Do Nursing Home Closures Increase Mortality Among Displaced Residents?

Olenski (2025) examined whether nursing home closures harm displaced long-stay residents or improve their long-term outcomes by moving them to better facilities. He analyzed nationwide administrative data on all long-stay U.S. nursing home residents, including 1,104 nursing home closures, using a matched difference-in-differences design. He found that closures increased short-term mortality by 1.18 percentage points, a 16.3% increase over the baseline quarterly mortality rate. Older and frailer residents experienced the largest risks. However, survivors generally transferred to higher-quality nursing homes, and cumulative mortality was 1.16 percentage points lower after three years, suggesting potential long-term benefits for surviving residents.

Why This Article Was Selected for The Policy Scientist

The long-term care sector is becoming increasingly important as populations age and governments devote larger shares of public spending to nursing home care. Understanding how provider closures affect vulnerable patients has direct implications for health care access, quality, and the organization of publicly financed services. This topic is especially timely because nursing home closures have accelerated in many regions due to financial pressures and workforce shortages. Olenski’s study makes an important contribution by distinguishing the immediate risks of displacement from the potential long-term benefits of reallocation to higher-quality facilities, extending earlier research that primarily examined short-term effects. The nationwide administrative data are exceptionally strong and support broad generalizability to similar health care systems. The matched difference-in-differences design represents a rigorous causal inference approach and provides substantially stronger evidence than conventional multivariate regression.

Full Citation and Link to Article

Olenski, A. (2026). Disruption and displacement in health care: Evidence from nursing home closures. American Economic Journal: Economic Policy. Advance online publication. https://doi.org/10.1257/pol.20240393

Central Research Question

This study examines whether nursing home closures improve or worsen outcomes for residents who are forced to relocate. Specifically, Olenski asks whether the immediate disruption caused by displacement increases mortality and whether any long-term benefits arise because surviving residents transfer to higher-quality nursing homes. The paper also investigates whether these effects differ according to patient health, local market conditions, and the quality of receiving facilities. More broadly, the study addresses an important policy question in health economics: whether the exit of lower-quality providers ultimately improves the allocation of patients or whether the disruption associated with provider closures outweighs any gains from improved quality.

The question is increasingly important because the U.S. nursing home industry has experienced a prolonged decline. Thousands of facilities have closed over the past two decades as reimbursement rates have failed to keep pace with costs and alternative long-term care options have expanded. These closures have generated concerns about access to care, particularly in rural communities, yet little evidence has existed regarding their causal effects on the health of displaced residents.

Previous Literature

Previous research has shown that provider exits can affect patient outcomes, but most studies have focused on hospitals or physician practices rather than nursing homes. Earlier studies examining hospital closures generally reported reduced access and poorer health outcomes, although some found evidence that patients benefited when care shifted toward higher-quality providers. Similar findings have emerged in studies of physician retirement and practice closures.

The nursing home literature has been considerably more limited. Earlier research largely consisted of individual case studies examining one or two facility closures. Although several studies reported higher mortality following forced relocation, others found little effect or even modest improvements. These conflicting findings reflected important methodological limitations, including small samples, lack of appropriate comparison groups, and limited ability to examine long-term outcomes. Most previous studies focused only on the period immediately surrounding relocation and could not distinguish temporary disruption from longer-term changes in care quality.

Olenski extends this literature by examining more than 1,100 nursing home closures using nationwide administrative data and modern causal inference methods. The study also contributes to research on health care competition and resource allocation by evaluating whether patients benefit when low-quality providers exit the market and residents relocate to higher-quality facilities.

Data

The analysis relies on several comprehensive administrative data sources from the Centers for Medicare & Medicaid Services. The primary dataset is the Minimum Data Set (MDS), which contains quarterly clinical assessments for essentially every resident of Medicare- and Medicaid-certified nursing homes between 2000 and 2017. These assessments provide detailed information on patient demographics, health conditions, functional status, and nursing home residence.

The author links these records to Medicare enrollment and claims data, allowing patients to be followed after transfer to another nursing home. This linkage makes it possible to observe mortality, hospitalizations, residential movement, and other health outcomes regardless of where patients relocate. The study also incorporates publicly available information on nursing home characteristics, including staffing levels, ownership, occupancy, payer mix, quality deficiency citations, and facility closures.

The final sample includes 1,104 nursing home closures involving approximately 42,900 displaced long-stay residents and more than 208,000 residents in matched comparison facilities. Because the data encompass nearly the entire population of long-stay Medicare nursing home residents over many years, they provide exceptional statistical power and broad external validity.

Methods

The study employs a matched difference-in-differences research design, a widely accepted causal inference approach. Each closing nursing home is matched with as many as four observationally similar facilities that remained open. Matching is based on characteristics such as occupancy, ownership status, Medicaid share, bed capacity, staffing, market concentration, and county population.

The author then compares changes in mortality before and after closure between residents of closing facilities and residents of matched control facilities. This design attempts to estimate the counterfactual outcome—that is, what would have happened to displaced residents had their nursing home remained open. Facility fixed effects, matched cohort-by-quarter fixed effects, and extensive patient-level controls further reduce potential sources of bias.

The analysis also estimates cumulative mortality over time rather than focusing solely on quarterly death rates. This distinction is important because an initial increase in mortality changes the composition of surviving patients. Additional analyses examine heterogeneity by age, dementia status, ownership type, and local market competition. Finally, the study evaluates mechanisms by examining transfers to higher-quality facilities, travel distance, hospitalization rates, and preventable hospital admissions.

Although no observational study can completely eliminate concerns about unobserved confounding, the combination of nationwide administrative data, careful matching, event-study analyses, robustness tests, and difference-in-differences estimation provides a strong causal framework that substantially improves upon conventional multivariate regression.

Findings/Size Effects

The study finds substantial evidence that nursing home closures create significant short-term health risks. During the quarter of closure, displaced residents experience a 1.18 percentage point increase in mortality relative to comparable residents whose facilities remain open. Given a baseline quarterly mortality rate of 7.2%, this represents a 16.3% increase in mortality risk.

The disruption disproportionately affects vulnerable populations. Residents aged 80 years and older experience a 2.03 percentage point increase in mortality, while residents with Alzheimer’s disease or related dementia experience a 1.71 percentage point increase. Younger and healthier residents experience substantially smaller effects.

Despite these immediate harms, long-term outcomes differ. After the initial mortality spike, surviving residents increasingly benefit from relocation to higher-quality nursing homes. Three years after closure, cumulative mortality is estimated to be 1.16 percentage points lower than would have been expected had the original facilities remained open, although the precision of these estimates varies across specifications and patient groups.

The study identifies quality improvement as the primary mechanism underlying these long-term gains. Residents who relocate transfer to facilities with approximately 29.5% fewer quality-of-care deficiency citations than their original nursing homes. They also move to facilities with higher risk-adjusted survival rates. Patients experiencing the largest improvements in facility quality also experience the greatest long-term survival gains.

The results also demonstrate important differences across markets. Residents in competitive markets benefit most from quality-driven reallocation, whereas those living in less competitive areas experience the largest short-term mortality increases without corresponding long-term improvements. Hospitalization rates increase by 2.81 percentage points immediately after closure, including a 1.10 percentage point increase in preventable hospitalizations, suggesting that disruption itself contributes directly to poorer short-term health outcomes.

Conclusion

Olenski concludes that nursing home closures generate two competing effects. Forced relocation imposes substantial short-term health risks, particularly for older and medically fragile residents. At the same time, surviving residents frequently transfer to higher-quality facilities, producing evidence of improved long-term survival for some patient groups.

These findings illustrate that provider exits involve both disruption and reallocation. The results suggest that maintaining continuity of care is particularly important for clinically vulnerable residents, while quality improvements associated with relocating patients may benefit healthier survivors. The study also demonstrates that market structure influences the consequences of closures, with competitive markets providing greater opportunities for beneficial reallocation.

More broadly, the paper advances understanding of health care provider exits by integrating patient displacement, provider quality, and market competition within a unified causal framework. Its use of comprehensive national administrative data and rigorous difference-in-differences methods provides some of the strongest evidence to date on the consequences of nursing home closures and represents an important contribution to the health economics and public policy literature.