量化网络成瘾对青少年抑郁影响中的性别差异：一种因果推断方法。

Quantifying gender differences in the impact of internet addiction on adolescent depression: A causal inference approach.

作者信息

Tarimo Clifford Silver, Feng Yifei, Jia Shiyu, Wu Xiaoman, Zhao Weijia, Zuo Yibo, Wang Yuhui, Bi Yuefeng, Wu Jian

机构信息

School of Pharmaceutical Science, Zhengzhou University, Ke Xue Da Dao 100, Zhengzhou 450001, Henan, China; Department of Health Management, College of Public Health, Zhengzhou University, Zhengzhou, Henan, China; Department of Science and Laboratory Technology, Dar es Salaam Institute of Technology, P.O. Box 2958, Dar es Salaam, Tanzania.

Department of Health Management, College of Public Health, Zhengzhou University, Zhengzhou, Henan, China.

出版信息

J Affect Disord. 2025 Jun 15;379:793-802. doi: 10.1016/j.jad.2025.03.079. Epub 2025 Mar 15.

DOI:10.1016/j.jad.2025.03.079

PMID:40097112

Abstract

BACKGROUND

The rising prevalence of adolescent depression in China has raised concerns about internet influence. This study identified predictors of internet addiction (IA) and quantified its gender-specific impact on depression.

METHOD

In April-May 2023, 8176 adolescents from six (6) junior high schools in Henan, China, were randomly sampled. IA and depression were assessed via the 20-item internet addiction test (IAT) and 10-item Center for Epidemiologic Studies Depression (CESD-10) scale, respectively. IA's impact on depression was analyzed using four approaches: IPW (confounder weighting), DML (high-dimensional control), PSM (covariate balancing), and RA (residual adjustment).

RESULTS

Among the males, 31.8 % reported depression and 17.7 % reported IA while females, 49.4 % reported depression and 21.3 % IA. The mean ages were 14.5 ± 0.94 (males) and 14.4 ± 0.93 (females). DML revealed the overall effect of 0.22 (95 % CI: 0.17-0.23; p < 0.001), higher in males (95 % CI: 0.23, 95 % CI: 0.18-0.31) than females (0.17, 95 % CI: 0.11-0.23). IPW estimated an effect of 0.30 (95 % CI: 0.25-0.34; p < 0.001), also higher in males (0.34, 95 % CI: 0.28-0.39) than females (0.29, 95 % CI: 0.23-0.36). PSM and RA yielded similar results. Predictors of IA included low self-esteem, history of negative life events, poor sleep quality, and one-child family status.

CONCLUSION

Internet addiction (IA) exacerbates adolescent depression, disproportionately affecting males. Addressing self-esteem, negative life events, poor sleep, and one-child family challenges can mitigate IA's effects. Gender-sensitive interventions, school-based programs, and parental guidance are critical.

摘要

背景

中国青少年抑郁症患病率的上升引发了对网络影响的关注。本研究确定了网络成瘾（IA）的预测因素，并量化了其对抑郁症的性别特异性影响。

方法

2023年4月至5月，从中国河南六所初中随机抽取8176名青少年。分别通过20项网络成瘾测试（IAT）和10项流行病学研究中心抑郁量表（CESD - 10）评估网络成瘾和抑郁症。使用四种方法分析网络成瘾对抑郁症的影响：逆概率加权法（IPW，混杂因素加权）、双机器学习法（DML，高维控制）、倾向得分匹配法（PSM，协变量平衡）和残差调整法（RA）。

结果

男性中，31.8%报告有抑郁症，17.7%报告有网络成瘾；女性中，49.4%报告有抑郁症，21.3%报告有网络成瘾。平均年龄分别为14.5±0.94岁（男性）和14.4±0.93岁（女性）。双机器学习法显示总体效应为0.22（95%置信区间：0.17 - 0.23；p < 0.001），男性（95%置信区间：0.23，95%置信区间：0.18 - 0.31）高于女性（0.17，95%置信区间：0.11 - 0.23）。逆概率加权法估计效应为0.30（95%置信区间：0.25 - 0.34；p < 0.001），同样男性（0.34，95%置信区间：0.28 - 0.39）高于女性（0.29，95%置信区间：0.23 - 0.36）。倾向得分匹配法和残差调整法得出类似结果。网络成瘾的预测因素包括自卑、负面生活事件史、睡眠质量差和独生子女家庭状况。