39 is proposed as a severity threshold for mania on the MRS by virtue of its ROC validation in a large representative sample of severe versus nonsevere manic patients whose severity was assessed according to DSM-IV subtyping. Copyright (c) 2007 S. Karger AG, Basel.