社区老年人健康状况综合评估量表的信效度评价

Reliability and validity of Comprehensive Geriatric Assessment Scale for community-dwelling elderly

目的评价社区老年人健康状况综合评估量表(CGA)在北京地区应用的信度、效度与可接受度。方法2020年8月至2021年1月,采用方便抽样法选取北京市3个区共1983名老年人进行CGA量表评价,有效回收问卷1980份。采用克朗巴赫α(Cronbach′sα)系数评价量表的内部一致性;采用组内相关系数(ICC)评价量表的重测信度及评定者信度;Pearson相关系数对量表进行项目分析,并评价量表的校标效度;探索性因子分析和验证性因子分析评价量表的结构效度;临界比值法评价量表的敏感性;量表有效回收率及填表时间评估量表的可接受性。采用SPSS 22.0及AMOS 22.0软件对数据进行统计分析。结果总量表的Cronbach′sα系数,重测信度及评定者信度的ICC分别为0.849、0.837及0.899,各维度Cronbach′sα系数,重测信度及评定者信度的ICC分别为0.467~0.833、0.579~0.910及0.280~0.860。项目分析结果显示,各条目得分与其所在维度得分的Pearson相关系数r为0.241~0.934(均P<0.01);与量表总分的相关性系数r为0.018~0.631,除条目5-5外,其余差异均有统计学意义(均P<0.01)。反映结构效度的Kaiser-Meyer-Olkin的取样适切性量数值为0.737,Bartlett球形检验的χ^(2)值为14420.98(P<0.001),探索性因子分析显示量表共提取13个公因子,累计方差贡献率为55.65%。验证性因子分析显示量表结构方程模型的拟合优度指数、增值拟合指数、比较拟合指数、规准拟合指数及简约拟合优度指数分别为0.961、0.927、0.926、0.903及0.753,近似误差均方根为0.039,均符合拟合值标准,模型结构拟合较好。除慢性病患病状况及社会支持外,其余各维度与其校标量表均呈显著正相关(均P<0.05)。各维度临界比值为8.09~28.10(均P<0.01),提示量表各维度均有较好的敏感性。量表填写时间为(12.8±4.8)min,有效回收率为99.0%(1980/2000)。结论本量表具有良好的内部一致性、重测信度及评定者信度;内容效度、结构效度及校标效度均通过验证,可全面反映社区老年人的健康综合状况;可接受度良好。

Objective To evaluate the reliability,validity and acceptability of the Community Comprehensive Health Assessment(CGA)scale for the elderly people in Beijing.Methods From August 2020 to January 2021,1983 elderly people living in 3 districts of Beijing were selected by convenient sampling and surveyed with CGA scale.And finally,1980 questionnaires were validly recovered.Kronbachα(Cronbach′sα)coefficient was used to evaluate the internal consistency of the scale.Intragroup correlation coefficient(ICC)was employed to analyze the test-retest reliability and evaluator reliability of the scale.Pearson correlation coefficient was used for item analysis of the scale.Exploratory factor analysis and confirmatory factor analysis were conducted to evaluate the structural validity of the scale.Pearson correlation coefficient was employed to evaluate the calibration validity of the scale.Critical ratio was used to evaluate the sensitivity of the scale.The acceptability of the scale was evaluated by the effective recovery rate and filling time of the scale.SPSS 22.0 and AMOS 22.0 software were used for statistical analysis.Results The Cronbach′sαcoefficient of the total scale,and ICC coefficients of retest reliability and evaluator reliability were 0.849,0.837 and 0.899,respectively,and the Cronbach′sαcoefficient of all dimensions,and ICC coefficients of retest reliability and evaluator reliability ranged from 0.467 to 0.833,from 0.579 to 0.910,and from 0.280 to 0.860,respectively.The results of item analysis showed that the Pearson correlation coefficient r between the score of each item and that of its corresponding dimension was 0.241-0.934(all P<0.01),and the correlation coefficient r with the total score of the scale was 0.018-0.631.Except for items 5-5,the differences were statistically significant(all P<0.01).The sampling suitability of Kaiser Meyer Olki,which reflects the structural validity,was 0.737,and the Chi-square value of Bartlett′s test was 14420.98(P<0.001).Exploratory factor analysis showed that there were 13 common factors extracted from the scale,and the cumulative variance contribution rate was 55.65%.The confirmatory factor analysis indicated that the goodness of fit index,incremental fit index,comparative fit index,normed fit index and parsimony goodness of fit index were 0.961,0.927,0.926,0.903 and 0.753,respectively,and the root mean square error of approximation was 0.039,which were in line with the standard of fitting values,and the model structure was well fitted.Except for the prevalence of chronic diseases and social support,the other dimensions were significantly positively correlated with the calibration scale(all P<0.05).The critical ratio of each dimension was 8.09-28.10(all P<0.01),indicating that each dimension of the scale had good sensitivity.The filling time of the scale was(12.8±4.8)min,and the effective recovery rate was 99.1%(1980/2000).Conclusion This CGA scale is internal consistent,retest and inter-rater reliable,with good content validity,structural validity,calibration validity,discriminant validity and excellent acceptability,and thus can be used to evaluate the comprehensive health status of the elderly in the community.