| 王洁.个体化预测精神科女性住院患者医院感染发生风险的列线图模型构建[J].浙江中西医结合杂志,2026,36(4): |
| 个体化预测精神科女性住院患者医院感染发生风险的列线图模型构建 |
|
| 投稿时间：2025-01-15 修订日期：2026-01-16 |
| DOI： |
| 中文关键词: 精神科 女性住院患者 医院感染 影响因素 列线图模型 |
| 英文关键词:Psychiatry Female inpatients Hospital infection Influencing factors Column chart model |
| 基金项目: |
|
| 摘要点击次数: 9 |
| 全文下载次数: 0 |
| 中文摘要: |
| 【】目的:探讨精神科女性住院患者发生医院感染的影响因素,构建列线图预测模型。方法:选取本院精神科2020年1月至2023年4月期间所有住院治疗且符合标准的699例女性患者为研究对象,随机分为建模组524例和验证组175例,建模组患者根据是否发生医院感染分为感染组49例和未感染组475例。采用logistic多因素回归分析筛选精神科女性住院患者医院感染的危险因素；采用R 3.6.1软件绘制预测精神科女性住院患者医院感染发生风险的列线图模型,采用Bootstrap法对列线图模型进行内部验证；采用校正曲线和受试者工作特征(ROC)曲线评估列线图模型的区分度、校准度和预测效能。结果:感染组49例精神科女性住院患者感染部位分别为呼吸道(57.14%)、泌尿道(16.33%)、胃肠道(10.20%)、口腔(6.13%)、其它(10.20%)。感染组年龄≥60岁、住院时间≥4周、有基础疾病、有药物不良反应、管理方式为封闭式管理的患者比例显著高于未感染组(P＜0.05)。年龄≥60岁、住院时间≥4周、有基础疾病、有药物不良反应、管理方式为封闭式管理是精神科女性住院患者医院感染的独立危险因素(P＜0.05)。列线图模型内部验证结果显示,一致性指数(CI)为0.914,Hosmer-Lemeshow拟合优度检验结果显示χ2=1.117,P＞0.05,校正曲线趋近于理想曲线。ROC曲线的曲线下面积为0.927(95%CI:0.883~0.970),敏感度为86.61%,特异度为90.16%。列线图模型外部验证结果显示,CI为0.922,Hosmer-Lemeshow拟合优度检验结果显示χ2=1.427,P＞0.05,校正曲线趋近于理想曲线。ROC曲线的曲线下面积为0.970(95%CI:0.944~0.997),敏感度为93.88%,特异度为87.64%。结论:本研究构建的列线图模型有利于个体化预测精神科女性住院患者医院感染的发生风险,有利于临床尽早制定干预策略。 |
| 英文摘要: |
| Objective: To explore the influencing factors of hospital infection in female psychiatric inpatients and construct a column chart prediction model. Methods: A total of 699 female patients met the standards and admitted to the psychiatric department of our hospital from May 2020 to April 2023 were selected as the study subjects, they were divided into a modeling group of 524 cases and a validation group of 175 cases randomly, the modeling group were 49 cases in the infected group and 475 cases in the uninfected group. Logistic multiple factor regression analysis was applied to screen risk factors of hospital infection in female hospitalized psychiatric patients; R 3.6.1 software was applied to draw a column chart model for predicting the risk of hospital infection in female psychiatric inpatients, Bootstrap method was applied to internal validation of the column chart model; calibration curve and receiver operating characteristic (ROC) curve were applied to evaluate the discrimination, calibration, and predictive performance of the column chart model. Results: The infection sites of 49 female inpatients in the infected group included respiratory tract (57.14%), urinary tract (16.33%), gastrointestinal tract (10.20%), oral cavity (6.13%), and others (10.20%). The proportions of age ≥ 60 years old, hospital stay ≥ 4 weeks, underlying diseases, adverse drug reactions, and closed management in the infected group was significantly higher than that in the uninfected group (P<0.05). Age ≥ 60 years old, hospital stay ≥ 4 weeks, presence of underlying diseases, adverse drug reactions, and closed management were independent risk factors for hospital infection in female psychiatric inpatients (P<0.05). The internal validation results of the column chart model showed that the consistency index (CI) was 0.914, and the Hosmer-Lemeshow goodness of fit test results showed χ2=1.117, P>0.05, the calibration curve approached the ideal curve. The area under the ROC curve was 0.927 (95% CI: 0.883~0.970), with a sensitivity of 86.61% and a specificity of 90.16%. The external validation results of the column chart model showed that the consistency index (CI) was 0.922, and the Hosmer-Lemeshow goodness of fit test results showed χ2=1.427, P>0.05, the calibration curve approached the ideal curve. The area under the ROC curve was 0.970 (95% CI: 0.944~0.997), with a sensitivity of 93.88% and a specificity of 87.64%. Conclusion: The column chart model constructed in this study is conducive to personalized prediction of the risk of hospital infection in female psychiatric inpatients, and is conducive to developing clinical intervention strategies as soon as possible. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
