| 周晓飞,彭金燕,彭一鹏,罗会林.基于列线图预测模型分析抑郁症患者发生院内感染的危险因素.[J].中南医学科学杂志.,2024,(5):804-807. |
| 基于列线图预测模型分析抑郁症患者发生院内感染的危险因素 |
| Risk factors for nosocomial infection in patients with depression based on nomogram prediction model |
| 投稿时间：2023-12-14 修订日期：2024-04-18 |
| DOI：10.15972/j.cnki.43-1509/r.2024.05.028 |
| 中文关键词: 抑郁症 院内感染 列线图 危险因素 |
| 英文关键词:depression nosocomial infection nomogram risk factors |
| 基金项目:武汉市医学科研项目(WX21B26) |
|
| 摘要点击次数: 0 |
| 全文下载次数: 0 |
| 中文摘要: |
| 目的应用列线图预测模型分析抑郁症患者发生院内感染的危险因素。 方法收集305例抑郁症住院患者的临床资料,多因素Logistic回归分析抑郁症患者发生院内感染的危险因素,建立列线图模型并进行预测模型的验证和效能评估。 结果抑郁症住院患者院内感染率为8.20%。感染组年龄＞60岁、重度抑郁、合并糖尿病、使用抗抑郁药≥2种、侵入性操作、有营养不良风险患者占比均高于非感染组,住院时间长于非感染组,血红蛋白水平低于非感染组(P＜0.05)。Logistic结果显示,年龄＞60岁、住院时间＞28天、使用抗抑郁药≥2种、侵入性操作、重度抑郁、营养不良风险是院内感染的独立危险因素(P＜0.05)。建立的列线图预测模型预测院内感染的校正曲线趋近于理想曲线(P＞0.05)；模型预测的ROC下面积为0.850(P＜0.05)。 结论年龄、住院时间、使用抗抑郁药种类数、侵入性操作、重度抑郁、营养不良风险均是抑郁症患者发生院内感染的独立危险因素,建立的列线图模型可直观评估院内感染的风险。 |
| 英文摘要: |
| AimTo analyze the risk factors for nosocomial infection in patients with depression using a nomogram prediction model. MethodsThe clinical data of 305 depressive inpatients were collected. Multivariate Logistic regression analysis was used to identify risk factors for nosocomial infection in patients with depression. A nomogram model was established, and validation and performance evaluation of the predictive model were conducted. ResultsThe nosocomial infection rate of inpatients with depression was 8.20%. The proportion of patients with age＞60 years, severe depression, complicated with diabetes, number of using antidepressants ≥2, invasive procedures and at risk of malnutrition in the infection group was higher than that in the non-infection group, with longer hospital staying time as well as lower hemoglobin level (P＜0.05) in infection group compared with non-infection group. Logistic analysis showed that age ＞60 years, hospital staying, number of use antidepressants≥2, invasive procedures, severe depression, and malnutrition risk were independent risk factors for nosocomial infection (P＜0.05). The correction curve of nosocomial infection predicted by the nomogram model was close to the ideal curve (P＞0.05). The area under ROC curve predicted by the model was 0.850 (P＜0.05). ConclusionAge, length of hospital staying, types of antidepressants, invasive procedures, severe depression, and risk of malnutrition are all independent risk factors for nosocomial infection in patients with depression. The established nomogram model can intuitively assess the risk of nosocomial infection. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|