Aniek Wols

2 67 REVIEW OF APPLIED & CASUAL GAMES FOR MENTAL HEALTH into account, which may have overestimated our effects. Moreover, the heterogeneity among included studies prevented the examination of effect sizes in the follow-up period. In order to draw conclusions about long-term effects of digital games on mental health, future research should include follow-up measurements. Furthermore, our method did not account for differences in the relative sample sizes of the studies (McKenzie & Brennan, 2022) and together with having the same study contributing multiple effect sizes this may have biased the results. Our aim, however, was to provide a complete overview of research that has evaluated applied and casual games for mental health improvement in youth and to examine current research trends by reviewing methodological characteristics such as type of control group(s) and the use of multi-modal interventions. Making decisions about which comparison or outcome variable from a specific study should be presented, would have undermined our aim. The strength of the current review is that it gives insight into the frequency of certain control groups and the relative effectiveness of the games to these control groups, which is crucial information to move the field forward. At the same time, we acknowledge that the breadth of the current review did not allow us to provide an in-depth examination of the effect of applied and casual games on specific mental health outcomes and this is a trade-off that future research should take into account. The second limitation of the current review is related to the reliability of the findings. Although the risk-of-bias judgements showed that the majority of included papers had a low or moderate overall risk of bias, the overall strength of the evidence is limited by the number of studies with (comparable) measurable outcomes per mental health area and the available data that could be obtained. For some mental health domains, the number of studies on which effect sizes could be calculated was small, limiting the conclusions that could be drawn from those analyses. Future research should attempt to harmonise outcome measures and to provide all necessary data for systematic reviews and meta-analyses in the paper itself or make them easily accessible through an online repository. Finally, the current review specifically included studies that employed a randomised controlled design with pre and post measurements and, as a result, studies evaluating digital games through a different design may have been overlooked. Randomised controlled designs, however, follow robust methodological procedures and are considered to provide the highest quality of evidence. Additionally, by far most games currently available have no experimental evidence (Donker et al., 2013) and Bakker et al. (2016) recommend that randomised controlled trials are required to validate future games. Thus,

RkJQdWJsaXNoZXIy MTk4NDMw