Building a Smarter Safety-Net: Evidence of Offsets between the SNAP and Medicaid Programs

Project Summary

The research team aims to examine whether changes in SNAP eligibility impact patterns of health service use and health outcomes among Medicaid beneficiaries. To do this, the team will construct a novel data linkage of administrative Medicaid and SNAP datasets to evaluate two natural experiments that lead low-income families to acquire or lose SNAP benefits.

Research Question(s)

  1. Does gaining access to SNAP benefits impacts health outcomes, health care utilization, and Medicaid spending?
  2. What are the implications of changes in access to SNAP benefits on state budgets and health disparities?


  • Provide policymakers with comprehensive evidence on the state-level return to investments in SNAP; and
  • Directly inform policy proposals that threaten to reduce the generosity of the SNAP program


  • Intermediate Outcomes
    • Rates of new chronic conditions (e.g., Diabetes, Hypertension, Hypercholesterolemia)
    • Management of existing chronic conditions (e.g., Type 2 diabetes mellitus with ketoacidosis)
    • Percentage of Infants Born with Low-Birth Weight
    • Mortality
  • Health Care Utilization Outcomes
    • Overall Acute Care Utilization
    • Overall Emergency Department Use
    • Overall Admission Rate
    • Ambulatory Care Sensitive ED Visits and Admissions
  • Health Care Spending Outcomes
    • Overall Health Care Spending


The investigators will evaluate two natural experiments that lead low-income families to acquire or lose SNAP benefits, including: 1) using a regression discontinuity design to estimate the effect of acquiring SNAP on care and outcomes for the near-elderly; and 2) leveraging policy-imposed recipient enrollment limits to estimate the effect of the loss of SNAP benefits on intermediate health outcomes using a difference-in-differences approach with a treatment group comprised of Medicaid beneficiaries that lose access to their SNAP benefit due to programmatic term limits set by the states. Additionally, in a related set of policy simulations, the research team will examine the effects of SNAP eligibility on joint state spending across the SNAP and Medicaid programs collectively, as well as the impact of SNAP eligibility on population-level disparities in health using data from an All-Payer Claims Dataset.

Image of a two-armed scale with vegetables and fruits on the left hand side and colorful pills on the right hand side.
Grantee and Partner organizations

Yale University
State of Connecticut Department of Social Services

Grant status
In Progress
Principal investigators
Chima Ndumele, PhD
Start date
Award amount
30 months

Stay Connected