Rural Communities in the United States Face Persistent Disparities in Access to Orthopaedic Surgical Care

Abstract

Background

Access to orthopaedic care across the United States (U.S.) remains an important issue, however, no recent study has examined disparities in rural access to orthopaedic care. The goals of the present study were to (1) investigate trends in the proportion of rural orthopaedic surgeons from 2013 to 2018 as well as the proportion of rural U.S. counties with access to such surgeons and (2) analyze characteristics associated with choice of a rural practice setting.

Methods

The study analyzed the Centers for Medicare and Medicaid Services (CMS) Physician Compare National Downloadable File (PC-NDF) for all active orthopaedic surgeons from 2013 to 2018. Rural practice settings were defined using Rural-Urban Commuting Area (RUCA) codes. Linear regression analysis investigated trends in rural orthopaedic surgeon volume. Multivariable logistic regression evaluated the association of surgeon characteristics with rural practice setting.

Results

The total number of orthopaedic surgeons increased 1.9%, from 21,045 (2013) to 21,456 (2018). Meanwhile, the proportion of rural orthopaedic surgeons decreased by roughly 0.9%, from 578 (2013) to 559 (2018). From a per capita perspective, the number of orthopaedic surgeons practicing in a rural setting per 100,000 population ranged from 4.55 orthopaedic surgeons per 100,000 in 2013 and 4.47 per 100,000 in 2018. Meanwhile, the number of orthopaedic surgeons practicing in an urban setting ranged from 6.63 per 100,000 in 2013 and 6.35 per 100,000 in 2018. The surgeon characteristics most associated with decreased odds of practicing orthopaedic surgery in a rural setting included earlier career-stage (OR: 0.80, 95% CI: [0.70-0.91]; p < 0.001) and sub-specialization status (OR: 0.40, 95% CI: [0.36-0.45]; p < 0.001).

Conclusion

Existing rural-urban disparities in musculoskeletal healthcare access have persisted over the past decade and could worsen. Future research should investigate the effects of orthopaedic workforce shortages on travel times, patient cost burden, and disease specific outcomes.

Level of Evidence: IV

Introduction

Recent estimates from the United States (U.S.) Census Bureau suggest the number of Americans older than 65 will increase from 15% in 2018 to nearly 25% in 2060.¹ Many common health problems faced by older populations are musculoskeletal in nature, including fractures of the hip, spine, or forearm, as well as degenerative joint diseases of the hip or knee.^2,3 Studies have projected the demand for orthopaedic surgical care to grow rapidly over the next several decades, with potential for substantial workforce shortages across the nation.^4-6 It has also been estimated that more than 1 in 5 older Americans live in rural areas and that these communities face major health inequities and poorer outcomes compared to their urban and suburban counterparts.^7-11

Prior research has demonstrated that the supply of orthopaedic surgical services is not spread evenly throughout the United States.^5,6,14 Specifically, one study surveying 145 hospital administrators similarly found that only 30% of rural hospitals had a full-time orthopaedic surgeon on staff.¹² Furthermore, 71% stated the need for additional orthopaedic surgical services within their community.¹² In order to ensure equity in musculoskeletal health and healthcare access, it is necessary to pursue an effective geographic distribution of the US orthopaedic workforce. Although previous research has demonstrated discrepancies in the workforce between rural and urban areas, there has yet to be an examination of rural-urban disparities in orthopaedic access over the past decade.¹¹

Therefore, the goals of the present study were to (1) identify trends in rural access to orthopaedic surgeons in the United States between 2013 and 2018 and (2) determine the surgeon characteristics associated with practice in a rural area.

Methods

Data Source

The present study retrospectively queried the Centers for Medicare and Medicaid Services (CMS) Physician Compare National Downloadable File (PC-NDF) for all orthopaedic surgeons with an active National Physician Identifier from January 1st, 2013, to December 31st, 2018.¹³ Following the initial data extraction, the dataset was linked with the Medicare Part B Provider and Other Supplier Payment and Utilization File (POSPUF) dataset to obtain additional surgeon level characteristics.¹⁴ For each orthopaedic surgeon, we compiled self-reported gender, year of medical school graduation, geographic practice location, number of annual Medicare beneficiaries, group practice size, and sub-specialization status. The number of years in practice was estimated by taking the difference between the reported medical school graduation year and year of data reporting, followed by subtracting five years for generalists and six years for subspecialists. These data were then categorized as early (≤14 years), mid (15-24 years), and late (≥25 years) career stages, in concordance with previously published methodology.¹⁵

Rural and Urban Classification

Within the CMS POSPUF dataset, each physician entry was assigned a Rural-Urban Commuting Area (RUCA) code based on their reported primary practice location. RUCAs are a classification system developed by the Economic Research Service of the United States Department of Agriculture (USDA) using 2010 US census tracts as well as the 2006-2010 American Community Survey. The RUCA classification utilizes whole numbers (1-10) and is utilized to distinguish zip code level geographic regions based on differences in population density, degree of urbanization, and daily commuting patterns.¹⁶ In concordance with previous literature, we grouped RUCA codes 1 through 3 as urban and RUCA codes 4 through 10 as rural.^17,18 These categories are defined by primary commuting flow to census tracts of size 50,000 or more (Urban, RUCA 1-3), 10,000 to 49,999 (large rural, RUCA 4-6), 2,500 to 9,999 (small rural, RUCA 7-9), and 2,499 or less (isolated rural, RUCA 10).^19,20 Zip code to county geographic crosswalk files were obtained from the Office of Policy Development and Research at the U.S. Department of Housing and Urban Development (HUD) and utilized to determine surgeon practice region and county level access trends.²¹

Statistical Analysis

Descriptive statistics analysis was conducted based on the surgeon characteristics, such as self-reported gender, number of years in practice, sub-specialization status, U.S. census region, Medicare patient volume, and group practice size. A bivariate analysis was conducted using Pearson’s chi-squared test. County population estimates were extracted from the U.S. Census Bureau website and used to calculate the population density of orthopaedic surgeons per 100,000 individuals in each U.S. counties.²² The above data sets were imported into Quantum Geographic Information System (QGIS) geospatial analysis software (version 3.12.1; open-source license GNU GPLv2), along with county boundary files from the U.S. Census Bureau website.²² Heatmaps were then constructed to visualize changes in the geographic distribution and relative population density of U.S. orthopaedic surgeons between 2013 and 2018. Linear regression analysis was utilized to determine the significance of trends in the number and proportion of rural orthopaedic surgeons.

A multivariable logistic regression was used to evaluate the impact of various surgeon characteristics, as well as practice setting characteristics, on rurality of practice setting. The constructed regression model also included gender, number of years in practice, sub-specialization status, census region, practice group size, and number of Medicare beneficiaries treated annually. Odds ratios (ORs) and their respective 95% confidence interval (95% CI) were determined to quantify the effects of the included variables on the rurality of practice setting. All statistical tests were two-sided and utilized a pre-determined significance threshold of p < 0.05. All statistical analyses were performed via RStudio R version 4.1.1 (R, Foundations for Computational Statistics, Vienna, Austria) using the dplyr, DescTools, gtsummary, and ggplot2 packages.

This study was considered exempt from review by the Cleveland Clinic Foundation ethical review board due to the public nature of all data included and lack of protected health information as defined by 45 CFR 46.102 of the Department of Health and Human Services’ Code of Federal Regulations.

Results

Trends in Rural Access to Orthopaedic Surgeons

Throughout the study period the total number of practicing orthopaedic surgeons in the dataset increased by approximately 1.9%, from 21,045 in 2013 to 21,456 in 2018. The proportion of these surgeons working in rural areas decreased significantly over the same timeframe, from 11.9% to 11.0% (p = 0.002) (Figure 1). This trend was driven primarily by a decrease in orthopaedic surgeons working in large rural areas (9.1% in 2013 to 8.4% in 2018, p = 0.001). Over the same period, no significant changes were seen in the proportion of surgeons working in small rural areas (p = 0.097) and isolated rural areas (p = 0.772) (Figure 1). Among rural U.S. counties specifically, roughly 66.6% (1,316 of 1,976) did not have access to a local orthopaedic surgeon in 2013 (Figure 2A). In 2018, only 33.5% of rural counties (664 of 1,976) possessed access to a local orthopaedic surgeon, compared to 67.8% (791 of 1,167) of urban counties (Figure 2B).

Figure 1.

Trends (2013-2018) in the Proportion of Total Orthopaedic Surgeons Practicing in Rural Areas, stratified by Rural Urban Commuting Area (RUCA) categories.*

*Urban (RUCA 1-3, Primary commuting flow to metropolitan area of size 50,000 or more); Large Rural (RUCA 4-6, Primary commuting flow to micropolitan area of size 10,000 to 49,999); Small Rural (RUCA 7-9, Primary commuting flow to town of size 2,500 to 9,999); Isolated Rural (RUCA 10, Primary commuting flow to an area of 2,499 or less).

Figure 2A to 2B.

Density of Practicing Orthopaedic Surgeons per 100,000 Population, across United States Counties in (2A) 2013 and (2B) 2018.

Surgeon Characteristics Associated with Rural Practice Setting

In 2018, we analyzed a national cross-section of 21,456 practicing orthopaedic surgeons, of which 89.0% (19,088 of 21,456) were determined to practice in an urban setting and 11.0% (2,368 of 21,456) were determined to practice in a rural setting (Table 1). Following multivariable logistic regression analysis, sub-specialization in any given orthopaedic sub-field was shown to be significantly associated with decreased odds of rural practice setting (OR: 0.40, 95% CI: [0.36-0.45]; p < 0.001) (Table 2). With respect to geography, U.S. orthopaedic surgeons were significantly more likely to practice in a rural setting if located in the Midwest (OR: 1.92, 95% CI: [1.64-2.25]; p < 0.001), South (OR: 1.23, 95% CI: [1.05-1.44]; p = 0.010), or West (OR: 1.39, 95% CI: [1.17-1.65]; p < 0.001) compared to the Northeast. Furthermore, the present data show that earlier career orthopaedic surgeons, defined as those who have practiced 14 years or less, are less likely to practice in a rural setting, compared to later-career orthopaedic surgeons (OR: 0.80, 95% CI: [0.70-0.91]; p < 0.001) (Table 2).

Table 1.

Orthopaedic Surgeon Characteristics, Stratified by Urban vs. Rural Practice Setting, 2018

		Rural (n = 2368)a	Urban (n = 19088)a	p-valueb
Gender				0.3
	Male	2247 (95)	18,007 (94)
	Female	121 (5.1)	1081 (5.7)
Region				<0.001
	Northeast	338 (14)	3190 (20)
	Midwest	729 (31)	4135 (22)
	South	819 (35)	6968 (37)
	West	482 (20)	4075 (21)
Number of Years Practiced				<0.001
	25 and more	605 (31)	4468 (27)
	15-24	596 (31)	4375 (27)
	≤ 14	747 (38)	7644 (46)
Patient Volume				0.072
	500 and more	526 (22)	4055 (21)
	201-499	1095 (46)	8564 (45)
	11-200	747 (32)	6469 (34)
Practice Group Size				<0.001
	200 and more	438 (24)	6147 (41)
	50-199	499 (28)	3286 (22)
	1-49	857 (48)	5600 (37)
Sub-specialization				<0.001
	General	1777 (79)	10707 (59)
	Sub-specialist	480 (21)	7447 (41)

^a n (%)

^b Pearson’s Chi-squared Test

Table 2.

Multivariable Logistic Regression for Orthopaedic Surgeon Characteristics Associated with Rural Practice Setting

		Odds Ratio	95% CI	p-value
Gender
	Male	Reference
	Female	1.13	0.89, 1.41	0.3
Region
	Northeast	Reference
	Midwest	1.92	1.64, 2.25	<0.001
	South	1.23	1.05, 1.44	0.010
	West	1.39	1.17, 1.65	<0.001
Number of Years Practiced
	25 and more	Reference
	15-24	1.07	0.94, 1.23	0.3
	≤ 14	0.80	0.70, 0.91	<0.001
Patient Volume
	500 and more	Reference
	201-499	1.09	0.96, 1.24	0.2
	11-200	0.98	0.85, 1.14	0.8
Practice Group Size
	200 and more	Reference
	50-199	2.10	1.83, 2.41	<0.001
	1-49	2.20	1.94, 2.49	<0.001
Sub-specialization
	General	Reference
	Sub-specialist	0.40	0.36, 0.45	<0.001

Discussion

In response to the growing demand for musculoskeletal care across the U.S., it is essential to understand access to orthopaedic services across a variety of marginalized communities. It has been well demonstrated that patients residing in rural areas face higher rates of chronic disease, poorer access to social and psychological support services, and lower life expectancies.^23,24 Although prior research has raised concerns regarding the uneven distribution of the orthopaedic workforce, there has been insufficient evidence to assess trends in rural access over the past decade. The present study determined that, as of 2018, approximately two thirds of rural counties did not have access to a local orthopaedic surgeon. Furthermore, the proportion of rural orthopaedic surgeons was shown to gradually decrease from 2013 to 2018. Residence in the Midwest, later career-stage, smaller practice size, and lack of sub-specialization were associated with greater odds of rural practice setting. Despite increased attention and discussion surrounding the musculoskeletal needs of rural communities in the U.S., these data suggest there have been no definitive improvements in rural-urban access disparities. Considering the aging current workforce and propensity for younger surgeons to practice near metropolitan areas, recruitment and retention of the rural orthopaedic surgeons remains an urgent policy priority.^6,25,26

Limitations

The above findings have several limitations. The data utilized for this study were confined to orthopaedic surgeons certified by the CMS and caring for at least 11 Medicare patients annually. It should be noted that surgeons and those in cash-based practices are not represented within the datasets used for the present study. However, recent survey estimates have shown that those not certified by the CMS and those not accepting Medicare patients to be a small minority of the overall orthopaedic workforce.²⁷ Additionally, more specific information describing employment characteristics, such as compensation models, part-time work, or locum tenens contracts, were not accessible for this study.^15,28,29 Such data may be helpful to understand the motivating forces behind workforce fluctuations in terms of geography and practice setting. Trends such as declining procedural reimbursements, alternative payment models, and health system consolidation have been well described, although their effects on the rural orthopaedic workforce remain unclear.³⁰ Finally, the current analysis was limited to be between 2013 to 2018 and was unable to account for the large workforce disruptions in the year 2020 related to the COVID-19 pandemic.³¹ It is possible the disparities highlighted above have been recently exacerbated by the closure of rural hospitals or safety hazards faced by older surgeons still in practice.^25,32 Further research on this subject is warranted as more recent workforce data becomes available.

Trends in Rural Access to Orthopaedic Surgeons

While the total number of practicing orthopaedic surgeons rose over the study period, the proportion of these surgeons in rural practice settings demonstrated a modest decrease. Furthermore, as of 2018, roughly two of every three rural counties did not have access to a local orthopaedic surgeon. These data support prior research suggesting disparities in rural access to a wide variety of surgical services.^26,33 A 2005 policy report from the University of Washington Rural Health Research Center showed rural orthopaedic surgeons to comprise 13.5% of the overall workforce. Additionally, the report determined that 10.2%, 2.7%, and 1.6% of orthopaedic surgeons practiced in large rural, small rural, and isolated rural areas, respectively. The present study findings for the year 2018, using the same categorization scheme, suggest a 2.5% decrease the proportion of orthopaedic surgeons in any rural practice setting. For those practicing in large rural, small rural, and isolated rural settings, the present study findings show decreases of 1.8%, 0.5%, and 1.1%, respectively. These data should raise concerns for impending exacerbations of existing rural-urban disparities. Prior research has indicated the rate of financial distress among rural hospital to be rapidly increasing, with more than 100 confirmed closures distributed throughout the U.S. between the years 2010 and 2018.³⁴ These closures have contributed to worsening access to essential medical, surgical, and emergency services in the surrounding communities.³⁵ A variety of interventions have been attempted to address workforce disparities including, but not limited to, telehealth expansions, visiting consultant clinics, and public service loan forgiveness programs.³⁶ However, the implementation of these programs has often lacked coordination beyond the local level.³⁷ Future research should investigate the impact of workforce changes on delays to orthopaedic care, travel times, and cost-burden on patients.³⁸

Surgeon Characteristics Associated with Rural Practice Setting

The surgeon characteristics most associated with increased odds of practicing orthopaedic in a rural setting included later career-stage, residence in the Midwest, smaller practice size, and generalist status. Recent evidence has shown trends towards increasing fellowship sub-specialization, health system consolidation, and difficulty of managing a sustainable solo-practice.^15,39,40 Considering these trends, these present study findings highlight the persisting rural-urban disparities in orthopaedic surgical care and suggest a potential for further decline in the rural orthopaedic workforce over the coming years. An aging rural physician workforce has been previously described in both the generalist and surgical subspecialty literatures. The present study underscores the continued trend of later career staged orthopaedic surgeons to be significantly associated with rural practice setting as demonstrated in previous literature. Among primary care physicians, a recent systematic review determined that rural upbringing, receipt of government scholarships, and enrollment in dedicated rural-health medical school programming to be the three most predictive factors for rural practice setting. Among orthopaedic surgeons, similar factors may be helpful to identify, recruit, and retain newly trained or earlier career surgeons in rural areas.

Conclusion

The present study determined that, as of 2018, approximately two-thirds of rural counties did not have access to a local orthopaedic surgeon and the proportion of rural orthopaedic surgeons has gradually decreased from 2013 to 2018. Additionally, surgeons practicing general orthopaedics in the Midwest, in small groups, in later stages of their careers had greater odds of working in a rural community. These findings suggest that persisting rural-urban disparities in musculoskeletal health may worsen if newly trained orthopaedic surgeons migrate to major metropolitan areas without filling the positions of retiring orthopaedic surgeons in rural regions. Future research should investigate the effects of orthopaedic workforce shortages on travel times, patient cost burden, and disease specific outcomes.