Medicaid Covers at Least One in Five Hospital Inpatient Days in Nearly Every State
The House and Senate are working on legislation to meet the requirements in the budget resolution, specifying cuts to Medicaid of up to $880 billion or more over 10 years. Large reductions in Medicaid spending are likely to have direct implications for the 83 million people covered by Medicaid, state budgets, and health care providers, including hospitals. Medicaid accounted for about one fifth (19%) of all spending on hospital care in 2023 and cuts to payments for care or loss of coverage could have implications for hospitals’ finances, the cost and quality of care, and people’s ability to access hospital services. There could be consequences for the broader economy too given that hospitals are the sixth largest employer in the country across industry subsectors.
To inform these discussions, this analysis describes the percent of inpatient hospital days that are covered by Medicaid (also referred to as the “Medicaid inpatient share”), nationally and by state. The analysis uses Medicare cost report data from 2023 (the most recent year available) and focuses on hospitals that are non-federal (see Methods for more details about the hospitals included and analysis).
Medicaid covered at least one in five inpatient hospital days in 48 states and the District of Columbia (hereafter referred to as a state) in 2023. Medicaid covered at least 25% of inpatient days in 30 states and at least 30% of days in 10 states (see Figure 1). The Medicaid share ranged from 11% in Wyoming to 37% in New Mexico. Medicaid covered about one in four (26% of) inpatient days nationally. Variation across states is driven by Medicaid eligibility levels—including whether a state has opted to expand under the Affordability Care Act—as well as demographics. Medicaid’s share of hospital use is likely higher than its share of hospital revenues in part because Medicaid payment rates are generally lower than what commercial insurers pay.
States with the highest Medicaid shares included a mixture of red and blue states. For example, among the 10 states with Medicaid shares of at least 30%, five were states that voted for President Trump in the 2024 election (Alaska, Kentucky, Louisiana, Nevada, and Oklahoma) and five were states that voted for Vice President Harris (California, Colorado, DC, New Mexico, and New York).
Medicaid covered about four in ten (41% of) births nationally in 2023, almost half (47%) of births in rural areas, and at least 40% of births in 26 states according to prior KFF analysis. The vast majority of births occur in hospitals.
With hospital care accounting for about one third of Medicaid spending in 2023, large Medicaid cuts would be likely to affect hospitals. Some policy options under discussion would affect hospitals directly by reducing the payments made to hospitals through managed care organizations (by limiting what are known as state directed payments) or restricting states’ ability to fund Medicaid through provider taxes, which often support higher payments for hospitals. Also being considered are options to reduce federal spending on the Affordable Care Act (ACA) Medicaid expansion, which has helped improve hospital finances and may have especially benefited rural hospitals. The hospital industry has been lobbying Congress against proposed cuts, arguing that reductions in Medicaid spending would threaten access to hospital care for all patients—not just Medicaid beneficiaries—and the National Rural Health Association has argued that Medicaid spending reductions would lead rural hospitals to reduce or eliminate the services they offer or close altogether.
Also on the horizon is the expiration of the enhanced Affordable Care Act (ACA) subsidies in 2026; there would likely be implications for hospitals if the subsidies are allowed to expire, as the number of uninsured people would increase by 3.8 million per year on average according to Congressional Budget Office (CBO) projections.
This work was supported in part by Arnold Ventures. KFF maintains full editorial control over all of its policy analysis, polling, and journalism activities.
Methods |
| Data and Hospital Inclusion Criteria: This analysis relied on the RAND Hospital Data, a cleaned and processed version of the annual cost reports that Medicare-certified hospitals are required to submit to the federal government. The analysis included non-federal general short-term hospitals in the 50 states and the District of Columbia. The analysis excluded 47 hospitals that did not report total inpatient days, reported negative total Medicaid or inpatient days, or had a Medicaid share greater than 100%. These hospitals together represented 1% of non-federal general short-term hospitals in the 50 states and the District of Columbia (ranging from 0% in several states to 7% in Connecticut). The final sample included 4,400 hospitals.
We conducted a sensitivity analysis to assess how restricting to hospitals reimbursed under the inpatient prospective system (IPPS) would affect the results and found that it would have changed the results in each state by at most, one percentage point. Most IPPS hospitals directly report their Medicaid inpatient share, which is used for determining eligibility for Medicare disproportionate share hospital (DSH) status and the amount of DSH payments. Calculation of Medicaid Shares: The analysis used the Medicaid inpatient share reported by hospitals to the federal government for the purposes of establishing Medicare DSH status and determining the amount of DSH payments for the 63% of hospitals in sample that reported that measure. For the 37% of hospitals that did not directly report their shares, the share was calculated from other lines in the cost reports. Medicaid days (the numerator) include Medicaid paid days and Medicaid eligible unpaid days. State shares reflect hospital shares weighted by the denominator (total inpatient days). When calculating shares or inpatient days, lines that were blank were treated as 0, including some instances where hospitals did not report any Medicaid days. Some hospitals had cost report periods that were less than or greater than 365 days, in which case days were scaled up or down to reflect a full year. Hospital data were sorted into fiscal year 2023 based on the mid-point of the reporting period. Limitations: Cost report data are reported by hospitals and likely have some degree of inaccuracy. For example, it is possible that hospitals that are not part of the Medicare DSH program—which account for 41% of the hospitals in the sample and 7% of inpatient days—may have less incentive to fully account for Medicaid days, given that they do not receive Medicare DSH payments. In that case, the estimates in this analysis would be conservative. It is also possible that the data are inaccurate in other ways that could lead to this analysis overreporting shares, and reporting issues could vary by state. |