艾滋病感染者抑郁症状相关因素分析的统计模型

Statistical model for analysis of factors related to depressive symptoms in HIV-infected patients

冯颖 ¹陈卫东 ²郑银霞 ²杨庚林 ²李月飞 ³何倩 ³倪明健⁴

扫码查看

作者信息

1. 新疆医科大学公共卫生学院,乌鲁木齐 830000;乌鲁木齐市妇幼保健院,乌鲁木齐 830000
2. 乌鲁木齐市妇幼保健院,乌鲁木齐 830000
3. 新疆医科大学公共卫生学院,乌鲁木齐 830000
4. 新疆维吾尔族自治区疾病预防控制中心,乌鲁木齐 830000
折叠

摘要

目的:运用3种统计模型分析育龄期女性HIV感染者抑郁症状的相关因素并进行比较.方法:选取育龄期女性HIV感染者553人,采用汉密顿抑郁量表(HAMD)评估感抑郁症状,选取logistic回归模型、人工神经网络模型、决策树模型分析抑郁症状的相关因素,采用ROC曲线比较3种模型的预测效果.结果:ROC曲线下面积由大到小依次排序为决策树模型、人工神经网络模型和logistic回归模型(AUC=0.813、0.707、0.701),两两比较显示,人工神经网络模型、logistic回归模型的ROC面积值均小于决策树模型的ROC面积值(均P＜0.01).结论:在预测HIV感染者抑郁症状方面,决策树模型的效果要优于人工神经网络模型和logistic回归模型.

Abstract

Objective:To analyze and compare the related factors of depressive symptoms in HIV-infected women of childbearing age using three statistical models.Methods:Totally 553 HIV-infected women of childbear-ing age were selected,and depressive symptoms of infected women were evaluated with the Hamilton Depression Scale(HAMD).Logistic regression model,artificial neural network model and decision tree model were selected to analyze the related factors of depressive symptoms of infected women,and ROC curve was used to compare the pre-diction effects of the three models.Results:The areas under ROC curve were sorted in order from large to small as decision tree model,artificial neural network model and logistic regression model(AUC=0.813,0.707,0.701).The ROC area values of artificial neural network model and logistic regression model were both smaller than those of decision tree model(P＜0.01).Conclusion:Decision tree model is better than artificial neural network model and logistic regression model in predicting depressive symptoms in HIV-infected patients.

关键词

育龄期女性/艾滋病感染者/歧视感知/抑郁/模型

Key words

women of childbearing age/infected with AIDS/perception of discrimination/depression/mod-el

引用本文复制引用

基金项目

新疆维吾尔自治区自然科学基金面上项目(2022D01A30)

新疆艾滋病防控研究重点实验室开放课题(XJYS1706-2021002)

出版年

2024

中国心理卫生杂志

中国心理卫生协会

中国心理卫生杂志

CSTPCDCHSSCD北大核心

影响因子：2.088

ISSN：1000-6729

段落导航