精神分裂症复发征兆问卷效度分析  

Validity Analysis of the Schizophrenia Relapse Symptoms Questionnaire

作　　者：闫丽琼[1] 马良[1] 马万欣[1] 丁璐[1] 冯萍[1] YAN Liqiong MA Liang MA Wanxing et al(The Third Hospital of Chaoyang District ,Beijing 100121 ,China)

机构地区：[1]北京市朝阳区第三医院,100121

出　　处：《中国健康心理学杂志》2017年第2期171-175,共5页China Journal of Health Psychology

基　　金：北京市朝阳区科学技术委员会课题(编号:SF1221)

摘　　要：目的:验证精神分裂症复发/恶化征兆问卷(家属版)的结构效度及预测效度。方法:对304例精神分裂症患者家属进行问卷调查,所得结果应用Lisrel 8.51软件采用结构方程模型的方法对问卷的原始六因素模型进行验证性因素分析(采用最大似然估计法),得到问卷的整体拟合指标并对比。对90例精神分裂症患者家属进行随访1年,研究问卷的敏感度及特异度。结果:验证性因素分析结果显示,问卷的NFI值为0.92,NNFI值为0.94,IFI值为0.95,CFI值为0.95,RMSEA值为0.077,χ2/df值为2.77。应用ROC曲线进行分析,曲线下面积0.914(95%CI:0.856~0.971),问卷的敏感度为0.826,特异度为0.881,复发临界分为13.5分。复发患者中,问卷总分复发前后差异有统计学意义(Z=-4.287,P<0.01),各维度量表分差异均有统计学意义(Z=-2.582,-4.112,-4.017,-4.128,-4.116,-3.103;P<0.01)。未复发患者中,问卷总分及各维度分前后比较均无统计学意义。结论:精神分裂症复发/恶化征兆问卷结构效度及预测效度较好,可用于指导患者家属评估患者复发/恶化征兆。Objective：To examine the construct validity and predictive validity of schizophrenia relapse/worsening symptoms questionnaire （family member）. Methods： To investigate 304 subjects who were the family members of schizophrenia patients by using questionnaire.The data had been produced by using Lisre 18.51.Confirmatory factor analysis used the structural equation model to test the original six factors model o{ the questionnaire（using the maximum likelihood estimation method）.We conducted follow-up studies on 90 subjects who were the family members of schizophrenia patients for 1 year to study the sensitivity and specificity of the questionnaire.Results：The results of confirmatory factor analysis were as {ollows：NFI value was 0.92, NNFI value was 0.94, IFI value was 0.95,CFI value was 0.95, RMSEA value was 0.077, chi-square/df value was 2.77. The area under the ROC curve was 0. 914 （95%CI：0.856-0.971）.The sensitivity of the questionnaire was 0.826.The specificity was 0.881.Critical value of recurrence was 13.5. Among recurrence to patients, there was statistically significant difference（Z = -4. 287, P〈0.01 ） in total scores between before and after recurrence. There was significant difference （P d0.01）in all dimensions of scale scores （Z = - 2.582,- 4.112, - 4.017,- 4.128, - 4.116,- 3.103 ; P〈0.01） ;=Among no recurrence patients, there was no statistically significant difference. Conclusion： The structure validity an- predictive validity of schizophrenia re- lapse/worsening symptoms questionnaire（family member）is good.It can be used for directing the family members of patients in monitoring the relapse/worsening symptoms.

关 键 词：精神分裂症 复发 恶化 预测 评估 

分 类 号：R749.3[医药卫生—神经病学与精神病学]