Authors
8 of 9Flávio ShansisMarcelo Pio de Almeida FleckIván IzquierdoAlexandre Willi SchwartzhauptEduardo ChachamovichEugenio Horácio GrevetBetina S. MatteviGabriela Maldonado
Topics
Bipolar Disorder and Treatment..•._----P187Which is lhe besl mania raling scale? A comparalive study of psychomelric propertie. of Young, CARS·M and Bech-Rafael.en Scale.F Shansis·.~' Izquierdo, A Schwartzhalpt. EChachamovich. E Grevet. B Mattevi. GMaldonado, M Ber1im Hospital de Clínicas de Porto Alegre - Federal University Df Rio Grande do Sul, Oepartments of Psychjatry and Biochemislry. Porto Alegre, BraziJ Int..-oduction: Clinicai and phannacological studies with manie paLienls requires lhe availabilily af a good rating scale lo assess both quantilative and qualitative changes in mood, cognitive and behaviour aspecls ar (his syndrome. Objectivt'S: The authors have compared eilher psychometric as quaJilalive aspecls of lhe lhree most used mania raling scales: Young (YM RS), Altman Rating Scale Clinician Version' (CARS-M) .nd Bech-Rafaelsen (BR). Mefhods: Fifteen bipolar type I inpalients in manie episode were videolaped and raled by three independent psychiatrislS resulting in 45 observalions. The sample was helerogenous regarding sex, age and clinicai aspects. A control group had 15 paoic eulhymic palients. Resutts: A high intraclass correlation coefficient was observed: YMRS = 0.86 (0.71--6.94), CARS-M = 0.86 (0.71--6.94) and BR = 0.84 (O.68--ú.94) (p < 0.00 I). A principal componenl analyses 3nd Cronbach a coefficienl were perfonned to asses validily. 11 was built in a matrix with atolaI variance explained for YMRS, CAR·M and BR af 61.03%, 72.63% and 71.83% of eigenvalues in Kaiser method > I resulting in lhree. five and four faclors. respectively. The Cronbach a coefficienl was 0.66. 0.80 and 0.79. As expecloo lhe discrimina0I validily showed a slatislical difference in rating manic symploms (p < 0.001). Discussion: Although Ihese re..<;ulls have showed lhal ali the Ihree scales have very similar psychometric properties, the Qualitative analyses regarding clinicaI evidence has showed Ihat each ooe is more specific to some manie presentations. 11 is noleworlhy that psychOlic symptoms (exceplion for grandiose deJirious) are nor well rdled in these scales. specifically paranoid symptoms. These lhree mania raling scales are retiable instrumenls for euphoric and non·psychotic mania bul not for other presenlalions. Conclusions: Our study suggest lhat lhese scales are more able lo rate 'ç1assical (euphoric non-psychotic)' (ype I bipolar patienls. KeYlI'ord.s: bipolar disorders. mania rating scales, psychometric properllcs.
About
PublishedJan 1, 2003
Citations0