Background and Aims: The co-occurrence of behavioral addictions (BAs) and substance use disorders (SUDs) or other mental health problems (MHPs) is well documented. However, there is limited evidence on associations between changes in the severity of BAs, SUDs, and MHPs, or their directions of influence or causation. Methods: A non-self-selecting sample of 5,611 young Swiss men (mean age 25.5 at baseline and 28.3 at follow-up) completed a self-reporting questionnaire on various BAs (gambling, gaming, internet, internet pornography, smartphone, work), SUDs (alcohol, cannabis) and MHPs (major depressive disorder, ADHD, borderline personality disorder, social anxiety disorder). Latent change score models were used to evaluate pairwise, bidirectional associations in symptom severity among different BAs, and between BAs and SUDs or MHPs. Results: Overall, changes in each BA?s symptom severity were significantly and positively correlated with changes in the symptom severity of other BAs, alcohol use disorder, and MHPs; for cannabis use disorder, such correlations were only found with gaming and work. Significant bidirectional cross-lagged associations were found between the severity of BAs and MHPs, and between the severity of internet and smartphone addiction and other BAs. For SUDs, cross-lagged pathways were often not significant (e.g., with gambling or pornography) or even negative (between cannabis use disorder and work). Discussion and Conclusions: This study provides strong evidence that BAs and MHPs mutually reinforce each other over time. While this interplay can develop and maintain dysfunction, it may also enable positive change, highlighting the need for a comprehensive theoretical framework and integrated intervention approaches.  Kulcsszavak: Cohort Study on Substance Use Risk Factors (C-SURF), behavioral addictions, substance use disorders, mental health, latent change score model, cross-domain coupling