Policy Recommendations for Improving Rural Cancer Services in the United States

Abstract

Compared with urban residents, rural Americans have seen slower declines in cancer deaths, have lower incidence but higher death rates from cancers that can be prevented through screening, have lower screening rates, are more likely to present with later-stage cancers, and have poorer cancer outcomes and lower survival. Rural health provider shortages and lack of cancer services may explain some disparities. The literature was reviewed to identify factors contributing to rural health care capacity shortages and propose policy recommendations for improving rural cancer care. Uncompensated care, unfavorable payer mix, and low patient volume impede rural physician recruitment and retainment. Students from rural areas are more likely to practice there but are less likely to attend medical school because of lower graduation rates, grades, and Medical College Admission Test (MCAT) scores versus urban students. The cancer care infrastructure is costly and financially challenging in rural areas with high proportions of uninsured and publicly insured patients. A lack of data on oncology providers and equipment impedes coordinated efforts to address rural shortages. Graduate Medical Education funding greatly favors large, urban, tertiary care teaching hospitals over residency training in rural, critical access and community-based hospitals and clinics. Policies have the potential to transform rural health care. This includes increasing advanced practice provider postgraduate oncology training opportunities and expanding the scope of practice; improving health workforce and services data collection and aggregation; transforming graduate medical education subsidies to support rural student recruitment and rural training opportunities; and expanding federal and state financial incentives and payments to support the rural cancer infrastructure.

INTRODUCTION

Cancer is the second leading cause of death in the United States.¹ Although cancer deaths have decreased nationally, declines have been lower in rural (nonmetropolitan) areas. Rural Americans have lower incidence but higher death rates from cancers that can be prevented through screening,^2,3 have lower screening rates than urban individuals,^4,5 are more likely to present with later stage cancers,³ and have poorer cancer outcomes and lower survival.^6,7 Rural health provider shortages and lack of cancer services may explain some disparities.^8-11

ASCO estimates just three percent of medical oncologists work in rural areas.¹² Rural communities have fewer per capita gastroenterologists, general surgeons, and radiation oncologists integral to colorectal cancer screening and treatment compared with nonrural areas.¹³ The American Society for Radiation Oncology found a decline in rural radiation oncologists (16%-13%) between 2012 and 2017.¹⁴ In Arizona, the lack of any gastroenterologists, medical oncologists, or pathologists in all nonmetropolitan counties and hematology/oncology physicians in the two most rural counties was unchanged from 2009 to 2019.^15,16

Rural areas lack cancer services. Over a quarter of Texas counties had zero mammography capacity, which did not improve between 2003 and 2009.⁹ Rural counties were significantly more likely to have zero mammography capacity compared with urban counties. Nationally, the proportion of Critical Access Hospitals providing oncology and chemotherapy services declined from 2008 to 2017.¹⁷ This care delivery review explores factors that contribute to insufficient cancer capacity and makes policy recommendations to improve rural cancer care delivery.

FACTORS CONTRIBUTING TO RURAL HEALTH CARE CAPACITY SHORTAGES

Rural Resident Demographics and Physician Recruitment

Rural populations often have lower incomes and are more likely to be uninsured or have Medicare or Medicaid compared with urban populations.¹⁸ Uncompensated care, unfavorable payer mix, and low patient volume pose challenges in rural physician recruitment and retainment.¹⁹ After adjusting for specialty, work hours, and provider and practice characteristics, rural physicians have lower incomes compared with urban physicians and experience longer work hours and more patient visits.²⁰

Physicians who grew up or lived in rural areas are more likely to practice there.²¹ However, rural students are less likely to attend and complete college compared with urban students and are, therefore, less likely to become physicians.^22,23 Because of rural physician recruitment challenges, international medical graduates (IMGs) on J-1 visa visitor exchange programs help fill the gaps in rural areas, especially in primary care.²⁴ IMGs are more likely to practice in rural needy areas compared with US medical graduates.²⁵ J-1 visa waivers can help rural health facilities recruit IMGs to fill open positions.

Cancer Infrastructure Financial Challenges

Between 2005 and 2022, 182 rural hospitals closed because of factors such as low profitability and patient volume²⁶ resulting in dramatic decreases in the availability of rural surgical oncology services.²⁷ A rural hospital closure has been associated with a 9% reduction in a rural county's physician supply, worsening existing provider shortages.²⁸ Even if efforts to increase the supply of rural oncology physicians are successful, access to care would not improve without the existence of rural health care facilities.

Compared with urban hospitals, rural hospitals tend to be smaller with fewer beds. Rural patients with complex medical needs, including cancer, prefer to seek care from urban hospitals offering more services, thus bypassing nearby rural hospitals.^29,30 Rural patients with high acuity care needs are often referred to larger hospitals, resulting in lower occupancy rates for rural versus urban hospitals and lower profit margins.^31,32 Rural patient decisions on where to seek cancer care can be complex. Even when services are available, some patients still bypass the closest rural hospital, seeking care from reputable hospitals known for quality in urban areas.^30,33

Although the number of US mammography machines is increasing, the number of facilities is decreasing, especially in rural counties.³⁴ Diagnostic radiologist shortages challenge mammography facility fiscal viability,^35,36 made worse by high liability insurance cost and low or no reimbursement.³⁷ University-based mammography programs often have a negative contribution margin (expenses exceeded revenues), primarily driven by diagnostic mammograms—complex, time-intensive procedures with higher staffing needs. Diagnostic mammograms take five times as long compared with screening mammograms, yet Medicare payment is only 11% higher than screening mammograms.³⁸ Centers for Medicare & Medicaid Services (CMS) has proposed mammography service payment reductions in recent years.^39-42

Measurement Issues—Lack of Data

Under the US Food and Drug Administration (FDA), the Mammography Quality Standards ACT requires mammogram locations to have annual inspections and FDA certification every three years.⁴³ The FDA's online, public Mammography Registry includes certified mammogram locations, facility names, and contact information.⁴⁴ However, it does not have capacity data (eg, number or type of mammography machines and hours of operation).

The ability to overlay oncology health care workforce, access, outcomes, and demographic/socioeconomic data sets is nonexistent. Data are fragmented because of siloed, uncoordinated public and private data collection systems, such as hospitals, community health centers, health plans, and state and federal governments.⁴⁵ Additionally, data were not intended for and thus are often not used for quality improvement purposes.⁴⁶ The right type of data can also be an issue. For example, the CMS National Provider Identification (NPI) database captures provider name, specialty, and practice location but does not provide full time equivalent (FTE) and is often outdated.^47,48 Development of effective strategies to reduce rural cancer disparities remains a challenge without relevant data coordinated by a single entity.

Graduate Medical Education

Compared with urban students, rural students have lower graduation rates, grades, and MCAT scores^21,23,49,50 and are thus less likely to apply to or be competitive for medical school admission. Geographic proximity to higher education is associated with an increased rate of high school students applying to college after controlling for socioeconomic factors.⁵¹ Geographic proximity to a medical school is also associated with an increased rate of medical school application and matriculation.⁵² Medical residents and fellows train primarily in urban hospital and outpatient settings. Physicians tend to go into practice near their Graduate Medical Education (GME) training location, but the vast majority of residency programs are in urban areas.^50,53 Rural GME community-based rotations can influence future rural practice location.⁵⁴

Despite spending $22 billion US dollars (USD) dollars to subsidize GME annually in the United States, efforts to address disparities in the physician workforce have had little success. Direct Medicare GME subsidies ($3.8B USD in 2019) cover resident stipends, supervising physician costs and administration while indirect subsidies ($10.1B USD) cover the higher costs of hospitals with GME training programs.⁵⁵ Subsidies are not tied to geographical workforce needs. Instead, they incentivize urban over rural locations and inpatient hospital settings over outpatient, ambulatory sites where most physicians practice when they finish GME training.⁵⁶

Medicare GME payments are based on antiquated inpatient formulas. Residents also train in outpatient clinic sites that are ineligible for the higher indirect GME payments. Medicare GME support greatly favors large, urban, tertiary care teaching hospitals and much lower payments for residency training in rural, critical access and community-based hospitals and clinics.⁵⁶ The rural health care infrastructure provides rich training milieus and would better align subsidies with unmet need.

POLICY RECOMMENDATIONS

We propose three overarching goals and nine policy recommendations to improve rural cancer care capacity.

Goal 1: Leverage and Incentivize the Rural Cancer Care Workforce

Recommendation 1.

Increase advanced practice provider (APP) oncology training. Oncology demand will outpace oncologist supply.⁵⁷ Nationally, APPs comprise a larger percent of providers in rural versus urban areas.⁵⁸ Postgraduate physician assistant and nurse practitioner (NP) oncology fellowships can help address rural oncology shortages.⁵⁹ There are only 11 NP oncology fellowships and six physician assistant oncology fellowships in the United States.^60,61 However, there are 53 National Cancer Institute designated Comprehensive Cancer Centers serving their communities and the broader public by integrating training and education for biomedical researchers and health care professionals.⁶² In partnership with National Cancer Institute, they could expand APP oncology fellowship training. Comprehensive Center Support Grants now require designated cancer centers to identify and work with their catchment areas, including minority and underserved populations for renewal.⁶³

Recommendation 2.

Expand infusion service scope of practice and eliminate payment barriers for appropriately trained APPs to improve rural radiation therapy access. CMS requires direct supervision for radiation therapy services either by a physician or nonphysician practitioner (ie, APP) with the ability to perform such procedures under the state's scope of practice for payment.⁶⁴ Most states do not allow APPs to supervise radiation therapy. Arizona allows APPs to order, interpret, and perform laboratory, radiographic, and other diagnostic tests as long as they meet training requirements.⁶⁵ The American College of Radiology raises concern that using ionizing radiation for clinical treatment is complex with risks. They oppose expanded APP scope of practice, with the rationale that radiation oncologists undergo significant training to deal with radiation therapy administration and complications.⁶⁶ An Australia-based study found that a NP-led model of emergency cancer care provided an equal standard of care to physician-led models.⁶⁷ States with expanded APP scope of practice should assure through licensing that they are properly trained to positively contribute to cancer care delivery.

Recommendation 3.

Increase the number of IMGs to fill gaps in the rural oncology workforce. J-1 Visa recipients who have completed their medical education or training in the United States are required to move back home for at least two years before reentry into the United States unless they receive a waiver.⁶⁸ The Conrad-30 J-1 Waiver program, the largest waiver program, requires physicians to work in a Health Professional Shortage Area of Medically Underserved Area for at least three years. However, states are limited to 30 waivers annually. Federal and state governments should collaborate to expand the number of waivers to meet rural workforce needs by specialty.

Goal 2: Improve Cancer Services Data Quality, Timeliness, and Accuracy

Recommendation 4.

Add FDA Mammography Registry data entry fields on the number of mammograms done at each site per year, the hours of mammogram availability (ie, hours per week, weeks per year), and the number and type of mammography machines at each facility (ie, film, digital, 3D). Current research on mammography capacity focuses on the existence of mammography services in a given area.^9,69 Additional data can more accurately assess rural mammography capacity.

Recommendation 5.

Have state licensing bodies collect clinical FTE and location data at the time of licensure and renewal. For oncologists, this should include the specific cancer they specialize in treating to better understand the availability of oncologists, incidence, and mortality by cancer site. Although state leaders may be aware of provider shortages in underserved and rural areas, many are unable to quantify these shortages because of a lack of detailed data collection.⁷⁰ These data can allow states to formally evaluate the effect of policy interventions. For example, Indiana implemented a loan repayment program to increase the number of mental health providers practicing in underserved areas.⁷¹ Using biennial licensure renewal data, they found that the program increased access to mental health professionals in priority areas.

Recommendation 6.

Create a CMS administered data aggregation and analysis system to study the effects of policies to address rural health provider shortages. State-level health workforce data on FTE and specialty type should be aggregated nationally and made available to those working to address rural provider shortages. CMS manages the NPI system. State licensing bodies could submit NPI linked provider data (FTE, practice site, and medical specialty) to evaluate and improve efforts to reduce rural provider shortages, barriers to health services by geography, and annual changes. Collecting more timely accurate data could track changes in rural providers, such as those in oncology, over time. Data could then be used to coordinate health care workforce efforts nationally.

It should be acknowledged that data collection comes with a cost. Although studies have not examined the costs and benefits of these recommendations specifically, other studies have examined return on investment for public health data tracking programs. For example, the CDC launched the Environmental Public Health Tracking Program in 2002 to understand links between the environment and health, especially chronic disease. The program's societal benefits ($5.2-$17.1 billion USD) far outweigh the annual costs ($25 million USD) and result in improved health outcomes.⁷²

Goal 3: Increase the Number of Facilities, Providers, and Specialists Providing Rural Cancer Care

Recommendation 7.

Require medical schools and residency training programs to recruit and admit students from rural backgrounds and promote rural job placement. Although chosen specialty and practice location are influenced by multiple factors, Rosenblatt argues that medical schools are not merely passive educational conduits training physicians without influencing their career paths.⁷³ They receive billions of dollars in federal support and remain medical education gatekeepers.⁷⁴

Decades of evidence demonstrate that medical schools can successfully produce rural physicians. Thomas Jefferson University implemented the Physician Shortage Area program in 1974, which focused on recruiting, prioritizing, and admitting medical school applicants from rural areas who plan to practice in rural communities after graduation and require rural clinical rotations.⁷⁵ A systematic review demonstrated that medical schools with targeted efforts to increase rural physicians are successful.⁷⁶ The University of Arizona launched a new Primary Care Physician Scholarship in 2019 with state allocated funds, which covers medical education tuition in exchange for a commitment to practice in a health professional shortage area after graduation.⁷⁷ It supports 86 active medical school students and could serve as a model for other states. If just half of medical schools in the United States enacted efforts aimed at recruiting and admitting students from rural backgrounds, subsidized their educational costs, and provided rural practice placement support post-graduation, rural physician placement would be improved.⁷³

The federal government should reform Medicare GME subsidies to address unmet rural needs and balance rural and urban training. The first step is implementing the Affordable Care Act's National Healthcare Workforce Commission, which was never funded by congress and or convened.⁷⁸ This could be the first body to oversee coordination of GME funding nationally to increase transparency and accountability. A second step involves tying the annual $22 billion USD in GME payments to workforce needs instead of perpetuating outdated bed to resident formulas that favor urban, inpatient hospital medical training. Payments are made to hospitals without specifying the types of physicians trained based on need.⁷⁹ Reforms should tie GME subsidies to specified training outcomes. States have the flexibility to align Medicaid GME payments to meet state and rural community needs—but most follow anachronistic Medicare direct and indirect payment.

Coordinated national efforts using workforce data can help target GME subsidies to resident training in rural/underserved areas with documented physician shortages in high need specialties (eg, oncology, primary care, general surgery). The National Rural Health Association suggests that the Accreditation Council for GME should require accredited residencies mandate a clinical rotation at a rural site (eg, Critical Access Hospital or Federally Qualified Health Center (FQHC)) to increase rural medicine exposure.⁸⁰

Recommendation 8.

Federal and state agencies should expand financial incentives and payments to support the rural cancer infrastructure. Government tax credits and loan repayment programs, such as the National Health Services Corps, have been successful in recruiting physicians to rural and underserved areas with demonstrated cost savings.^81-83 Each additional National Health Services Corps supported behavioral health provider results in significant cost reductions in per community health center behavioral health visit with the largest savings observed in rural sites.⁸⁴ Many incentive programs focus on high needs specialties (eg, primary care) and may be unavailable to other specialties (eg, oncology). These programs are cost effective and should be expanded to train, recruit, and retain rural oncology physicians identified as high/critical needs by state or national workforce data.

Medicaid should increase reimbursement for rural cancer services. The number of rural mammography facilities has declined in recent years because of financial challenges.³⁴ Rural individuals are more likely to be enrolled in Medicare or Medicaid, which usually pay less than private payers.^85,86 Illinois' state health department raised Medicaid mammography payment to match Medicare rates as long as the facility agrees to annual mammography data reporting.⁸⁷ Additionally, mobile mammography programs are effective in ensuring access to breast cancer screening among rural populations but may require increased reimbursement rates because of hidden costs.^88,89 Breast cancer screening has a demonstrated ability to save Medicaid programs money with a positive return on investment, justifying enhanced reimbursement.⁹⁰

All states should expand Medicaid to support rural access to care. Medicaid expansion significantly decreases rural hospital closure rates compared with nonexpansion states by reducing uncompensated care.⁹¹ Studies have consistently shown that Medicaid expansion results in a positive return on investment for states.^92-94

Recommendation 9.

Payers and health systems should transform rural payment models to ensure financial sustainability and quality cancer care. Payment models that prioritize value-based rather than volume-based care in low patient volume settings can help ensure rural hospital access and quality. The 2019 Pennsylvania Rural Health Model focuses on transforming rural hospital payments to global budget payments from public and private payers.⁹⁵ This model stabilizes cash flows and incentivizes rural hospitals to invest in expanded services that reduce potentially avoidable utilization, including access to specialty care. Initial evaluation data indicate that participating hospitals have experienced improvements in operating margins, quality of care, and significant value in the cost/benefit ratio.⁹⁶ Strategies that focus on improving the quality of care and expanding rural hospital services may be effective in reducing bypass.⁹⁷

In conclusion, rural versus urban cancer outcome disparities, an unequal distribution of health providers, and lack of access to specialized cancer care services in rural areas must be addressed. Federal and state level policies have the potential to increase the rural health workforce. This includes leveraging APPs, improving health workforce and services data collection, and aligning GME subsidies and payments to support high quality, accessible rural oncology care.

AUTHOR CONTRIBUTIONS

Conception and design: Adrienne B. Lent, Elizabeth T. Jacobs, Leila Barraza, Elizabeth A. Calhoun

Administrative support: Daniel Derksen

Collection and assembly of data: Adrienne B. Lent, Elizabeth A. Calhoun

Data analysis and interpretation: Adrienne B. Lent, Daniel Derksen, Elizabeth A. Calhoun

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Policy Recommendations for Improving Rural Cancer Services in the United States

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

No potential conflicts of interest were reported.