简体版改良Barthel指数在脑卒中恢复期中的信度与效度研究

目的:观察简体版改良Barthel指数量表(MBI)在评定脑卒中患者恢复期基础性日常生活活动(BADL)的信度和效度。方法:由2名研究员对100名符合入选标准的脑卒中恢复期患者进行BADL评定。在患者入院的次日,上午由评定员A、B使用MBI分别对每位患者的BADL进行评定,下午由评定员A使用MBI、评定员B使用BI分别对每位患者的BADL进行再次评定,收集整理数据并加以分析。采用相关系数(ICC)值描述评价者的重测信度、评价者间信度,采用Cronbach’sα系数进行计算量表的内部一致性;邀请6名高级职称康复专家对简体版MBI的内容效度(CVI)进行评估,采用Spearman相关性检验分析简体版MBI的校标效度(以BI为金标准),采用KMO检验、Bartlett球形检验和因子分析方法来分析构想效度。结果:6名专家对简体版MBI的条目水平CVI为0.83~1,量表水平CVI均值为0.95;简体版MBI各个条目、总分的重测信度为ICC=0.909~0.991;简体版MBI各个条目、总分的评定者间信度为ICC=0.880~0.992;简体版MBI内部一致性Cronbach’α系数为0.929;以BI为金标准,简体版MBI的校标效度(Spearman相关系数)依次为:进食(0.562)、个人卫生(0.701)、洗澡(0.715)、如厕(0.785)、穿衣(0.684)、大便控制(0.690)、小便控制(0.630)、床椅转移(0.925)、步行/轮椅(0.864)、上下楼梯(0.877)、总分(0.941),均P<0.001;简体版MBI构想效度结果显示,KMO统计量为0.877,Bartlett球形检验拒绝相关阵是单位阵(P<0.05),80.28%的总方差可由2个公因子解释,旋转后因子1有:洗澡、穿衣、如厕、床椅转移、步行/轮椅、上下楼;旋转后因子2有:进食、个人卫生、大便控制、小便控制。结论:简体版MBI量表不仅具有良好内容效度、重测信度、评定者间信度、内部一致性、校标效度和构想效度,而且易于掌握、使用方便。

Reliability and validity of the simplified version Modified Barthel Index in convalescence period of stroke

Objective:To evaluate the reliability and validity of the simplified version Modified Barthel Index(MBI)for basic activity of daily living(BADL)in convalescence period of stroke.Methods:Two researchers evaluated BADL of 100 patients with stroke in recovery phase.On the day after admission,BADL was assessed by raters A and B using MBI in the morning,and again by raters A using MBI and B using BI in the afternoon,and the data were analyzed.Intraclass correlation Coefficient(ICC)was used to describe the retest reliability and interevaluator Reliability.Cronbach’s α coefficient was used to calculate the internal consistency of the scale.Content validity index(CVI)was used to evaluate the content validity.Spearman correlation test was used to analyze the calibration validity of simplified MBI(BI as the gold standard).KMO test,Bartlett Spherical test and factor analysis were used to analyze construct-related validity of simplified MBI.Results:The CVI of six experts was 0.83 to 1,and the average value of CVI was 0.95.The test-retest reliability of each item and total score of the simplified version MBI was 0.909-0.991 for ICC.The inter-raters reliability of each item and total score of the simplified version MBI 0.880-0.992 for ICC.Cronbach's α coefficient of internal consistency of the simplified version MBI was 0.929.Using BI as the gold standard,the calibration validity(Spearman correlation coefficient)of the simplified version MBI was:eating(0.562),personal hygiene(0.701),bathing(0.715),going to the toilet(0.785),dressing(0.684),stool control(0.690),urination control(0.630),bed-chair transfer(0.925),walking/wheelchair(0.864),going up and down stairs(0.877),total score(0.941),all P<0.001.The simplified MBI construct validity analysis showed that the KMO statistic was 0.877,the Bartlett spherical test reject correlation Matrix was a unit matrix(P<0.05),and 80.28%of the total variance could be explained by two common factors.Factor 1 after rotation:Bathing,dressing,going to the toilet,bed and chair transfer,walking/wheelchair,going up and down stairs;factor 2 after rotation:eating,personal hygiene,stool control and urination control.Conclusion:The simplified version of MBI has good content validity,retest reliability,inter-evaluator reliability,internal consistency,calibration validity and construct validity,and also is easy to master and use.