目的 构建急性白血病化疗患者肛周感染风险预警模型，以期为护理人员提供肛周感染风险筛查工具。 方法 2019年 1月至2021年12月，采用便利抽样法选取宁夏某三级甲等医院血液内科急性白血病化疗患者233例为研究对象，按其是否发生肛周感染分为感染组（68例）和非感染组（165例）。通过单因素分析构建二分类Logistic回归风险预警模型。以受试者特征曲线下面积验证模型区分度，以Hosmer-Lemeshow拟合优度检验评价模型拟合度。 结果 中性粒细胞计数(OR=7.738)、肛周基础疾病(OR=5.758)、住院天数(OR=4.809)、便秘(OR=4.112)、年龄(OR=3.452)是发生肛周感染的独立危险因素（均P＜0.05）。受试者特征曲线下面积为0.939（P＜0.001），最大约登指数为 0.804，最佳临界值为7.5，灵敏度为95.60%，特异度为84.80%，模型拟合度较好（P=0.717）。 结论 急性白血病化疗患者肛周感染风险预警模型有良好预测价值，得分越高，发生率越高，可作为临床肛周感染风险筛查的工具推广应用。

Objective To construct an early warning model of perianal infection,and to provide a risk screening tool for nursing staff. Methods From January 2019 to December 2021,233 patients with acute leukemia undergoing chemotherapy in the Department of Hematology of a tertiary A hospital in Ningxia were selected by convenience sampling method.According to whether they had perianal infection,they were divided into infected group (68 cases) and non-infected group (165 cases).Binary Logistic regression risk warning model was constructed by single factor analysis.The area under the receiver operating characteristic curve was used to verify the model differentiation,and the Hosmer-Lemeshow goodness of fit test was used to evaluate the model fit. Results Neutrophil count(OR=7.738),perianal underlying disease (OR=5.758)，length of stay(OR=4.809)，constipation(OR=4.112),and age(OR=3.452) were independent risk factors for perianal infection(all P＜0.05). The area under the receiver operating characteristic curve was 0.939(P＜0.001),the maximum approximate entry index was 0.804,the optimum critical value was 7.5,the sensitivity was 95.60%,the specificity was 84.80%,and the model fit was good(P=0.717). Conclusions The early risk warning model of perianal infection in patients with acute leukemia undergoing chemotherapy has good predictive value.The higher the score,the higher the incidence.It can be used as a tool for clinical risk screening of perianal infection.

R473.55

宁夏护理学会科研课题（NXHL182）