中文版改良诺丁汉感觉功能评价量表的建立及信效度检验

关键词：脑梗死  感觉功能  量表  信度  效度  跨文化调试

Keywords:stroke  sensory ability  scale  validity  reliability  cross-cultural adaption

摘要：

      摘要 目的：对改良版诺丁汉感觉功能评价量表(revised Nottingham sensory assessment scale，reNSA)进行跨文化调试与信效度检验，建立本土化的中文版reNSA量表。 方法：依据美国矫形外科医师学会循证医学委员会(American Academy of Orthopaedic Surgeons Committee on Evidence-based Medicine， AAOS)推荐的跨文化调适过程指南，按照“翻译-整合-回译-小组评判-专家咨询-预测试-形成初步工具”的程序翻译并调试reNSA量表。对50例梗死后康复期患者进行量表测评，通过临界比率值(critical ratio value， CR)和同质性检验进行条目分析，通过Cronbach α系数、分半信度和重测信度进行信度分析；通过条目内容效度指数(item content validity index， I-CVI)、量表内容效度指数(scale content validity index， S-CVI)和结构效度进行效度分析。 结果：本次测评共回收问卷47份，回收率94.0%。量表纳入8个条目，CR值均>3.0，各个条目与总分的Pearson系数均>0.5。8条条目的整体Cronbach α系数为0.908，半分信度为0.949，重测信度为0.479—0.968。I-CVI和S-CVI分别为0.953和0.938；主成分分析法提取公因子3个，累积解释方差94.662%。此外，reNSA量表得分与Fugl-Meyer感觉评价量表（Fugl-Meyer sensory assessment scale）得分呈直线正相关性(r=0.729，P<0.001)。 结论：本研究采用跨文化调试的方法建立中文版reNSA量表，信效度高，可为脑梗死的预后评估提供方法学依据。

Abstract:

      Abstract Objective： To establish a Chinese version revised Nottingham sensory assessment (reNSA) scale through cross-cultural adaption methods， and to test its validity and reliability. Method： According to the guidelines for cross-cultural adaptation recommended by the American academy of orthopaedic surgeons committee on evidence-based medicine (AAOS)， a Chinese version reNSA scale was established through the program of "translation-integration-back-translation-group assessment-expert consultation-predictive test-forming preliminary tool". Fifty patients at the stage of post-infarction rehabilitation were rescued to evaluate the Chinese version reNSA scale. The critical ratio value (CR) and homogeneity test were used for item analysis. The reliability was analyzed by Cronbach coefficient， fractional and semi-reliability and retest reliability. The validity analysis was carried out through item content validity index (I-CVI)， scale content validity index (S-CVI) and structure validity. Result： A total of 47 questionnaires were collected， with a recovery rate of 94.0%. Eight items were included in the scale， and their CR values were all >3.0， and the Pearson conffcient of each item with the total score was >0.5. The overall Cronbach coefficient of items was 0.908， the half-fraction reliability was 0.949， and the retaken reliability was 0.479—0.968. I-CVI and S-CVI were 0.953 and 0.938， respectively. Three common factors were extracted by principal component analysis， and the cumulative interpretation variance was 94.662%. In addition， the score of reNSA scale was positively correlated with the score of Fugl-Meyer sensory assessment scale in a straight line (r=0.729， P<0.001). Conclusion： A Chinese version reNSA scale was established by cross-cultural adaption with high reliability and validity， which can provide methodological basis for the prognosis assessment of cerebral infarction.