脑卒中后生活质量量表3.0代理人版中文版的测试及评价

背景对于卒中的幸存者来说,生存质量是患者对于自身健康、生活满意程度、物理功能的主观评定.对于那些不能自己完成量表的卒中患者,我们可以尝试替代测量.然而迄今还没有学者将QOL的代理人量表引入中国.我们的研究目的是编译脑卒中生存质量影响量表(SIS 3.0)代理人版本,考察中文版的信度、效度和反应度,探讨代理人代替病人测试生存质量(QOL)的可行性.方法翻译(将原量表译成中文)和回译(将译文译回英文,与原文比较)原量表后,并作适当改进,然后对量表作心理学考评.选择10对病人及其代理人进行初试,再另选231对病人及代理人用于正式研究.病人的入选条件年龄＞18岁;脑卒中患者(包括脑出血和脑梗死);脑卒中的诊断符合WHO 诊断标准;且都有CT/MR证据.排除条件为TIA;卒中前有永久性残疾;严重的合并症(Ⅲ-Ⅳ级心衰,必须血液透析,严重的肌肉骨骼疾患影响物理功能,癌症,活动期精神病或痴呆,AIDS).代理人的入选条件年龄＞18岁;认识病人至少1年;每周与病人会面的次数至少1次,每次见面的时间至少1小时.排除条件精神状态简易速检表(MMSE)量表评分＜16分.对这些病人和代理人进行一系列量表的测试,包括SIS 3.0代理人版中文版,和其他公认的用于做测评工具的量表,包括医疗结局研究简表(medical outcomes study form 36, MOS SF-36)的中文版、Barthel指数(Barthel Index, BI)、汉密顿抑郁量表(HAMD)考察各个领域的标准效度.使用牛津残障评分(OHS)作为分组依据来考察SIS 3.0的反应度.我们选择了量表完成率和完成量表的时间两个指标来考察可行性;选用分半信度和克朗巴赫系数(α)来考察信度;选用内容效度、标准效度、结构效度来考察效度;通过比较不同残疾程度的病人的SIS 3.0得分来考察反应度.此外,我们比较了病人同代理人得分的差异.结果该中文版量表可行性好.分半信度和α系数均大于0.8,表明信度好.经专家评审后,认为表的内容与测试目的吻合.此外,量表各个测试条目得分与所属领域总分的Pearson相关系数在0.620～0.969之间,P值＜0.01.因子分析结果表明量表的结构与设计时的构想基本吻合.因为目前缺乏"交流"领域金标准,因此没有作该领域的标准效度检验.剩余的七个领域与已知公认量表的相应领域高度相关,Pearson相关系数＞0.5,P=0.000,具有良好的相关性,说明标准效度好.牛津残障评分(OHS)分组下的单因素方差分析显示量表可以区分不同残疾程度的病人,且随着病人残疾程度的加重,QOL评分总体呈下降趋势,表明反应度好.使用配对t检验和组间相关系数,分领域考察病人和代理人得分的差异,结果表明无统计学差异.结论 SIS 3.0代理人版中文版的信度、效度和反应度是满意的,用代理人代替病人进行QOL研究是可行的.

Research on translation and psychometric evaluation and test of chinese version of Stroke Impact Scale 3.0 for Proxy

Background Quality of life(QOL) refers to the person's subjective appraisal of well-being,life satisfaction,health and functional performance. For those stroke patients who cannot finish the scale themselves,we can try the proxy measurement. However,until recently no QOL scale for proxies were induced to Chinese stroke patients. The goal of this research is to translate and test the SIS 3.0 for proxy,and differences between patient and proxy scores. To translate and test the SIS 3.0 for proxy, and analyzed differences between patient and proxy scores.Methods Ten pairs of patients and their proxies were involved in the primary test. Two hundreds and thirty-three pairs were involved in the formal test. We analyzed the validity, responsiveness, reliability and feasibility of the SIS 3.0 for proxy, as well as the validity in proxy assessment. Results The feasibility was sufficient. Both Split reliability and α coefficient were more than 0.8, demonstrating SIS a reliable instrument. SIS had a good content validity with correlation coefficient more than 0.6. Good criterion validity was established by comparing the scores on various domains to standardized measures with P=0.000. Construct validity was also good as indicated by factor analysis. Proxy scores were significantly different across OHS scales which showed domain responsiveness was good. Comparison of patient and proxy responses resulted in no significant difference. Conclusions SIS for proxy is satisfactory for chinese patients. It is feasible to use a proxy respondent to answer questions on the patient's behalf.