【作者】 钟丽娟； 高婧； 尤雪婷； 王晨光； 张芹； 江钟立； 林枫；

【Author】 Li-Juan ZHONG;Jing GAO;Xue-Ting YOU;Chen-Guang WANG;Qin ZHANG;Zhong-Li JIANG;Feng LIN;Department of Rehabilitation Medicine,Sir Run Run Hospital,Nanjing Medical University;School of Rehabilitation Medicine,Nanjing Medical University;Department of Rehabilitation Medicine,The First Affiliated Hospital of Nanjiing Medical University;

【机构】 南京医科大学附属逸夫医院,康复医学科； 南京医科大学,康复医学院； 南京医科大学第一附属医院,康复医学科；

【摘要】 目的:国际功能、残疾和健康分类(International Classification of Functioning,Disability and Health,ICF)从"生物-心理-社会"医学模式的角度为疾病和健康提供标准化术语体系,为患者的功能提供了一个标准化的理论框架。本研究拟采用项目反应理论(Item response theory, IRT)分析技术,对失语症患者(People with aphasia, PWA)的ICF核心集条目进行量化,为条目的筛选和优化提供一个全面的基础。此外,本研究将探讨患者个人能力值与条目难度之间的关系,从而指导临床应用。方法:选取75例失语症患者组成便利样本,进行横断面研究。通过与失语症患者及其家属的访谈完成了失语症ICF核心集中66个条目的评估。对条目进行非参数型和参数型项目反应理论分析,构建项目反应理论模型。此外,为了确保量表的公平性,还评估了性别之间的项目功能差异(Differentialitem functioning, DIF)。结果:结果表明优化后的量表满足单维性、单调性和局部独立性假设,且四参数模型与数据拟合最佳。Cronbach’s α值为0.8998,Guttman’sλ2值为0.9103,Molenaar-Sijtsma Statistic值为0.9248,Latent Class Reliability Coefficient(LCRC)值为0.9489,表明量表具有良好的内部一致性。由四参数模型估计的个人能力值与失语商及其子项得分之间存在中等强度(0.4<r<0.6)显著相关关系(p<0.001),表明该模型所计算的个人能力值可反映个人健康水平。DIF结果表明该量表不具有性别组差异(p>0.05)。结论:优化后的失语症ICF核心集条目均符合单调同质模型假设,表明该量表具有良好的心理测量特征。缩减后的量表仅具有23个条目,有利于提高临床评估效率。四参数模型计算的个人能力值与条目难度关系图有利于指导失语症患者制定与其能力更加匹配的康复目标,有望为患者制定康复干预计划提供合理依据。更多还原

【Abstract】 Objective:The International Classification of Functioning,Disability,and Health(ICF) provides a standardized theoretical framework for patient functioning.By applying item response theory(IRT),we quantify the ICF core set for patients with aphasia(PWAs) and provide a comprehensive basis for item screening and optimization.Additionally,the study will explore the relationship between personal ability values and item difficulty,which will guide clinical application.Methods:A cross-sectional study of 75 PWA was conducted using a convenience sampling.A speech therapist evaluated PWAs’ language ability by Western Aphasia Battery-Revised(WAB-R).All subjects and their relatives completed the 66 items in the core set.Each item’s psychological characteristics were analyzed using Mokken scale analysis and parametric item response theory technique.We also evaluated differential item functioning(DIF) based on gender differences to ensure scale fairness.Results:A 4-PL IRT analysis provided the best fit for the data,and the assumptions of unidimensionality,monotonicity,and local independence were all satisfied.The Cronbach’s α was 0.8998,Guttman’s λ2 was 0.9103,Molenaar-Sijtsma Statistic was 0.9248,and the Latent Class Reliability Coefficient(LCRC) was 0.9489,suggesting a good internal consistency.The 4PLM was to estimate personal ability values.It showed that personal ability values were moderately correlated with each WAB-R subset score and Aphasia Quotient(AQ)(p < 0.001,0.4 < r < 0.65).DIF analysis suggested no significant gender differences(p > 0.05).Conclusion:The IRT assumptions were met for the shortened PWA ICF core set,indicating that the scale has good psychometric properties.The reduction of items to only 23 items is useful for matching rehabilitation goals with person-item threshold distribution maps.更多还原