阳性和阴性症状量表的中国常模和因子分析

制订阳性和阴性症状量表 (PANSS)的中国常模 ;分析其因子结构 ;为在中国精神分裂症病人中的应用提供参数。方法 :在全国 11个精神卫生机构对不同病程和亚型的精神分裂症病人进行PANSS评定。将原始评定结果进行T分和百分位数转换 ;用最大方差正交旋转法进行因子分析。结果 :常模样本为 190例正接受治疗的精神分裂症病人 ,来自全国 9个地区 ,包含了各种亚型和不同病程的急慢性精神分裂症 ,总病程平均 10 .93± 18.77月 ,平均住院 1.86次 ,目前86.8%为住院病人。各分量表得分呈正态分布。因子分析得出五因子模型 :阴性因子 ,认知因子 ,阳性因子 ,兴奋因子 ,焦虑抑郁因子。各因子与总分的相关系数在 0 .2 7～ 0 .76之间 ,与原作者相应因子的相关系数在 0 .83以上。结论 :各分量表的评定结果呈正态分布。换算后的百分位数分布与原作者相仿。五因子模型与许多报道一致。总体上 ,5因子的项目组成比较合理。

The Chinese Norm and Factor Analysis of PANSS

To establish Chinese norms and factor structure of PANSS in schizophrenic patients. Methods: PANSS was administered to 190 schizophrenic patients of different subtypes and courses of illness, sampled from 11 psychiatric institutions in China. The PANSS raw scores were transferred into T score and Percentile. Principal component analysis was used to examine the factor structure. Results: The subscale scores of PANSS were normally distributed. Factor analysis resulted in a five-factor model: negative, cognitive, positive, excitement, and anxiety/depression. Correlation between the five factors and the total score ranged from 0.27 to 0.76. Correlation with corresponding factors reported by Kay et al., was above 0.83. Conclusion: The PANSS subscale scores were normally distributed, resulting in compatible percentile values reported by the original author. The five-factor model was supported by existing research, showing good consistency and interpretability.