The role of state social service spending in moderating mental health outcomes of community violence: A spatial meta-analysis

Community violence is an epidemic in the United States. Studies estimate almost 30% of youth experience community violence, with higher rates in low-income urban settings. Community violence exposure is consistently linked to adverse mental health outcomes, including post-traumatic stress, internalizing, and externalizing symptoms. In response, nationwide efforts have emphasized increased financial investment in social services as a potential means to mitigate the impact of violence exposure. Yet, gaps remain in understanding whether structural investments (i.e., social service expenditures) moderate mental health outcomes. This study aimed to: (1) provide an updated examination of the associations between community violence exposure and mental health outcomes, and (2) test whether state-level social service expenditures moderated these associations. We conducted a spatial meta-analysis using a three-level random-effects model, linking study data to publicly available Census expenditures data. A total of 155 studies representing 23 states yielded 769 effect sizes. Results indicated a significant small-to-moderate positive association between community violence and mental health (Fisher's z = 0.22, p < .0001; 95% CI [0.20, 0.24]), with heterogeneity across studies (Q = 9524.4, df = 768, p < .0001). Public welfare, hospital, and veterans' services spending weakened the positive associations with somatic, post-traumatic stress, and general stress outcomes. By contrast, public welfare, employment security, and health spending amplified the positive associations with internalizing and somatic outcomes. Findings underscore the enduring mental health burden of community violence, the potential role of targeted policy investments, and the need for cross-disciplinary research to clarify how psychological and structural factors impact mental health outcomes among those exposed to community violence.