ADAS-Cog中文版信效度分析

Reliability and Validity of Chinese Version of Alzheimer’s Disease Assessment Scale-Cognitive Part

目的:评价阿尔茨海默病评估量表认知部分(ADAS-Cog)的信度与效度。方法:采用ADAS-Cog评估135例阿尔茨海默病(AD)患者与78例健康对照老人。4周后对95名AD患者重测。评估ADAS-Cog量表的重测信度、平行效度和效标效度。结果:重测与初评结果除注意力项目不相关外,总分与其它11个项目分均显著相关(P<0.001)。ADAS-Cog总分与MMSE、ADL、NPI评分显著相关(P<0.01)。以DSM-Ⅳ为诊断AD的金标准,ADAS-Cog总分15.5分为区分AD与健康对照老人的最佳划界值,灵敏度与特异度分别为91.9%和89.5%,ROC曲线下面积(AUC)为0.95。结论:ADAS-Cog具有良好的信度与效度,适合在AD药物临床试验或纵向研究中使用。

Objective： To evaluate the validity and reliability of the Chinese version of Alzheimer＇s Disease Assessment Scale-Cognitive part （ADAS-Cog）. Methods： 135 subjects with Alzheimer＇s disease （AD） and 78 normal control subjects were assessed by ADAS-Cog. 95 subjects with AD were retested after 4 weeks. The test-retest reliability, parallel validity and criterion validity were examined. Results： Except the item of concentration, the other item scores of ADAS-Cog and its total score were significantly correlated to retest ones （P〈0.001）, ADAS-Cog total was correlated to the scores of MMSE, ADL and NPI （P〈0.01）. By the diagnostic criterion for AD from Diagnostic and Statistical Manual of Mental Disorders, the fourth version （DSM-IV）, optimal cut-off points of ADAS-Cog was 15.5, with sensitivity 91.9% and specificity 89.5%, and the area under the curve （AUC） was 0.95. Conclusion： With satisfactory reliability and validity, ADAS-Cog can be applied to clinical researches or longitudinal studies on AD.