Efficacy of User Self-Led and Human-Supported Digital Health Interventions for People With Schizophrenia: A Systematic Review and Meta-Analysis

Background Digital health interventions (DHIs) may enable low cost, scalable improvements in the quality of care for adults with schizophrenia. Given the fast-growing number of studies using these tools, this review aimed to assess the efficacy and feasibility of randomized controlled trials (RCTs) of DHIs among people with schizophrenia, focusing on human support. Design A systematic search of PubMed, Embase, PsycINFO, CINAHL, Web of Science, and Cochrane databases was conducted in January 2024 to identify relevant RCTs. Random effects meta-analyses were undertaken to evaluate the effects on psychosis symptoms, cognition, and other health-related outcomes. Results Twenty-six RCTs (n = 2481 participants) were included. Pooled recruitment and retention rates were 57.4% and 87.6%, respectively. DHIs showed no statistically significant effect sizes across all examined outcomes, including psychosis symptoms, depression, quality of life, global and social cognition, global and social functioning, and medication adherence. A third (34.6%) of DHIs were developed using co-design while the majority (70%) personalized the intervention to its users and 69.2% were deemed to be at high risk of bias. There were trends toward effects in favor of DHIs with human support for social cognition and quality of life. Conclusions DHIs are feasible for people with schizophrenia and potentially useful for improving health outcomes, particularly when including human support. More high-quality studies are required to examine the benefits of human support within DHIs. Future research should examine the feasibility of sustained adherence and benefits from digital interventions, possibly incorporating human interaction complemented with artificial intelligence, in real-world clinical settings.