59 Review of social network intervention studies 2 DISCUSSION This is the first comprehensive meta-analysis, including 29 randomized controlled trials and four quasi-experimental designs, quantifying the effectiveness of social network interventions in psychiatric patients on a broad range of outcomes. Main findings support modest additive effects of social network interventions on patients’ positive social networks and support, general functioning, mental health treatment adherence, days substance use, and abstinence. No superior effects of social network interventions were found for psychiatric and psychological symptom reduction. If groups of psychiatric patients consisted of more female patients, psychiatric symptoms reduced more in those who participated in the social network intervention compared to control groups. In addition, effects on general functioning were larger at follow-up assessment, whereas effects on mental health treatment adherence were larger at post-assessment. These findings suggest that a focus on social network enhancement in addition to mental health treatment, might cause sustainable improvement of general functioning in psychiatric patients. Besides, patients might be more satisfied and adhere better during the course of treatment. Therefore, the addition of social network interventions could actively support mental health treatment. Our findings are consistent with previous reviews that provided modest evidence of comparable social network interventions among psychiatric populations (Anderson et al., 2015; Barnett et al., 2022; Brooks et al., 2022; Killaspy et al., 2022; Ma et al., 2020; Pinto, 2006; Shorey & Chua, 2022; Siette et al., 2017; Smit et al., 2022; Webber & Fendt-Newlin, 2017). Although modest, these effects are considered meaningful, as effect sizes in studies examining the additive effects of experimental interventions in clinical populations are expected to be smaller (Götz, Gosling, & Rentfrow, 2022; Kraft, 2020). In contrast to our expectations, and one previous review (Mead et al., 2010), no evidence of psychiatric and psychological symptom reduction in psychiatric populations was found. Conflicting findings may be explained by the differences in study population. Less symptom reduction could be expected in difficult-to-treat psychiatric populations included in our study. Furthermore, no clear benefits were found on criminal behavior as well as the vast majority of specific outcome measures. This could be explained by the small number of studies that examined these specific outcomes in psychiatric populations, making it impossible to draw firm conclusions at this point (Deeks, Higgins, Altman, & Cochrane Statistical Methods Group, 2019; Siette et al., 2017). To our knowledge, no prior studies demonstrated subgroup effects (i.e., sex and assessment time) of social network interventions on treatment outcomes.
RkJQdWJsaXNoZXIy MjY0ODMw