Universal free school meals and Medicaid spending: evidence from linked education and claims data in Arkansas.

INTRODUCTION: Universal free meal (UFM) policies provide school breakfast and lunch at no cost to all students, increasing participation, improving dietary quality, and reducing household food insecurity. Whether these policies affect children's health care spending is unknown. METHODS: Using linked Arkansas education and Medicaid claims data (2013-2020) and a stacked difference-in-differences design across 4 UFM adoption cohorts, we examined changes in Medicaid spending among 131 851 continuously enrolled Medicaid children before and after school-level UFM adoption. RESULTS: The aggregate effect across the 2 post-adoption years was not statistically significant (spending ratio 0.973; 95% CI 0.932-1.016). However, spending declined significantly in the second post-adoption year (spending ratio 0.937; 95% CI 0.889-0.987), a 6.3% reduction. Reductions were concentrated in inpatient and pharmacy spending for mental, behavioral, and neurodevelopmental disorders, alongside increased outpatient spending in these categories. CONCLUSIONS: UFM adoption was associated with a delayed but significant reduction in Medicaid spending, driven largely by shifts in mental and behavioral health care utilization. These findings suggest school nutrition policy may influence health care spending among publicly insured children.