A County-level Analysis of the Association Between Medicaid Expansion and Medicare Spending

Abstract

Background: If access to Medicaid improves health outcomes, it may also result in lower long-term spending, however the association between Medicaid expansion and Medicare spending is unknown. In this analysis we sought to investigate the association between Medicaid expansion and per capita Medicare spending at the county level. Methods: This is an observational analysis of all U.S. counties in the ten years from 2010 to 2019. We used a difference-in-difference event study to investigate the difference in per capita Medicare spending between counties in states that expanded Medicaid and counties in states that did not expand Medicaid. The exposure was treatment year, which characterized whether a county was in an expansion state and when expansion occurred. In non-expansion counties, treatment year was assigned 0 for all observations. In expansion counties, treatment year ranged from -3 to +6, with treatment year 1 corresponding to the first full year of expansion. The primary outcome was fee-for-service Medicare spending per capita in each county. A secondary analysis investigated subcategories of per capita spending including inpatient, outpatient, skilled nursing care, inpatient rehabilitation, home health, and hospice care. Results: We analyzed 1,648 expansion and 1,494 non-expansion counties, with ten observations per county, one for each year between 2010 and 2019. In the adjusted event study analysis, the difference between expansion and non-expansion counties in expansion year 5 compared to pre-expansion was -200 [95% Confidence Interval (CI): -406, 6] dollars. In the subcategory analysis, the difference in inpatient care, skilled nursing care, outpatient care, and home health spending were -46 [95% CI: -103, 12], -92 [95% CI: -194, 11], 57 [95% CI: -67, 181], and 55 [95% CI: -17, 126] dollars per capita respectively. Conclusions: Medicaid expansion is not consistently significantly associated with lower Medicare spending compared to pre-expansion. However, observed trends towards lower spending and cost-shifting from inpatient to outpatient settings in expansion counties warrant additional investigation.