中文版脑瘫患儿粗大运动功能分级系统的信度和效度研究

Study on reliability and validity of the Chinese version of the gross motor function classification system for cerebral palsy

目的确定中文版脑瘫患儿粗大运动功能分级系统(Gross Motor Function Classification System,GMFCS)的信度和效度。方法共有来自上海3家康复机构91例0-12岁脑瘫患儿参加了此项研究,其中35例测定GMFCS的重测信度;66例测定GMFCS的评价者间信度;分别以88例脑瘫患儿的粗大运动功能评估量表(GMFM)和54例脑瘫患儿的 Peabody粗大运动发育量表(PDMS-GM)的各项测试结果为效标确定GMFCS的平行效度;对88例同时接受GMFCS和 GMFM评价的患儿的测试结果进行结构效度检测,以GMFCS为应变量,GMFM 5个功能区的百分比为自变量进行多元逐步回归分析,判断粗大运动中5个分区功能对GMFCS的影响程度。结果 GMFCS具有良好的重测信度(等级间相关系数为0．99),同时具有良好的评估者间信度(等级间相关系数为0．95-0．98);GMFCS与GMFM和PDMS-GM各项分值之间有良好的平行效度,Spearman r为-0．57--0．84;粗大运动功能中的坐位能力和行走能力是影响GMFCS的主要因素,校正决定系数为0．709(P<0．001)。结论中文版GMFCS具有良好的信度和效度,适用于国内对脑瘫患儿进行功能分级。

Objective To evaluate reliability and validity of the Chinese version of the Gross Motor Function Classification System. Methods 91 children（58 males and 33females ; mean age： 49.4 months, SD： 33months, Age range ： 7 to 144 months） with cerebral palsy（CP） were involved from three different Rehabilitation Centers in Shanghai. Types of CP in the children were hemiplegia （ n = 12）, spastic diplegia （ n = 31 ）, spastic quadriplegia （ n = 41 ）, spastic triplegia （ n = 1 ）, dyskinetic and mixed （ n = 5） ,ataxic （n = 1 ）. 35 children received a second assessment to assess test-retest reliability after a 10 day interval. Inter-tester reliability was tested in 66 CP children. Criterion-related validity was evaluated by comparing GMFCS levels with tests of Gross Motor Function Measure （ GMFM,n = 88 ） and Peabody Developmental Motor Scale-Gross Motor（ PDMS-GM, n = 54）. Construction validity was analyzed in 88 CP children who were tested with GMFCS and GMFM at the same time. Stepwise regression was performed to examine the multivariate associations between GMFCS levels and the five explanatory variables （ scores of five dimensions of GMFM）. Results Chinese version of GMFCS was found to possess excellent test-retest reliability（ ICC = 0. 99） as well as excellent inter-tester reliability reliability （ ICC = 0. 95 - 0. 98 ）. Criterion-related validity was excellent between GMFCS levels and scores of GMFM and PDMS-GM （Spearman rs = -0. 57 to -0. 84）. Sitting ability and walking ability of gross motor function were two main factors to determine GMFCS levels in CP children（ Adjusted R^2 = 0. 709,P 〈 0. 001 ）. Conclusions This study extends the reliability and validity of the Chinese version of GMFCS with supporting its use in clinical practice and research. GMFCS is suitable to classify the gross motor functions for children with CP in China.