Objective: The development of serious games for mental wellbeing is a topic of growing interest. The increase in acceptance of games as a mainstream entertainment medium combined with the immersive qualities of games provides opportunities for meaningful support and intervention in mental wellbeing. Method: We conducted a systematic review and exploratory meta-analysis to examine if aspects of the interventions influenced outcomes as measured via overall effect sizes. We employed a multilevel meta-analytic approach to accommodate the interdependency of effect sizes (18 effect sizes from 14 studies, with 2027 participants). Results: Overall, the main effect for gaming interventions on any outcome variable was small to medium sized, d =.35 (confidence interval [.23, 47], p <.001). Results revealed that the only significant moderator was the nature of the intervention. Specifically, only interventions that included a rational emotional behavioural focus significantly predicted an improvement in depression and/or anxiety in participants. Conclusion: The findings reveal promising effects for therapeutic games for mental health, but replications are needed, alongside the addressing of methodological and procedural concerns.