Recent scholarship has emphasised the importance of social interventions in mental health, broadly defined as interventions designed to deliver social benefits that drive positive mental health consequences to their beneficiaries. While more is understood about the impact of some types (i.e., social-relational interventions) globally, impacts of social interventions addressing broader systemic and material conditions linked to mental health, such as poverty and access to welfare systems, are less understood. Addressing this evidence gap is critical for advancing a holistic agenda in the global mental health field that reflects the social realities and mental health needs of communities. We conducted a systematic review in accordance with PRISMA guidelines to advance understandings of social interventions and their role in addressing mental ill-health in low resource settings. We focused on interventions targeting conditions within the WHO MHGAP (version 2.0) guidelines, delivered in LMIC countries defined by World Bank criteria. Our protocol was registered with Prospero (CRD42022357660). Searches were conducted in October 2022 and updated in June 2025. 22,553 records were identified and independently double-screened, with 83 studies meeting inclusion criteria. We structured our discussion of findings, by level of intervention delivery: Individual and family level; Household and working life level; community level; and population level. We report on findings from 36 countries, with studies from the African continent being most represented. Meta-analysis of our most commonly explored mental health condition (depression) identified a moderate effect size (Cohen’s d = 0.42, P = 0.005) for social interventions, with a high level of heterogeneity. Service level interventions (primarily school-based), were most common, followed by community-level interventions. Overall, our findings indicate that the primary level of delivery did not preclude actions and impact across other levels, as many social interventions were flexible in responding directly to local needs. Implications for intervention programming in LMIC landscapes are discussed.
Ionescu et al. (Wed,) studied this question.