指南临床适用性评价工具(2.0版)信效度分析

Reliability and validity analysis of the instrument for evaluating clinical applicability of guidelines(version 2.0)

目的分析指南临床适用性评价工具(2.0版)的信度和效度。方法分别从东、中、西部地区抽取已开展手癣、足癣和脑出血相关诊治工作的各级医疗机构专家,采用指南临床适用性评价工具(2.0版)评价《手癣和足癣诊疗指南(2017修订版)》和《中国脑出血诊治指南(2019)》。采用Cronbach’sα系数和Spearman-Brown系数评价该工具的内在信度和分半信度,通过计算条目水平内容效度指数及调整后的Kappa值评价该工具的内容效度,通过计算各条目与所在维度的相关系数及假设检验评价集合效度和区分效度,采用结构方程模型进行验证性因子分析评价工具的结构效度。结果指南临床适用性评价工具(2.0版)评价两部指南的Cronbach’sα系数和Spearman-Brown系数均大于0.7,内容效度指数S-CVI/Ave均大于0.8,集合效度均为100%,区分效度试验成功率分别为100%和96%,结构方程模型进行验证性因子的χ^(2)/df均小于3,模拟比较拟合指数CFI、拟合优度指数GFI、增值拟合指数IFI均大于0.9,均方根残差RMR均小于0.05,近似误差均方根RMSEA均小于0.09,RESEA假设检验的P值均大于0.05。结论指南临床适用性评价工具(2.0版)具有良好的信度和效度,未来可实际应用进一步验证其信度和效度。

Objective To evaluate the reliability and validity of the instrument of clinical applicability of guidelines(version 2.0).Methods The experts of domestic medical institutions were investigated by questionnaire,and the instrument of clinical applicability of guidelines(version 2.0)were evaluated the guidelines for the diagnosis and treatment of tinea mantis and tinea pedis(revised edition 2017)and the guidelines for the diagnosis and treatment of cerebral hemorrhage in China(2019).Using Cronbach’sαcoefficient and Spearman-Brown coefficient to evaluate the inherent reliability and split-half reliability.The content validity was evaluated by calculating the content validity index of the item level and the adjusted Kappa value.The correlation coefficient between each item and the dimension and the hypothesis test were used to evaluate the convergent and discriminant validity.The structural validity was evaluated by using structural equation model to evaluate the structural validity of the tool.Results The Cronbach’sαcoefficient and Spearman-Brown coefficient of the instrument of clinical applicability of guidelines(version 2.0)were both greater than0.7,the content validity index(S-CVI/Ave)were more than 0.8,the success rates of convergent were 100%,and the success rates of discriminant validity calibration were 100%and 96%.In the second-order confirmatory factor analysis model,theχ^(2)/df were less than 3,the fitting index(CFI),the goodness of fit index(GFI)and the adjustment goodness of fit index(AGFI)were all greater than 0.9.The root mean square residual(RMR)were all less than 0.05,and approximate error root mean square(RMSEA)were less than 0.09.The P value of RESEA hypothesis test were more than 0.05.Conclusion The instrument of clinical applicability of guidelines(version 2.0)has good reliability and validity,which can be further verified in practical application in the future.