Abstract: Background　Alzheimer's disease （AD），as a chronic neurodegenerative disease，cannot be cured.Thus，it is especially important to achieve “early detection，early diagnosis and early treatment”.Subjective cognitive decline（SCD）has attracted attention in the field of AD as it can indicate the preclinical stage of AD.The development of simple screening tools can help general practitioners to identify possible preclinical stage of AD，meanwhile reduce the burden on patients and doctors further.Objective　To explore the early simple screening tools for AD by translating English-version SCD questionnaire 9 （SCD-Q9）．Methods　The English version of SCD-Q9 was translated into Chinese via “two-person literal translation，back-translation” method of Brislin after obtaining the original author's authorization.Then the original author compared the nuances in language and culture between the Chinese and the original version.The first draft of SCD-Q9 questionnaire was revised and culturally adjusted after the expert panel discussion and pre-survey.From September 6 to 15，2016，two hundred local residents in ShunYi District were selected by random sampling method.The questionnaire which included general situation and Chinese version SCD-Q9 were used for the investigation.A total of 200 questionnaires were distributed and all of them were effectively recovered，with an effective recovery rate of 100.0%.Project analysis was carried out by using the comparative method between extreme groups （the top 27% of high group and the bottom 27% of low group）of project differentiation degree and the correlation coefficient method of project homogeneity.Cronbach's α coefficient and Spearman-Brown coefficient method were used for reliability analysis.And content validity was analyzed by expert evaluation.KMO test，Bartlett's sphericity test and exploratory factor analysis （EFA）were used for structural validity analysis.Results　The Chinese version of SCD-Q9 contained 2 dimensions and 9 items，including the overall memory function and time comparison （4 items）and daily activity ability （5 items）．The subjects were divided into high score group （54 cases）and low score group （54 cases）by item differentiation（top 27% as the high group and bottom 27% as the low group）．The differences between the two groups were statistically significant in each item （P<0.05）．The correlation analysis of each item and the total score showed that there was a negative correlation between them （P<0.05），the absolute value of the correlation coefficient was >0.400，and all items could be retained.Results of reliability test were showed，the Cronbach's α coefficient of the Chinese version of SCD-Q9 was 0.886.After deleting any item of the questionnaire，the Cronbach's α coefficient ranged from 0.870 to 0.881.With no specific high value，there was no prompt to delete entries.Spearman-Brown's split-half coefficient was 0.892.We also found that the KMO value of the questionnaire was 0.883 and the Bartlett's spherical test χ2 was 1 059.320，df=36，P<0.001，which indicated that the questionnaire was suitable for further factor analysis.Two common factors were extracted by EFA with eigenvalue≥1，and the eigenvalue of common factor 1 was 5.116，whose variance contribution rate was 56.846%，the eigenvalue of factor 2 was 1.231，whose variance contribution value was 13.673%，and the cumulative variance contribution value of two common factors was 70.519%.Conclusion　The internal validity and reliability of the Chinese version of SCD-Q9 questionnaire is good.The development of this questionnaire lays the foundation for its subsequent application in the Chinese population，and provides a simple screening tool for the exploration of SCD.

Key words: Alzheimer's disease, Subjective cognitive decline, Questionnaires, Screening, Reliability, Validity, SCD questionnaire 9

摘要： 背景　阿尔茨海默病（AD）作为一种慢性神经退行性疾病，无法治愈，实现早发现、早诊断、早治疗尤为重要。主观认知下降（SCD）因对AD临床前期具有重要的提示作用而备受关注，其简易筛查工具的研发可以帮助社区医生早期识别可能的AD临床前期患者，同时进一步减轻患者及医生负担。目的　通过汉化英文版主观认知下降问卷9（SCD-Q9），探索AD早期简易筛查工具。方法　取得英文版SCD-Q9原作者的问卷授权，采用Brislin“两人直译、回译”翻译法汉化SCD-Q9，原作者对比回译中文版与原版语言和文化的细微差别。并通过专家小组讨论及预调查对中文版SCD-Q9初稿进行条目修订及文化调适，形成中文版SCD-Q9。2016年9月6—15日，采用随机抽样法，选取200例顺义区当地居民为调查对象。采用一般情况调查问卷、中文版SCD-Q9对其进行调查。共发放问卷200份，回收有效问卷200份，有效回收率为100.0%。采用项目区分度（前27%高分组和后27%低分组）极端组比较法及项目同质性检验相关系数法进行项目分析；采用Cronbach's α系数、分半信度Spearman-Brown系数法进行信度分析；采用专家评价法进行内容效度分析；采用KMO检验、Bartlett's球形检验、探索性因子分析（EFA）进行结构效度分析。结果　中文版SCD-Q9包含2个维度、9个条目，分别是整体的记忆功能及时间对比（4个条目）、日常活动能力（5个条目）。采用项目区分度（前27%高分组和后27%低分组）极端组比较法将被调查者分为高分组（54例）、低分组（54例），各条目两组间比较，差异均有统计学意义（P<0.05）。各条目与总分的相关性分析结果显示，中文版SCD-Q9各条目和总分之间均呈负相关（P<0.05），相关系数绝对值均>0.400，各条目均可保留。信度检验结果显示Cronbach's α系数为0.886，在删除问卷中的任一条目后，问卷的Cronbach's α系数为0.870~0.881，无特异高值，无提示需要删除的条目。分半信度结果显示，Spearman-Brown分半系数为0.892。中文版SCD-Q9 KMO值为0.883，Bartlett's球形检验χ2=1 059.320，df=36，P<0.001，表明该问卷适合进一步做因子分析。利用EFA对问卷提取公因子，特征值≥1.000的公因子有2个，公因子1的特征值为5.116，方差贡献率为56.846%，因子2的特征值为1.231，方差贡献值为13.673%，2个公因子对总方差的累积方差贡献率为70.519%。结论　中文版SCD-Q9内部效度及信度良好，该问卷的研发，为后续中文版SCD-Q9在中国人群中的运用奠定基础，同时为SCD早期简易筛查工具的探索提供参考。

关键词: 阿尔茨海默病, 认知下降, 问卷调查, 筛查, 信度, 效度, 主观认知下降问卷9