The Impact of a Statewide Regionalization Policy on Travel Distance for Initial Breast Cancer Surgery

Academic Background

Breast cancer is one of the most common cancers among women globally, and the quality of treatment and patient survival rates are closely related to the allocation of medical resources. Studies have shown that high-volume medical institutions demonstrate better treatment outcomes in breast cancer surgery, particularly in significantly improving five-year survival rates. Based on this evidence, many regions and countries have begun implementing regionalization policies for medical services, concentrating complex surgeries in high-volume medical institutions to enhance treatment outcomes. However, while regionalization policies improve treatment efficacy, they may also negatively impact patient access to care, especially for low-income groups and patients living in remote areas.

In 2009, New York State (NYS) implemented a regionalization policy that restricted reimbursement for breast cancer surgery at low-volume hospitals for low-income patients (Medicaid beneficiaries). The goal of this policy was to improve patient outcomes by concentrating breast cancer surgeries in high-volume medical institutions. However, whether this policy would require patients to travel longer distances to access surgical care, thereby increasing their travel burden, has not been thoroughly studied. Therefore, this paper aims to explore the impact of this policy on travel distance for Medicaid and non-Medicaid patients, particularly for patients in different residential areas.

Source of the Paper

This paper was co-authored by Nina A. Bickell, Ann B. Nattinger, Emily L. McGinley, Maria J. Schymura, Purushottam W. Laud, and Liliana E. Pezzin, affiliated with institutions such as the Icahn School of Medicine at Mount Sinai, the Medical College of Wisconsin, and the New York State Department of Health. The paper was published on September 30, 2024, in the Journal of Clinical Oncology, with the DOI: https://doi.org/10.1200/jco.23.02638.

Research Process

Study Population

The study data were derived from a linked dataset combining the New York State Cancer Registry (NYSCR) with hospital discharge data. The study population consisted of women under 65 years old diagnosed with stage I-III breast cancer between 2004-2008 (pre-policy) and 2010-2013 (post-policy). The study excluded patients aged 65 and older, Medicare beneficiaries, those without surgical records, those whose residential addresses could not be mapped to geographic areas (RUCA classification), and those with missing race/ethnicity information. Ultimately, the study included 46,029 women, of whom 13.5% were Medicaid beneficiaries and 86.5% had other insurance.

Primary Study Measures

The primary measure of the study was the distance (in miles) from the patient’s residence to the surgical hospital. The distance was calculated based on the network distance from the population-weighted centroid of the residential census block group to the surgical hospital’s address, using GIS software. Additionally, the study collected information on patients’ insurance status, race/ethnicity, age, tumor stage, hormone receptor status, and comorbidities.

Statistical Analysis Methods

The study employed a multivariable difference-in-difference-in-differences (D-in-D-in-D) model to estimate the impact of the regionalization policy on travel distance for Medicaid and non-Medicaid patients. The model controlled for pre- and post-policy periods, insurance status, residential areas (New York City, other large urban areas, suburban/large towns, small towns/rural areas), and interactions between these factors. Additionally, the model controlled for patients’ age, race, tumor stage, hormone receptor status, and number of comorbidities.

Main Results

Sample Characteristics

Among the 46,029 patients, Medicaid beneficiaries were more likely to reside in large urban areas, particularly in New York City (65%). In contrast, only 32.9% of patients with other insurance lived in New York City. Medicaid beneficiaries tended to be younger, had a higher proportion of non-white individuals, a greater comorbidity burden, and more advanced tumor stages.

Changes in Travel Distance

Regardless of insurance status, all patients experienced an increase in travel distance post-policy. Medicaid beneficiaries consistently traveled shorter distances than patients with other insurance. Although all patients traveled longer distances post-policy, the increase was not significantly different between Medicaid and non-Medicaid patients, except for Medicaid patients living in suburban areas, whose travel distance increased significantly (from 14.4 miles to 22.1 miles, p=0.007).

Estimation of Policy Impact

By retransforming the model coefficients, the study calculated the predicted travel distances for Medicaid and other insurance patients in different residential areas before and after the policy. The results showed that the increase in travel distance was smaller for residents of New York City (Medicaid patients: from 5.5 miles to 6.4 miles; other insurance patients: from 7.3 miles to 8.3 miles), while the increase was significant for Medicaid patients in suburban areas (from 14.4 miles to 22.1 miles).

Conclusion

The study findings indicate that while NYS’s regionalization policy improved breast cancer treatment outcomes, its impact on patient travel distance was minimal. Although all patients traveled longer distances post-policy, the increase was not significantly different between Medicaid and non-Medicaid patients, except for Medicaid patients in suburban areas. Overall, the regionalization policy improved treatment outcomes without significantly reducing patient access to care.

Research Highlights

1. Quantification of Policy Impact: The study is the first to quantify the impact of NYS’s regionalization policy on travel distance for breast cancer patients, particularly for low-income patients.
2. Revelation of Regional Differences: The study found that Medicaid patients in suburban areas experienced a significant increase in travel distance post-policy, providing important insights for policymakers.
3. Application of Multivariable Models: The study employed a D-in-D-in-D model, effectively controlling for temporal trends and regional differences, enhancing the reliability of the results.

Research Significance

This study provides important empirical evidence for the implementation of regionalization policies, demonstrating that while improving treatment outcomes, the policy’s impact on patient travel distance is minimal. This serves as a reference for other regions considering similar regionalization policies, particularly in balancing treatment efficacy and patient access. Additionally, the findings offer policymakers differentiated recommendations for patients in different residential areas to mitigate potential negative impacts on low-income groups.

Other Valuable Information

The study also found that non-white patients and those with more comorbidities traveled shorter distances, possibly due to transportation barriers faced by these patients. Furthermore, the study excluded data from the COVID-19 pandemic, so future research could further explore the impact of the pandemic on patient travel distance.

This study provides significant scientific evidence for the implementation and evaluation of regionalization policies, holding high academic and policy application value.