Machine learning estimates SSI eligibility for Medicaid children, revealing substantial unclaimed benefits.
Though safety net programs offer important benefits, take-up is often incomplete. Using machine learning models on Medicaid data, we estimate the take-up rate for children’s Supplemental Security Income (SSI) and the characteristics of potentially eligible children with disabilities who do not receive benefits. Using more than 1,000 measures of health-care utilization, we estimate state-specific models that generate the probability that each child not receiving SSI is eligible. Using the expected number of potentially eligible children, the implied take-up of SSI is approximately 70 percent. Potentially eligible children have intensive health-care usage, often more intensive than current child SSI recipients.
No takes yet. Share an insight, caveat, or question.
Levere et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: