摘要
目的:为研发适合中国人群的记忆力筛查量表,采用瑞金简易记忆测试(RISMET)量表对阿尔茨海默病(AD)患者进行检测,评价其效用。方法:分别对57例AD患者和46名正常对照者行RISMET量表和简易智能状态检查(MMSE)量表测评,比较2组得分情况;以内部相关系数(ICC)计算重测信度,Cronbach’sα系数评价内部一致性;以受试者工作特征(ROC)曲线评价RISMET量表对于筛查AD患者的敏感性和特异性。结果:①MMSE、RISMET量表及其分项得分(包括定向力、瞬间记忆、逆行性记忆、相似性、画钟测验、计算力、语言流畅性和回忆8项)均AD组<正常组(均P=0.000);②RISMET量表重测信度较好(ICC=0.950),内部一致性Cronbach’sα系数为0.80,各分项Cronbach’sα系数均>0.7;③Spearman相关分析显示RISMET量表与MMSE量表呈正相关(P=0.726,P=0.000),每一分项与总分的相关性均大于2个分项的相关性,具有统计学差异(P<0.01);④以RISMET量表筛查AD的ROC曲线下面积为0.959(P=0.000),当取值≤23分时,检出AD的灵敏度和特异度分别为89.5%和89.1%,Youden指数为0.786。结论:RISMET量表具有良好的信效度,进一步的大样本研究将评估其在人群中筛查AD患者的敏感性和特异性。
Objective To evaluate the reliability and validity of a novel memory rating scale developed by us, the Ruijin simplify memory test (RISMET), for screening patients with Alzheimer's disease (AD) and cognitive disorders. Methods Mini mental state examination (MMSE) and RISMET were used to test 57 patients with AD and 46 normal controls. The scores of these two tests were compared. Interelass correlation coefficient (ICC) was used to calculate the reliability of repeated testing, and Cronbach's α coefficient was calculated for evaluating the internal consistency. Moreover, 20 subjects were re-assessed after one month of the first survey, and the correlation of scores between the two surveys was tested for assessing the stability of the scale. Receiver operating characteristic (ROC) curve was used for evaluating the specificity and sensitivity of RISMET for screening AD. Results①The scores of MMSE, RISMET and its sub-items (including orientation, episodic memory, retrograde memory, similarity, clock-drawing, calculation, language fluency and recall) were significantly higher in AD patients than in normal controls (P=0.000); ② The reliability of RISMET was excellent (ICC=0.950), and the Cronbach's α coefficient was 0.80 for RISMET overall items and the Cronbach's α coefficient for all the sub-items was 〉 0.7; ③Spearman's correlation analysis showed that there was a positive correlation between RISMET and MMSE (P=0.000); correlation between score of every item and total score was higher than that between two items(P〈0.01);④ Area under ROC curve for AD screening was 0.959 (P=0.000) in RISMET,and when cut-off was set at ≤ 23, the specificity and sensitivity for screening AD were 89.5% and 89.1%, respectively, and Youden index was 0.786. Conclusions The RISMET has reasonable validity and reliability. Further study in large samples could evaluate its sensitivity and specificity for screening AD in large populations.
出处
《内科理论与实践》
2014年第2期143-147,共5页
Journal of Internal Medicine Concepts & Practice
基金
上海交通大学晨星青年学者奖励计划(2011)
上海交通大学医学院新百人计划(2010)