青少年游戏障碍症状潜在类别及其与昼夜节律的关系

Latent Profiles of Gaming Disorder Symptoms in Adolescents and Their Relationship with Circadian Rhythm

摘要: 背景：游戏障碍在青少年中流行率高、损害显著，当前研究多以变量为中心，难以揭示群体内部的异质性；同时昼夜节律与游戏障碍症状的关联机制尚未明确，限制精准干预策略的制定。目的：基于潜在剖面分析(LPA)探索青少年游戏障碍症状的潜在类别并考察其与昼夜节律的关联，为针对性干预提供依据。方法：采用横断面设计，于2024年4月采用整群随机抽样，对15,721名初高中学生进行线上问卷调查，使用青少年游戏障碍量表(GADIS-A)、清晨–夜晚型量表(MEQ-19)对青少年的游戏障碍症状和昼夜节律进行评估。采用LPA识别游戏障碍症状潜在类别，通过多元逻辑回归分析昼夜节律类型对各类别的预测效应。结果：根据模型拟合指标均随类别数增加而下降，且5类别模型的Entropy = 0.871，LMR-LRT与BLRT均P < 0.001，将青少年游戏障碍症状分为“无症状型”(19.74%)、“轻度症状型”(42.34%)、“高沉迷–低损害型”(10.76%)、“中度症状型”(21.44%)及“重度症状型”(5.71%) 5种潜在类型，回归分析显示，昼夜节律类型为独立预测因素(χ² = 972.82, P < 0.001)：以中间型昼夜节律为参照组，夜晚型青少年归属重度及高沉迷型的概率显著升高(重度型：OR = 4.32，95%CI: 3.59~5.20；高沉迷型：OR = 3.69，95%CI: 3.19~4.28)；而清晨型对各类别均呈保护作用。结论：青少年游戏障碍症状存在5种潜在异质性分化，昼夜节律类型(尤其是夜晚型)可作为较高风险群体的早期筛查指标。

Abstract: Background: Gaming disorder is highly prevalent among adolescents and is associated with significant impairment. Most current research adopts a variable-centered approach, which fails to capture the heterogeneity within populations. Moreover, the mechanisms linking circadian rhythm to gaming disorder symptoms remain unclear, limiting the development of targeted intervention strategies. Objective: This paper aims to identify latent profiles of gaming disorder symptoms among adolescents using latent profile analysis (LPA), and to examine their association with circadian rhythm types, thereby providing evidence for precise intervention. Methods: A cross-sectional study was conducted in April 2024 using convenience sampling. A total of 15,721 middle and high school students completed an online survey assessing gaming disorder symptoms and circadian rhythm using the Gaming Disorder Scale for Adolescents (GADIS-A) and the Morningness-Eveningness Questionnaire (MEQ-19), respectively. LPA was employed to identify latent profiles of gaming disorder symptoms, and multinomial logistic regression was used to analyze the predictive effects of circadian rhythm types on each profile. Results: Model fit indices decreased as the number of profiles increased, and the five-profile model showed an entropy value of 0.871, with both LMR-LRT and BLRT tests significant (P < 0.001). Five distinct latent profiles were identified: “asymptomatic” (19.74%), “mild symptoms” (42.34%), “high impulsivity-low harm” (10.76%), “moderate symptoms” (21.44%), and “severe symptoms” (5.71%). Regression analysis indicated that circadian rhythm type was an independent predictor (χ² = 972.82, P < 0.001): compared with the intermediate type, evening-type adolescents had significantly higher odds of being classified as “severe” (OR = 4.32, 95%CI: 3.59~5.20) and “high impulsivity–low harm” (OR = 3.69, 95%CI: 3.19~4.28), while the morning type showed a protective effect across all profiles. Conclusion: There is notable heterogeneity in adolescent gaming disorder symptoms, with five latent profiles identified. Circadian rhythm type—especially the evening type—may serve as an early screening indicator for high-risk groups.