Frenchay活动量表的预测效度和信度研究

Predictive Validity and Reliability of Frenchay Activities Index

目的:评价中文版Frenchay活动量表(FAI)在脑卒中住院恢复期患者的信度与效度,为临床该量表的使用提供依据。方法:遵循国际标准化的慢性病治疗功能评价(FACIT)原则对原英文FAI进行量表翻译和修订,由4名康复治疗学教授、1名康复医学教授和1名神经康复心理学教授组成专家评审小组进行内容效度评审,形成中文版FAI量表,选取复旦大学附属华山医院、暨南大学附属第一医院、广东省中医院、新疆维吾尔自治区中医院和福建中医药大学附属康复医院5家医院共403名40～80岁的脑卒中住院恢复期患者,由同一名评定员以面对面访谈的方式对患者进行FAI和改良Barthel指数(MBI)的评定。以检验量表的内部一致性、结构效度、项目区分度,及评估MBI进行校标关联效度检验,1周之内随机抽取94例受试者,由同一名评定员进行重测检验。结果:中文版FAI量表的内部一致性Cronbach'sα系数为0.937。重测信度显示组内相关系数(ICC)为0.810(95%CI:0.672～0.952)。项目区分度方面,各项目高分组和低分组得分比较,差异均有统计学意义(P<0.01),该量表各项目区分度总体良好;中文版FAI量表各条目、各维度与总分的相关分析发现,中文版FAI中各项目与总分的相关系数在0.535～0.832(P<0.01)范围内,家居活动、休闲活动、外出活动3个维度与总分的相关系数分别是0.867、0.913、0.948(P<0.01)。采用探索性因子分析(EFA)检验中文版FAI量表的结构效度KMO统计量为0.924,Bartlett球形检验χ~2值为4535.558(df=105,P<0.01)。利用主成分分析法对量表提取公因子,结果特征值≥1的公因子有3个,公因子1的特征值为4.776,方差贡献率为31.842%,公因子2的特征值为3.437,方差贡献率为22.914%,公因子3的特征值为2.436,方差贡献率为16.238%,3个公因子对总方差的累积贡献率为70.994%。效标关联效度检验显示,FAI总分与MBI总分呈正相关,相关性一般(r=0.322,P<0.01),FAI的3个维度(家居活动、休闲活动、外出活动)与MBI总分的相关性一般(r=0.320、r=0.267、r=0.319,P<0.01)。结论:中文版Frenchay活动量表在脑卒中住院恢复期的患者中具有较好的信度和效度。

Objective: To evaluate the reliability and validity of the Chinese version of the Frenchay activities index(FAI) in the recovery period of hospitalized stroke, and to provide evidence for the clinical use of the scale. Methods: The original English FAI was translated and revised according to the international standardized functional assessment of chronic illness therapy(FACIT) principle.An expert review group were formed with four rehabilitation professors, one professor of rehabilitation medicine and one professor of neurorehabilitation psychology, to conduct content validity review for Chinese FAI scale. A total of 403 hospitalized patients from 40 to 80 years old with recovery in five affiliated rehabilitation hospitals, from Huashan Hospital Affiliated to Fudan University, First Affiliated Hospital of Jinan University, Guangdong Provincial Hospital of Traditional Chinese Medicine, Xinjiang Uygur Autonomous Region Chinese Medicine Hospital and Fujian University of Traditional Chinese Medicine, they were assessed by FIA and modified Barthel index(MBI) by face-to-face interviews with the same assessor. The internal consistency, structural validity, project discrimination of the test scale, and the evaluation of the improved Barthel index were used to test the validity of the calibration. Ninety-four subjects were randomly selected within one week and retested by the same assessor. Results: The internal consistency of the Chinese version of the FAI scale was Cronbach’s α coefficient of 0.937. The test-retest reliability showed that the intra-group correlation coefficient(ICC) was 0.810(95%CI: 0.672 to 0.952). In terms of project discrimination, the differences between the high and low group scores of each project were statistically significant(P<0.01), and the discrimination of each item in the scale was generally good;the correlation analysis of the entries, dimensions and total scores of the Chinese version of the FAI scale found that the correlation coefficient between the items in the Chinese version of FAI and the total score was in the range of 0.535-0.832(P<0.01), and the correlation coefficient of the three dimensions and total scores of household activities, leisure activities and outing activities were 0.867,0.913, 0.948 respectively(P<0.01). The exploratory factor analysis(EFA) was used to test the structural validity of the Chinese version of the FAI scale to a KMO statistic of 0.924 and the Bartlett spherical χ~2 test to 4535.558(df=105, P<0.01). Using the principal component analysis method to extract the common factors from the scale, there were three common factors with eigenvalue ≥1, the eigenvalue of common factor 1 was 4.776, the contribution rate of variance was 31.842%, and the eigenvalue of common factor 2 was3.437. The contribution rate of variance was 22.914%, the eigenvalue of common factor 3 was 2.436, the contribution rate of variance was 16.238%, and the cumulative contribution of the three common factors to the total variance was 70.994%. The validity-related validity test showed that the FAI total score was positively correlated with the MBI total score, the correlation was general(r=0.322, P<0.01), and the three dimensions of FAI(home activities, leisure activities, outing activities) was correlated with the MBI total score, the correlation was general(r=0.320, r=0.267, r=0.319, P<0.01). Conclusion: The Chinese version of the Frenchay activities index has good reliability and validity in patients with recovery from stroke hospitalization.