Department of Psychiatry, Instituto de Investigación Hospital 12 de Octubre (i+12), Avda. de Córdoba s/n, 28041, Madrid, Spain.
Schizophr Res. 2013 Jan;143(1):77-83. doi: 10.1016/j.schres.2012.10.020. Epub 2012 Nov 30.
Different exploratory and confirmatory factorial analyses of the Positive and Negative Syndrome Scale (PANSS) have found a number of factors other than the original positive, negative, and general psychopathology. Based on a review of previous studies and using confirmatory factor analyses (CFA), Wallwork et al. (Schizophr Res 2012; 137: 246-250) have recently proposed a consensus five-factor structure of the PANSS. This solution includes a cognitive factor which could be a useful measure of cognition in schizophrenia. Our objectives were 1) to study the psychometric properties (factorial structure and reliability) of this consensus five-factor model of the PANSS, and 2) to study the relationship between executive performance assessed using the Wisconsin Card Sorting Test (WCST) and the proposed PANSS consensus cognitive factor (composed by items P2-N5-G11). This cross-sectional study included a final sample of 201 Spanish outpatients diagnosed with schizophrenia. For our first objective, CFA was performed and Cronbach's alphas of the five factors were calculated; for the second objective, sequential linear regression analyses were used. The results of the CFA showed acceptable fit indices (NNFI=0.94, CFI=0.95, RMSEA=0.08). Cronbach's alphas of the five factors were adequate. Regression analyses showed that this five-factor model of the PANSS explained more of the WCST variance than the classical three-factor model. Moreover, higher cognitive factor scores were associated with worse WCST performance. These results supporting its factorial structure and reliability provide robustness to this consensus PANSS five-factor model, and indicate some usefulness of the cognitive factor in the clinical assessment of schizophrenic patients.
不同的探索性和验证性因子分析表明,阳性与阴性综合征量表(PANSS)除了最初的阳性、阴性和一般精神病学之外,还有许多其他因素。基于对以往研究的回顾,并使用验证性因子分析(CFA),Wallwork 等人(Schizophr Res 2012; 137: 246-250)最近提出了 PANSS 的共识五因素结构。该解决方案包括认知因素,这可能是精神分裂症认知的有用衡量标准。我们的目标是 1)研究该共识五因素模型的心理测量特性(因子结构和可靠性),以及 2)研究威斯康星卡片分类测试(WCST)评估的执行性能与提出的 PANSS 共识认知因素(由项目 P2-N5-G11 组成)之间的关系。这项横断面研究包括 201 名西班牙门诊精神分裂症患者的最终样本。对于我们的第一个目标,进行了 CFA,并计算了五个因素的克朗巴赫阿尔法;对于第二个目标,使用顺序线性回归分析。CFA 的结果显示出可接受的拟合指数(NNFI=0.94,CFI=0.95,RMSEA=0.08)。五个因素的克朗巴赫阿尔法是足够的。回归分析表明,PANSS 的这个五因素模型比经典的三因素模型解释了更多的 WCST 方差。此外,认知因子得分越高,WCST 表现越差。这些结果支持其因子结构和可靠性,为该共识 PANSS 五因素模型提供了稳健性,并表明认知因子在精神分裂症患者的临床评估中有一定的用处。