[Bech-Rafaelsen Mania Scale and Young Mania Rating Scale--comparison of psychometric properties of the two instruments for rating a manic syndrome].

AIM

The Bech-Rafaelsen Mania Rating Scale (BRMAS) and Young Mania Rating Scale (YMRS) are both popular rating scales for mania. Until now BRMAS has not had a Polish language version BRMAS, although it may be particularly useful because of its twin depression (melancholia) rating scale (BRMES). No attempt of evaluation of the scales were made in Poland until now. Attempts to compare their worth are rare in the literature. The aim of this study was to evaluate and compare the two scales.

METHOD

Thirty-five patients hospitalised for mania in the course of affective, schizophrenic and organic disorders were assessed by means of two mania rating scales (BRMAS, YRMS), as well as with clinical global impression (CGI) scale, and its parallel versions for mania (CGI-M) and treatment adverse events (CGI-SE). Two rating scales for psychotic syndromes (PANSS, KOSS-S) was also used. Patients were assessed three times. First three diagnosticians evaluated simultaneously and independently the patients' mental state, and then--after one and two weeks--one of them. As reliability measures, inter-rater, test-retest, and internal consistency indices were analysed. As validity measures, a clinical meaning of descriptive characteristics of the scales (external validity), and a comparison with alternative tools for mania rating (criterion, diagnostic validity) were performed. Specificity and sensitivity of BRMAS and YMRS in relation to clinical criterion of severe mania (CGI > or = 5) was analysed by means of ROC curve. As feasibility measure observations made during assessments were used.

RESULTS

BRMAS and YMRS revealed satisfactory convergence between assessments of the three diagnosticians (Spearman's rho > 0.72, no significant difference in non-parametric Friedman test), repeatability after a week (rho > 0.52) and internal consistency (Cronbach's alpha for BRMAS > 0.86, for YMRS > 0.73). Profile and dynamics of symptoms in the three successive assessments (each week) depicted changes consistent with expectations, which proved external validity of the scales. Mutual, high correlations between the BRMAS and YMRS (rho > 0.90), between them and CGI-M scale (rho >77), and between them and some meaningful items of psychosis rating scales (PANSS, KOSS-S) may be a mark of their diagnostic validity. ROC for both scales revealed similar, extensive areas under curve (AUC > 0.947; p < 0.000). BRMAS score above 24 points predicted severe mania with 92% sensitivity and 91% specificity. By analogy, YRMAS score above 29 points allowed this diagnosis with 83% sensitivity and 91% specificity. Practically no difficulties in application of the scale under study were noted.

CONCLUSIONS

Both compared mania scales revealed satisfactory indices of reliability and validity. They are also feasible tools. However limited size of the group under study has not allowed more penetrating analysis, particularly that of validity. This is why the study requires continuation.