中华护理杂志 ›› 2021, Vol. 56 ›› Issue (11): 1612-1617.DOI: 10.3761/j.issn.0254-1769.2021.11.002

• 论著 • 上一篇    下一篇

肠造口周围潮湿相关性皮肤损伤风险预测模型的构建及应用

刘莺歌(), 吴燕(), 曹秋君, 邱群, 吕桂芬, 徐洪莲, 高键   

  1. 200032 上海市 复旦大学附属中山医院护理部(刘莺歌,吴燕,曹秋君),营养科(高键);海军军医大学第一附属医院肛肠外科(邱群,吕桂芬,徐洪莲)
  • 收稿日期:2021-01-20 出版日期:2021-11-15 发布日期:2021-11-16
  • 通讯作者: 吴燕,E-mail: wu.yan@zs-hospital.sh.cn
  • 作者简介:刘莺歌:女,本科(硕士在读),E-mail: 18211170014@fudan.edu.cn
  • 基金资助:
    复旦大学-复星护理科研基金(FNF201942)

The development and validation of a risk prediction model for peristomal moisture-associated skin damage

LIU Yingge(), WU Yan(), CAO Qiujun, QIU Qun, LÜ Guifen, XU Honglian, GAO Jian   

  1. 【First-author’s address】 Nursing Department,Zhongshan Hospital,Fudan University,Shanghai,200032,China
  • Received:2021-01-20 Online:2021-11-15 Published:2021-11-16

摘要:

目的 构建肠造口周围潮湿相关性皮肤损伤(peristomal moisture-associated skin damage,PMASD)风险预测模型并应用于临床,验证该模型的预测效果。方法 采用前瞻性研究设计,便利选取2019年11月—2020年9月在上海市4所三级甲等医院接受肠造口术的329例患者作为建模组,根据患者术后3个月内是否发生PMASD分为并发症组(141例)和非并发症组(188例),基于单因素分析、Lasso变量筛选结果构建Logistic回归预测模型。采用Hosmer-Lemeshow检验判断模型的拟合度,受试者操作特征曲线下面积检验模型的区分度,Bootstrap重抽样法进行内部验证。于2020年10月—11月选取109例患者作为验证组对模型进行外部验证。结果 建模组术后3个月内PMASD发生率为42.9%(141/329),验证组为37.6%(41/109)。最终纳入模型的预测变量为性别(OR=0.497)、造口部位(OR=4.338)、造口开口高度(OR=0.304)、底盘区域手术切口(OR=3.479)、底盘贴合情况(OR=2.399)、软食(OR=3.155)、半流质饮食(OR=3.325)。Hosmer-Lemeshow检验P=0.274,受试者操作特征曲线下面积为0.812,约登指数为0.500,灵敏度为0.688,特异度为0.798。外部验证Hosmer-Lemeshow检验P=0.515,受试者操作特征曲线下面积为0.860,灵敏度为0.610,特异度为0.882。结论 该研究构建的预测模型预测效能良好,可为临床护理人员筛选PMASD高风险患者提供参考。

关键词: 肠造口术, 造口周围潮湿相关性皮肤损伤, 皮肤保护, 模型, 统计学, 护理

Abstract:

Objective To develop a risk prediction model of peristomal moisture-associated skin damage(PMASD),and to apply to clinical practice to validate its prediction performance. Methods The prospective study design and the convenience sampling method were adopted. A total of 329 patients from 4 hospitals in Shanghai were enrolled from November 2019 to September 2020,and predictors between the complication group(141 cases)and the non-complication group(188 cases) were compared using Logistic regression based on Lasso selection for model development. The goodness of fit of the model was verified by Hosmer-Lemeshow test. The discrimination was evaluated by the area under the ROC curve(AUC). The Bootstrap resampling was used for internal validation. From October to November 2020,109 patients were recruited for external validation. Results The incidence of complication in the development cohort was 42.9%(141/329),and 37.6%(41/109) in the validation cohort. The predictors that ultimately entered the prediction model were gender(OR=0.497),type of stoma(OR=4.338),height of stoma(OR=0.304),surgical wound in plate area(OR=3.479),plate attachment(OR=2.399),soft diet(OR=3.155),and semi-fluid diet(OR=3.325). The H-L test P=0.274;the AUC was 0.812. The Youden index was 0.500,with the sensiti-vity of 0.688 and the specificity of 0.798. While the external validation H-L test P=0.515,the AUC was 0.860 with the sensitivity of 0.610 and the specificity of 0.882. Conclusion The prediction model developed in this study has a good performance,which can provide a reference for clinical screening of high-risk patients with PMASD.

Key words: Enterostomy, Peristomal Moisture-Associated Skin Damage, Skin Care, Models, Statistics, Nursing Care