2 37 REVIEW OF APPLIED & CASUAL GAMES FOR MENTAL HEALTH 4. Outcome: Changes in mental health outcomes were studied and/or a (sub)clinical population was used (i.e., participants were either currently diagnosed with a mental health disorder, experienced elevated mental health symptoms, or had a past diagnosis). Mental health was defined as including internalising (e.g., anxiety, depression, mood, stress), externalising (e.g., aggression, delinquency, conduct-related, oppositional defiant, disruptive), neurodevelopmental (e.g., autism, ADHD), psychotic or personality-related outcomes, symptoms or disorders. Studies were also included when positive mental health outcomes were measured (e.g., happiness, relaxation) or when physiological markers were measured as an outcome (e.g., heartrate, cortisol). Populations and outcomes related to, for instance, addictive or eating disorders, dementia, and stroke were excluded. 5. Study design: The study employed an RCT design that adhered to the following three components (Bhide et al., 2018): use of only one single population, eligible participants have an equal chance to be allocated to each condition (i.e., (pseudo) random allocation), and all participants are treated the same with the single and only exception of the actual intervention or control condition (i.e., equal conduct). Both fully randomised and pseudo-randomised controlled trials were included, as well as papers that reported on the same dataset (e.g., primary and secondary outcome papers), on aggregated RCT results or results from one RCT-arm (e.g., secondary analyses). Studies that did not include a comparison condition (e.g., all participants played the same game) were excluded. Editorials, letters to the editor, book chapters, unpublished reports, studies that included no original data (e.g., reviews), study protocols and RCT-studies that did not report on the effectiveness of the intervention were also excluded. Review Process and Procedure Screening of the titles and abstracts was done by two independent reviewers (i.e., the first author and either a research assistant or the second author) and a record was kept regarding the reasons for exclusion, checking all four eligibility criteria. After every ± 250 references a reliability meeting was done, in which disagreements about in- or exclusion were resolved. For the original search, the mean percent agreement (kappa statistic) for eligibility was 95.8% (k = 0.76), with the range over all seven reliability meetings being 95.1–96.4% agreement and k being between 0.73 and 0.78. Due to slight refinement of the eligibility criterion for games (i.e., excluding non-digital games and games used as an assessment or manipulation tool), a second screening round was performed (percent agreement for eligibility 95.9% and k = 0.87). For the first
RkJQdWJsaXNoZXIy MTk4NDMw