Multi dimensional host susceptibility indicators in ICU in-hospital infection nursing risk assessment weight construction and verification
多维度宿主易感性指标在ICU院内感染护理风险评估中的权重构建与验证

摘要

目的 探讨多维度宿主易感性指标在ICU院内感染护理风险评估中的价值。方法 通过回顾性分析我院2024年1月1日至12月31日期间ICU住院时间超过48小时的重症患者临床资料，重点分析感染病例的宿主特征，包括基础疾病、营养状态、侵入性操作、免疫功能、抗菌药物使用等维度。建模组数据进行单因素分析，筛选候选变量后纳入多因素Logistic回归，构建感染风险预测模型。结果 多因素分析显示，合并慢性基础病、营养风险评分（NRS≥3）、呼吸机使用时间＞5天、中心静脉置管＞2次、CRP＞50mg/L、多重抗菌药物联合应用等6项为ICU院内感染的独立危险因素（P＜0.05）。构建的评分模型总分范围为0～18分，截断值定为≥10分判定为高风险。结论 基于多维度宿主易感性指标的护理风险评分模型可有效预测ICU院内感染发生风险。通过该模型进行高危人群筛查，有助于早期干预、优化护理流程。

Abstract

Objective To explore the value of multidimensional host susceptibility indicators in risk assessment of ICU infection nursing.
Methods A total of 213 patients with ICU stay exceeding 48 hours in our hospital from January 1 to December 31, 2024 were retrospectively included. They were randomly divided into a modeling group (n=149) and a validation group (n=64) in a 7:3 ratio using a random number table method. The system collects multidimensional host characteristics including underlying diseases, nutritional status, frequency of invasive procedures, levels of immune indicators, and use of antibiotics. Perform univariate analysis on the modeling group data, screen candidate variables, and include them in multivariate logistic regression to construct an infection risk prediction model.
Results Multivariate analysis showed that the presence of chronic underlying diseases, nutritional risk score (NRS ≥ 3), duration of ventilator use>5 days, central venous catheterization>2 times CRP＞50mg/L、 The combination of multiple antibiotics and other six factors were independent risk factors for ICU infections (P<0.05). The total score range of the constructed scoring model is 0-18 points, with a cutoff value of ≥ 10 points to determine high risk.
Conclusion   A nursing risk scoring model based on multidimensional host susceptibility indicators can effectively predict the risk of ICU hospital acquired infections. Screening high-risk populations through this model can help with early intervention and optimize nursing processes.