Author ORCID Identifier
https://orcid.org/0000-0002-2504-1437
Date of Award
Spring 6-2026
Document Type
Thesis (Ph.D.)
Department or Program
Health Policy and Clinical Practice
First Advisor
Douglas O. Staiger
Second Advisor
Anna N.A. Tosteson
Abstract
Statement of the Problem
Hospital closures have become increasingly common in the United States, raising concerns about access to and outcomes of hospital-based care. Prior research shows that closures can disrupt access and worsen outcomes, particularly for time-sensitive conditions. However, evidence is limited for non-urgent procedures, such as cancer surgery. This thesis examines how hospital closures affect cancer surgical quality, access, outcomes, and patient hospital choice.
Methods
Using Medicare administrative data, I identified hospital closures and Medicare patients diagnosed with or undergoing colon or lung cancer resection in 2008–2019. First, I conducted descriptive analyses to understand the extent of cancer surgical hospital closures and associations between undergoing surgery at a closing hospital and patients’ travel and post-operative outcomes. Second, I estimated heterogeneous effects of hospital closures on colon cancer surgical quality, access, and outcomes using a difference-in-difference framework. Finally, I modeled patients’ choice of surgical hospitals using a nested logit model to examine patient reallocation and simulate counterfactual scenarios of closing low-volume hospitals and improving low-performing hospitals.
Results
From 2008–2019, 267 (6.7%) hospitals performing colon cancer surgery and 108 (4.9%) hospitals performing lung cancer surgery closed. Among 558,087 Medicare patients undergoing surgery (64.5% colon, 35.5% lung), approximately 1% received care at hospitals that subsequently closed. Undergoing surgery at closing hospitals was significantly associated with higher post-operative mortality and complications. Difference-in-differences analyses showed that closure effects were concentrated in areas where closing hospitals had high market shares. Closure of higher-quality hospitals shifted patients to relatively lower-quality hospitals and worsened post-operative outcomes, whereas closure of lower-quality hospitals had the opposite effect. Hospital choice modeling indicated that higher exposure to closure was associated with higher probability of forgoing surgery or traveling farther to better quality hospitals. Closing low-volume hospitals would lead to large distance increases with modest quality improvements, while improving low-performing hospitals would induce reallocation away from high-quality hospitals to mid-quality hospitals.
Discussion
Hospital closures generate complex trade-offs between quality, access, and outcomes. Closure effects are highly heterogeneous, highlighting the importance of targeting investments into areas most vulnerable to adverse effects of closures on patient access and outcomes. Policies aimed at improving quality should carefully weigh intended gains against unintended consequences across patients.
Original Citation
Kim MY, Staiger DO, Brooks GA, Wang Q, Wong SL, Tosteson ANA. Outcomes Among Medicare Beneficiaries After Cancer Surgery in Hospitals That Subsequently Closed. JAMA Network Open. 2026;9(1):e2553704. doi:10.1001/jamanetworkopen.2025.53704
Recommended Citation
Kim, Min-Young, "HOSPITAL CLOSURES AND CANCER SURGERY IN THE UNITED STATES: EVIDENCE FROM COLON AND LUNG CANCER" (2026). Dartmouth College Ph.D Dissertations. 518.
https://digitalcommons.dartmouth.edu/dissertations/518
Included in
Health Economics Commons, Health Policy Commons, Health Services Research Commons, Population Health Commons
