预测大学生的抑郁症状：结合近期和早期生活事件以及社会支持的诺模图模型

Predicting depressive symptoms in college students: A nomogram integrating recent and early life events with social support

预测大学生的抑郁症状：结合近期和早期生活事件以及社会支持的诺模图模型

Authors: Runyu Wei, Lanlan Li, Ying Zhang, Yu Liu, Rui Wang, Shuqin Li, Yuhui Wan

Source: Journal of Affective Disorders

PMID: 40393544

DOI: 10.1016/j.jad.2025.119444 

Abstract

Objective: Depressive symptoms is a common psychological problem among college students, although it lacks a prognostic prediction model. This study aims to develop a nomogram for depressive symptoms in first-year college students.

Method: A longitudinal survey was conducted among 6672 first-year students from three colleges in Anhui province, China. Logistic regression, lasso regression, and random forest models were combined to identify the most significant predictors of depressive symptoms. A nomogram was constructed based on multifactor logistic regression models.

Results: Seven risk factors for depressive symptoms were identified: emotional abuse, peer violence, academic stress, punishment, health adaptation, positive childhood experiences (PCEs), and support utilization. The training set, validation set (internal validation) and testing set (external validation) of the binary logistic regression-based model showed good discrimination (area under the curve (AUC) 0.757, 95%CI: 0.726-0.789; 0.699, 95%CI: 0.648-0.751; 0.709, 95%CI: 0.670-0.748, respectively), and accuracy (Brier scores of 0.061, 0.066, and 0.073, respectively). The nomogram shows good prediction of discrimination, calibration and generalization.

Conclusion: The comprehensive nomogram constructed in this study is a useful and convenient tool for assessing the risk of depressive symptoms among first-year college students. It will help healthcare professionals to assess the risk of depressive symptoms among first-year college students, identify high-risk groups and take more effective preventive measures.

Keywords: Adverse childhood experiences; College students; Depressive symptoms; Nomogram.

摘要

目的：抑郁症状是大学生中常见的心理问题，尽管目前尚缺乏预后预测模型。本研究旨在为大学一年级学生的抑郁症状建立一个列线图模型。

方法：在中国安徽省的三所高校中对 6672 名一年级学生进行了纵向调查。采用逻辑回归、套索回归和随机森林模型相结合的方法来确定抑郁症状的最重要预测因素。基于多因素逻辑回归模型构建了诺模图。

结果：确定了七个导致抑郁症状的风险因素：情感虐待、同伴暴力、学业压力、惩罚、健康适应、积极童年经历（PCEs）以及支持利用情况。基于二元逻辑回归模型的训练集、验证集（内部验证）和测试集（外部验证）均显示出良好的区分度（曲线下面积（AUC）分别为 0.757，95%CI：0.726-0.789；0.699，95%CI：0.648-0.751；0.709，95%CI：0.670-0.748），以及准确性（布里尔分数分别为0.061、0.066 和 0.073）。诺模图显示出良好的预测区分度、校准度和泛化能力。

结论：本研究构建的综合列线图是评估大学一年级学生抑郁症状风险的有用且便捷的工具。它将有助于医疗保健专业人员评估大学一年级学生的抑郁症状风险，识别高危人群，并采取更有效的预防措施。

关键词：不良童年经历；大学生；抑郁症状；诺模图。