以患者为主的炎症性肠病患者PRO量表特异模块条目筛选

Screening of Special Abnormal Module Items in PRO Scale for Patients with Inflammatory Bowel Disease.

目的研制出适合我国国情的炎症性肠病患者报告临床结局测定量表(PRO-IBD)特异模块,以构建完整的消化系统体系式量表。方法通过定性访谈和定量结合,应用临界比法、线性相关系数法、克朗巴赫系数a法、离散趋势法、因子分析法、医生重要性评分6种方法对32名消化科医生与69名IBD患者的访谈及问卷数据进行统计分析,最后再组织专家讨论确定最终条目。结果临界比值法显示高分组与低分组在每个条目的平均数都有差异(P<0.05);离散趋势法中第5、8条目的SD<0.8,予以删除;线性相关系数法结果显示各条目与总分的相关系数均大于0.5,不予删除任何条目;Cronbach系数α法中分别删除条目14、15计算得到的领域Cronbach系数α<0.890,予以删除;15个条目中医生重要性评分平均数在3.57~4.36分之间(>3.5分),满分比为19.0%~61.9%,可说明专家对条目可操作性的集中程度高,且最终通过专家和患者角度所筛选的条目一致。结论按以上步骤筛选出的15个条目有较好的信度、效度和代表性,在以后的研究里可进一步考核该量表的推广性,并检验其在临床中的应用价值。

Objective To develop a special module of PRO-IBD for patients with inflammatory bowel disease in China,so as to construct a complete digestive system scale.Methods The interview and questionnaire data of 32 gastroenterologists and 69 IBD patients were statistically analyzed by six methods:critical ratio method,linear correlation coefficient method,Cronbach coefficient A method,discrete trend method,factor analysis method and doctor importance score.Finally,experts were organized to discuss and determine the final items.Results The critical ratio method shows that the average number of each item is different between the high group and the low group(P<0.05)The SD of items 5 and 8 in the discrete trend method is less than 0.8,which should be deleted;The results of linear correlation coefficient method show that the correlation coefficient between each item and the total score is greater than 0.5,and no item will be deleted;In Cronbach coefficientαmethod,items 14 and15 are deleted respectively,and the Cronbach coefficientα<0.890 is calculated and deleted;Among the 15 items,the average score of doctors’importance is between 3.57 and 4.36(>3.5),and the full score ratio is between 19.0%and 61.9%,which shows that experts have a high degree of concentration on the operability of items,and the items screened by experts and patients are consistent.Conclusion The 15 items selected according to the above steps have good reliability,validity and representativeness,which can further evaluate the popularization of the scale and test its clinical application value in future research.