目的 探讨尿路造口周围潮湿相关性皮肤损伤风险预测模型预测患者发生造口周围潮湿相关性皮肤损伤的效果。 方法 便利选取某三级医院2018年1月至2020年12月接受尿路造口术的169例患者作为研究对象，根据其术后3个月内是否发生造口周围潮湿相关性皮肤损伤分为并发症组（50例）和非并发症组（119例），利用Logistic回归建立风险预测模型，应用ROC曲线下面积检验模型预测效果，并选取2021年1-12月的54例患者进行模型预测效果验证。 结果 建模组术后3个月内造口潮湿相关性皮肤损伤发生率为29.6%（50/169），验证组为31.5%（17/54）。最终纳入模型的预测变量为体质量指数（body mass index,BMI）＜18.5（OR=0.101）、BMI 18.5~24（OR=0.134）、是否放化疗（OR=5.531）、造口定位（OR=3.853）、 造口部位（OR=5.788）、造口底盘渗漏次数＞3次（OR=5.211）、造口自护能力（OR=56.296）。Hosmer-Lemeshow检验P=0.775，受试者操作特征曲线下面积为0.951，约登指数为0.819，灵敏度为0.904，特异度为0.915。模型验证结果：灵敏度为0.882，特异度为0.948，准确率为96.3%。 结论 该研究构建的预测模型预测效能良好，可为临床护理人员及时对尿路造口周围潮湿相关性皮肤损伤高危患者采取预防性治疗和护理提供参考。

Objective To develop a risk prediction model of peristomal moisture-associated skin damage,and to apply the clinical practice to validate its prediction performance. Methods By convenience sampling method,169 patients in a hospital from January 2018 to December 2020 were enrolled and divided into a complication group (50 cases) and a non-complication group (119 cases) according to whether there was peristomal moisture-associated skin damage within 3 months after operation.Logistic regression was used to establish the risk prediction model,and the area under ROC curve was used to test the prediction effect of the model.About 54 patients from January to December 2021 were selected to verify the prediction effect of the model. Results The incidence of moisture related skin injury within 3 months after operation was 29.6% (50/169) in the modeling group and 31.5% (17/54) in the validation group.The final predictive variables included in the model were BMI＜18.5 (OR=0.101),BMI 18.5 to 24 (OR=0.134),radiotherapy and chemotherapy (OR=5.531),stoma location (OR=3.853),stoma site (OR=5.788),stoma chassis leakage times＞3 (OR=5.211),and stoma self-care ability (OR=56.296).Hosmer Lemeshow test P=0.775,the area under the operating characteristic curve of the subject was 0.951,the Youden index was 0.819,the sensitivity was 0.904,and the specificity was 0.915.The model validation results showed that the sensitivity was 0.882,the specificity was 0.948,and the accuracy was 96.3%. Conclusions The prediction model developed in this study has a good performance,which can provide a reference for preventative treatment and nursing measures for high-risk patients with peristomal moisture-associated skin damage.

R473.6