Bidirectional relationship between (Internet) Gaming Disorder, internalizing psychopathology, psychological distress, and well-being: A systematic review with meta-analysis of longitudinal studies

There is a lack of a synthesis of longitudinal studies that clarify the relationship between gaming disorder (GD) symptoms and mental health issues, psychological distress, and well-being.To address this gap, we undertook a systematic review with meta-analysis of longitudinal studies. The study protocol was preregistered online on PROSPERO (CRD42023407665). We included studies analyzing the association between GD or problematic/pathological videogame use, and internalizing psychopathology (i.e., depression and anxiety), psychological distress (i.e., loneliness, stress), or well-being (i.e., life satisfaction, quality of life, well-being) over time. Studies that only measured gaming time or video game use, as well as treatment and experimental studies, were excluded. We rated the risk of bias (study quality) using the National Institutes of Health quality assessment tool for observational cohort studies. Research articles were searched in PsycInfo, PsycArticles, PubMed, and Web of Science up until December 29, 2022. 36 articles (N subjects = 34,288) were included in the qualitative synthesis whereas 30 were considered for the quantitative synthesis (meta-analysis) (N = 28,782). Six studies were excluded because they lacked sufficient data to compute partial correlations. Effect sizes (i.e., partial correlation) were pooled using random-effects models. Sensitivity analyses were carried out excluding studies rated as “poor” and Chinese studies. Effect sizes for the association between GD and subsequent depression (n studies = 17, n subjects = 17,017, ρ = .077, 95% CI: .035, .118), emotional symptoms (i.e., a combination of internalizing symptoms) (n studies = 4, n subjects = 8,054, ρ = .083, 95% CI: .049, .116) and life satisfaction (n studies = 8, n subjects = 11,058, ρ = -.060, 95% CI: -.104, -.015) were weak but statistically significant. GD was not associated with subsequent anxiety (n studies = 9, n subjects = 7,149, ρ = .020, 95% CI: -.024, .064). On the other hand, depression (n studies = 18, n subjects = 17,756, ρ = .068, 95% CI: .020, .115), anxiety (n studies = 9, n subjects = 7,149, ρ = .076, 95% CI: .039, .113), emotional symptoms (n studies = 4, n subjects = 8,054, ρ = .075, 95% CI: .001, .148), life satisfaction (n studies = 8, n subjects = 11,058, ρ = -.053, 95% CI: -.080, -.025) and loneliness (n studies = 3, n subjects = 1,584, ρ = .096, 95% CI: .033, .160) were significantly associated with subsequent GD. Overall, sensitivity analyses supported the robustness of the results from the main analysis. In conclusion, this study found reciprocal associations between GD and depression, GD and life satisfaction, and GD and mixed emotional symptoms over time, a longitudinal effect of anxiety on GD (but no evidence for the inverse), and of loneliness on GD. However, the causal nature and practical relevance remain uncertain because effect sizes were small and based on observational studies. This research was supported by a Swiss Government Excellence Post-Doctoral Scholarship [2022.0033] awarded to the first author (SA).Keywords: gaming disorder; addictive disorder; meta-analysis; longitudinal; development.