中国成人斜视生存质量量表的Rasch分析

Evaluation of the Chinese Adult Strabismus-20 （CAS-20） questionnaire using Rasch analysis

目的 通过Rasch分析提炼中国成人斜视生存质量量表（CAS-20）,以能更好地用于评估成人斜视患者的生存质量状况.方法 横断面研究.246例成人斜视患者,独立完成CAS-20问卷调查.根据量表每一条目的得分情况进行Rasch分析,分析条目及答案的心理计量特征.分析的项目包括分解维度、选项排序、局部独立性、适合度指标、项目差别功能、项目适中性分析、最终确定维度等.结果 原CAS-20包括2个维度,为“社会心理维度”和“视觉功能维度”.通过Rasch分析可将量表分为4个维度,分别为“自我评价”、“与他人交往”、“阅读功能”和“一般功能”.原先的选项排序对每个维度均非常合理.每个维度的各条目之间标准残差相关系数均＜0.7,说明同一维度内条目之间具有一定的独立性.通过适合度指标分析,除“与他人交往”维度外,其他3个维度均需剔除1个条目.项目差别功能分析提示年龄和性别在4个维度中均无明显差异.项目适中性分析发现项目难度和研究对象个体能力之间匹配较好,阅读功能维度条目相对于患者个体能力来说略为简单.结论 CAS-20通过Rasch分析,20个条目缩减为17个条目,并且分为4个维度,分别是“自我评价”、“与他人交往”、“阅读功能”和“一般功能”,每个条目的选项仍为5个.优化的CAS-20可以更敏感地反映中国成人斜视患者的生存质量状态.

Objective To further refine the Chinese Adult Strabismus-20 （CAS-20） health-related quality of life （HRQOL） questionnaire using Rasch analysis.Methods In this cross-sectional study,the CAS-20 questionnaire was independently completed by 246 adult strabismus patients.Based on the score for each item,Rasch analysis was performed on the original CAS-20 using the following aspects：dimensionality,response order,local dependence,infit and outfit analysis,differential item function,subject targeting and confirmatory dimensionality.Results The original CAS-20 was divided into 4 subscales,which were labeled self-perception,interaction,reading function,and general function.Response ordering was appropriate for all of the 4 subscales.As a result of infit and outfit analysis,3 items were removed,one each from self-perception,reading function,and general function.No significant differential item function was found for sex or age.Compared with the ability of the patients and item difficulty,the 3 subscales matched well except for reading function.Conclusion The original CAS-20 was reduced to 17 items,and divided into 4 subscales,which were self-perception,interaction,reading function,and general function.Subject response choices saved 5 test options.The refined CAS-20 may prove to be more responsive to HRQOL changes in adult strabismus patients.