手足口病重症病例病情评分量表的ROC曲线分析

目的建立手足口病重症病例的症状评价指标,探索方便、快捷、有效的小儿手足口病重症病例评分法,为基层医疗机构早期快速判定手足口病重症病例和病情严重程度提供参考依据。方法对2014—2016年成都市重症手足口病病例及同期住院轻症病例资料进行回顾性分析,参照小儿危重病例评分法(Pediatric Critical Illness Score,PCIS)、既往文献研究资料,筛选出相对简单易行的症状作为主要评价指标,建立适用于快速判定重症手足口病患者及其病情严重程度的评分法。通过受试者工作特性(receiver operating characteristic,ROC)曲线下面积(area under the ROC curve,AUC)评价并比较评分表和评分系统工作效能。结果应用重症手足口病评分法得到重症组、Ⅳ期重症组病例的ROC曲线AUC分别为0.99、0.97,最优临界值评分对应的最大约登指数分别为0.96、0.89,说明此评价方法对重症和Ⅳ期重症手足口病例的分辨力较好。该评分表用于界定Ⅱ期与Ⅲ期重症手足口病病例的效果欠佳,AUC仅为0.62。结论此重症手足口病评分法能评判手足口病重症和Ⅳ期重症病例病情,为基层医疗机构儿科医生临床重症手足口病病例的早诊早治提供参考依据。

Receiver operating characteristic curve of illness scoring scale for cases of severe hand,foot and mouth disease

Objective To establish symptom evaluation indexes of cases of severe hand, foot and mouth disease (HFMD), to explore a convenient, fast and effective method for scoring cases of children with severe HFMD, and to provide a reference basis for early and rapid determination of severe HFMD cases and severity of HFMD in primary-level medical institutions. Methods The data regarding severe HFMD cases and hospitalized cases with mild HFMD in Chengdu city during 2014-2016 were retrospectively analyzed. According to Pediatric Critical Illness Score (PCIS) and previous literature research data, relatively simple and feasible symptoms were selected as the main evaluation indexes, and a scoring method suitable for rapid determination of severe HFMD patients and their severity was established. The receiver operating characteristic (ROC) curve and the corresponding area under the ROC curve (AUC) were used to evaluate and compare the performance of the scoring scale and the scoring system. Results The AUC of cases in severe HFMD group and Ⅳ severe group was 0.99 and 0.97, respectively, and the maximum Yoden indexes corresponding to the optimal threshold score were 0.96 and 0.89, respectively, showing that this evaluation method had a good discrimination for severe and IV severe HFMD cases. However, the scoring scale was not effective for defining cases of stages Ⅱ and Ⅲ HFMD, and the AUC was only 0.62. Conclusions This scoring method of severe HFMD can judge the condition of cases of severe and IV severe HFMD, and provide a reference basis for pediatricians of primary-level medical institutions to early diagnose and treat clinical cases of severe HFMD.