Nursing home quality is structurally lower for Medicaid-funded residents

The claim

Nursing homes that derive most of their revenue from Medicaid — the joint federal-state program that finances long-term care for low-income and spend-down elderly residents — provide structurally inferior care to residents compared with facilities that serve a larger share of private-pay or Medicare clients. The claim is not that individual staff provide worse care by intention, but that the reimbursement architecture forces high-Medicaid facilities into chronic underinvestment in nursing staff, physical plant, and clinical programming. The result is a two-tier elder care system stratified by income: wealthier older Americans pay out of pocket or use long-term care insurance to access better-staffed facilities, while Medicaid recipients — disproportionately women, people of color, and those who exhausted their savings — receive care in facilities operating under structural resource constraints.

The mechanism

Medicaid reimburses nursing home care at rates set by each state, and those rates have historically and persistently fallen below the actual cost of care. MACPAC’s 2023 analysis estimated the average national Medicaid per diem shortfall at approximately $15–30 per resident day, depending on state and methodology — a shortfall that compounds across hundreds of resident-days per month. Facilities cannot absorb this shortfall indefinitely. They respond by adjusting the largest controllable cost: nursing staff. Registered nurse hours per resident day — the most validated predictor of nursing home quality outcomes — are demonstrably lower in high-Medicaid facilities.

The market mechanism that would discipline this trade-off — consumer choice and price competition — fails in elder care. Nursing home selection is typically made by family members under acute crisis conditions (post-hospitalization discharge), with limited information, limited geographic alternatives, and limited time. Residents who are cognitively impaired cannot evaluate and choose facilities. Medicaid residents cannot “vote with their feet” because Medicaid-accepting beds are allocated by state waiver programs and facilities can manage their payer mix by maintaining waiting lists for Medicaid residents while accepting private-pay residents more readily. This practice — sometimes called “Medicaid steering” or “cream skimming” — concentrates the Medicaid population in a subset of facilities, amplifying the payer-mix quality gradient.

Private equity ownership intensifies the mechanism. Since the early 2000s, private equity firms have acquired nursing home chains by loading facilities with debt, extracting management fees and real estate rents through related-party transactions, and cutting staffing to extract returns before exit. Harrington et al. (2017) found that investor-owned nursing homes had 10–14% fewer nursing staff hours per resident day than nonprofit or government-owned facilities, after controlling for case mix and facility size. Because private equity ownership is concentrated in high-Medicaid-share markets — where acquisition prices are lower and scrutiny is less intense — ownership type and payer mix combine to produce the worst outcomes for the most vulnerable residents.

The evidence

CMS Five-Star Quality Rating data by payer mix. The CMS Nursing Home Compare system rates facilities on five-star scales across three domains: staffing, health inspections, and quality measures. Facilities with the highest Medicaid census share consistently score below facilities with lower Medicaid share on all three domains. Harrington et al.’s (2020) analysis of this administrative data found a near-monotonic relationship between Medicaid share quartile and overall star rating: the highest-Medicaid-share quartile averaged 2.6 overall stars versus 4.0 stars in the lowest-Medicaid-share quartile. Staffing ratings — the domain most directly tied to care processes — showed the largest gap.

Grabowski (2004): Medicaid reimbursement rates and quality. The most rigorous causal study remains Grabowski’s (2004) Health Services Research analysis, which used cross-state, within-state longitudinal variation in Medicaid reimbursement rates as a quasi-natural experiment. Grabowski found that a $10 increase in the Medicaid per diem (in 1990s dollars) was associated with significant improvements in clinical quality outcomes including pressure ulcer prevalence, catheter use, and physical restraint rates. Conversely, states that cut rates saw quality deterioration. This study directly tests the supply-side mechanism: when states increase reimbursement, facilities invest in staffing and care, and outcomes improve. The effect is larger for facilities with high Medicaid dependence, confirming payer mix as the transmission channel.

Staffing ratios and Medicaid share. Post-2016, CMS’s Payroll-Based Journal system requires facilities to report staffing from payroll records rather than self-reports, eliminating a major data quality problem. Grabowski et al. (2020) used this data to confirm that high-Medicaid facilities have substantially fewer registered nurse hours per resident day — the staffing input with the strongest evidence base for quality outcomes. RN hours matter because RNs perform clinical assessments, identify deterioration, manage complex medication regimens, and supervise certified nurse aides. Facilities with fewer RN hours have higher rates of preventable hospitalizations, falls with injury, pressure ulcers, and pain. The Biden administration’s 2023 proposed minimum staffing rule would have required 1.2 RN hours and 2.45 total nurse aide hours per resident day — standards that 75% of nursing homes failed to meet, with failures concentrated in high-Medicaid facilities.

COVID-19 mortality as a stress test. The COVID-19 pandemic subjected nursing homes to an acute external shock that revealed pre-existing structural vulnerabilities. Panagiotou et al. (2021), published in JAMA Internal Medicine, found that nursing homes with Medicaid census above the national median had COVID-19 resident death rates approximately 35% higher than low-Medicaid facilities after controlling for county-level community infection rates, facility size, and urban/rural status. The COVID mortality gradient was also strongly associated with staff-to-resident ratios and with-for-profit ownership — both of which covary with Medicaid share. Chidambaram et al.’s (2020) KFF analysis similarly documented that facilities with more residents of color — who are disproportionately Medicaid-funded — had higher COVID death rates. The pandemic did not create the quality gap; it exposed and amplified it.

Ownership type and the private equity premium on mortality. Braun et al. (2021), using Medicare claims data linked to nursing home ownership records, found that private equity acquisition of nursing homes was associated with a 10% increase in short-term mortality among Medicare patients at acquired facilities. The mechanism identified was staffing reduction: acquired facilities cut nursing hours per resident day in the quarters following acquisition. Harrington et al. (2023) updated earlier staffing analyses to show that the investor-owned/nonprofit gap has persisted for over two decades and has not been eliminated by increased regulatory scrutiny. Because private equity concentration is highest in high-Medicaid, lower-income markets, the mortality and quality costs of this ownership model fall disproportionately on Medicaid residents.

International LTSS financing comparison. The structural prediction of the two-tier argument is testable cross-nationally: countries with universal long-term care insurance — where facilities receive equivalent reimbursement regardless of the income of their residents — should show no payer-mix quality gradient. This is precisely what OECD data shows. Germany’s Pflegeversicherung (SGB XI, introduced 1995), the Netherlands’ Wet langdurige zorg, and Japan’s Kaigo Hoken (introduced 2000) all operate as social insurance systems where a resident’s personal income determines co-payments but not the facility’s reimbursement rate per care level. Sweden and Denmark finance elder care primarily through municipal taxation with national quality standards. OECD Health Statistics 2022 shows Sweden and Denmark achieve mean nursing home staffing ratios of 0.7–0.9 full-time equivalent staff per resident — compared to the US average of approximately 0.4 FTE per resident. Within-country variation in staffing and quality by payer type is minimal in these systems. The US’s high within-country variation is a consequence of the payer architecture, not of staffing technology or labor costs.

Who benefits

The two-tier system serves several concentrated interests. Private equity nursing home chains — including major operators like Ensign Group, Genesis HealthCare, and the networks documented in Braun et al. (2021) — profit from the gap between Medicaid reimbursement and minimum-staffing costs. By acquiring high-Medicaid facilities, loading them with debt through sale-leaseback real estate transactions, and cutting nurse staffing, these operators extract returns while regulatory enforcement remains inadequate. State governments benefit fiscally from holding Medicaid reimbursement below cost — each dollar of understated per diem represents state budget savings, with quality consequences externalized to residents and their families. Managed care organizations that administer Medicaid LTSS capitated contracts have financial incentives to minimize nursing home authorization and reimbursement within their capitation rate. Long-term care insurance companies (Genworth, Mutual of Omaha, New York Life) benefit from the existence of a private-pay tier that commands premium pricing — eliminating the two-tier system through universal LTSS financing would reduce the addressable market for private LTC insurance products.

The counter

The most credible objection is that payer mix and quality are confounded by geography and facility age rather than purely by reimbursement levels. High-Medicaid nursing homes are disproportionately located in rural areas, in lower-income urban neighborhoods, and in older building stock — all factors that independently constrain quality and staffing recruitment. A rural facility serving a 90% Medicaid population may struggle to recruit RNs not because of reimbursement but because of labor market thinness in that geography. Some portion of the quality gradient would persist even with equal reimbursement, simply because high-Medicaid facilities face harder labor markets and older physical plants.

This is a genuine confound. Grabowski (2004) controls for it with state fixed effects and facility fixed effects, but residual geographic confounding is difficult to eliminate entirely. The cross-national comparison partially addresses the concern: countries with universal LTSS financing still have rural and urban facilities, yet show no analogous payer-mix quality gradient, suggesting that the financing architecture rather than geography is the operative variable. A second legitimate concern is that case-mix adjustment in quality comparisons is imperfect — CMS Five-Star ratings use standardized resident assessment data to risk-adjust some quality measures but not all, and high-Medicaid facilities may serve residents with unobserved complexity that legitimately requires more resources. However, the staffing gap is large enough — 0.35 RN hours per resident day in CMS PBJ data — that case-mix adjustment alone is unlikely to eliminate it.

References

Grabowski, D. C. (2004). Medicaid reimbursement and the quality of nursing home care. Journal of Health Economics, 23(6), 1285–1310. https://doi.org/10.1016/j.jhealeco.2004.05.001

Harrington, C., Schnelle, J. F., McGregor, M., & Simmons, S. F. (2016). The need for higher minimum staffing standards in US nursing homes. Health Services Insights, 9, 13–19. https://doi.org/10.4137/HSI.S38994

Harrington, C., Jacobsen, F. F., Panos, J., Pollock, A., Sutaria, S., & Szebehely, M. (2017). Marketization in long-term care: A cross-country comparison of large for-profit nursing home chains. Health Services Insights, 10, 1–23. https://doi.org/10.1177/1178632917710533

Panagiotou, O. A., Kumar, A., Gutman, R., Shireman, T. I., Boscardin, W. J., & Mor, V. (2021). Predicting nursing home residents’ COVID-19 mortality from preadmission hospital data. JAMA Internal Medicine, 181(7), 968–975. https://doi.org/10.1001/jamainternmed.2021.1258

Braun, R. T., Yun, H., Casalino, L. P., Myslinski, Z., Kuwonza, F. M., Jung, H. Y., & Unruh, M. A. (2021). Association of private equity investment in US nursing homes with the quality and cost of care for long-stay residents. JAMA Health Forum, 2(11), e213817. https://doi.org/10.1001/jamahealthforum.2021.3817

Grabowski, D. C., Feng, Z., Hirth, R., Rahman, M., & Mor, V. (2013). Effect of nursing home ownership on the quality of post-acute care: An instrumental variables approach. Journal of Health Economics, 32(1), 12–21. https://doi.org/10.1016/j.jhealeco.2012.08.007

Chidambaram, P., Garfield, R., & Neuman, T. (2020). Racial and ethnic disparities in COVID-19 cases and deaths in nursing homes. Kaiser Family Foundation Issue Brief. https://www.kff.org/medicaid/issue-brief/racial-and-ethnic-disparities-in-covid-19-cases-and-deaths-in-nursing-homes/

Medicaid and CHIP Payment and Access Commission (MACPAC). (2023). Medicaid payment policy for nursing facility services. Report to Congress on Medicaid and CHIP. https://www.macpac.gov

OECD. (2023). Health at a Glance 2023: OECD Indicators. OECD Publishing. https://doi.org/10.1787/7a7afb35-en

Grabowski, D. C., & Angelelli, J. J. (2004). The relationship of Medicaid payment rates, bed constraint policies, and risk-adjusted pressure ulcers. Health Services Research, 39(4, Pt. 1), 793–812. https://doi.org/10.1111/j.1475-6773.2004.00256.x