Analyzing trends of Medicare-participating nurse… : Journal of the American Association of Nurse Practitioners

Introduction

The nurse practitioner (NP) and physician associate (PA) workforces have become increasingly integral to the delivery of primary and specialty care in the United States. The number of NPs and PAs, as well as the number of patient visits by these providers, has expanded substantially over the past couple of decades, with current rates of NP and PA entry into the workforce far outpacing that of physicians (Auerbach et al., 2018; Hooker et al., 2022; Xue et al., 2017, 2019). This rapid growth of NPs and PAs has been driven by a diverse array of factors, including shorter training times relative to physicians, fewer constraints on creating new training programs, rising recognition of the importance of team-based clinical care, and expansion of scope of practice regulations (Auerbach et al., 2018; Hooker & Christian, 2023; Xue et al., 2020).

Nurse practitioners and PAs in the United States are integral to addressing rising health care demands stemming from increasing physician workforce shortages, rising chronic disease burden, and an aging and growing U.S. population (Hacker, 2024; The Demographic Outlook, 2025; Walensky & McCann, 2025; X. Zhang, Lin, et al., 2020). These changes are occurring in a U.S. health care system that is already burdened by significantly worse health outcomes than peer countries, coupled with exceedingly high health care spending (Global Perspective on U.S. Health Care. Commonwealth Fund, 2023). Nurse practitioner and PA workforce growth trends are projected to continue in the coming years, driving a rapidly growing provider workforce (Auerbach et al., 2018; Hooker et al., 2022; Hooker & Christian, 2023). An increase in these providers could be a key contributor to mitigating workforce gaps, particularly in underserved and rural communities (Xue et al., 2019).

Despite the documented growth in NPs and PAs, there has been little research exploring how the geographic distribution of the NP and PA workforce has evolved over the past decade.

Geographic maldistribution of providers remains a persistent challenge with numerous studies documenting lower clinician density in rural and underserved areas (Machado et al., 2021; D. Zhang, Lin, et al., 2020). Some specialty-specific analyses have found that geriatric NPs and dermatology PAs practice predominantly in metropolitan areas (Xue et al., 2024; Young et al., 2022), whereas primary care NPs have increased notably in low-income and rural areas (Barnes et al., 2018; Xue et al., 2019). Understanding the distribution of NPs and PAs is particularly relevant for Medicare populations given the aging U.S. population. However, to date, there are no known temporal analyses of the geospatial distribution of all Medicare NPs and PAs on a per-county level.

Updated analyses of workforce distribution patterns can help policymakers identify where health care gaps may widen as the number of NPs and PAs increases and can inform initiatives to improve recruitment, training, and retention strategies. This study addresses gaps in the literature by leveraging a Centers for Medicare and Medicaid Services (CMS) national dataset and using spatial analysis methods. We aimed to characterize geographic and temporal trends in the Medicare-participating NP and PA workforce in the United States from 2017 to 2025 and to identify regions of disproportionately high and low clinician density in 2025, with the goal of generating evidence to inform policy, training, and workforce planning decisions that address geographic disparities in access to care. Guided by existing literature, we expected to find overall increases in NP and PA representation with significant urban-rural and geographic variation.

Methods

Data resources

Because of the availability of geographical data over time, the CMS Doctors and Clinicians national downloadable file, which includes providers who submit Medicare claims, was used as the primary dataset for these analyses (Doctors and Clinicians. Provider Data Catalog, 2025). Our study included data from 2017 to 2025. The dates of all files included are shown in Supplemental Digital Content 1, http://links.lww.com/JAANP/A456. Linkage between years was performed using the National Provider Identifier (Parsons et al., 2017), with variable names linked across releases. We obtained 2025 state-level scope of practice data for NPs and PAs from the American Association of Nurse Practitioners (AANP) and the American Academy of Physician Associates (“PA State Practice Environment,” 2025; State Practice Environment, 2025). Our total dataset contains 348,004 unique nurse practitioners and 162,377 unique physician associates.

Variables

Providers were included if they were NPs or PAs, identified using the primary specialty code. The number of providers per 100,000 population was our primary density metric. We used the 2023 Rural-Urban Continuum Codes (RUCC), which provides a detailed classification of rurality based on population size, urbanization, and proximity to metropolitan areas, to classify county rurality and obtain county and state population counts (Rural-Urban Continuum Codes. Economic Research Service, 2023). Following standard convention, we designated RUCC 1–3 as urban and RUCC 4–9 as rural for all analyses (Rhudy et al., 2020; Schmitz et al., 2025; Shen et al., 2024). We first mapped the ZIP code of providers to county codes and then mapped county codes to RUCC values. Providers who submitted claims from ZIP codes present in both urban and rural counties were categorized as rural providers and only mapped in rural counties. For each state, we classified regulatory environments into categorical practice restriction levels based on organizational definitions. There were four categories provided for PAs (“optimal,” “advanced,” “moderate,” and “reduced”) and three categories provided for NPs (“full practice,” “reduced practice,” and “restricted practice”). The NP criteria categorize states based on policies relating to the ability to independently practice under the sole authority of the state board of nursing versus limitations on NP practice such as career-long supervision, delegation, or team management by another health provider. The PA criteria categorize states based on policies relating to factors including legal requirements regarding relationships between PAs and physicians, the existence of direct payment to PAs rather than practice/employer, and the presence of a separate PA regulatory board or having at least one full PA voting member on the medical/healing arts board (“PA State Practice Environment,” 2025.; State Practice Environment, 2025). The categorizations of each state are shown in Supplemental Digital Content 2, http://links.lww.com/JAANP/A456.

Analytic plan

We assessed differences in the distribution of NPs and PAs over time. Trends of time series data for each year between 2017 and 2025 were assessed using the Mann Kendall Trend Test with the pyMannKendall Python package (Hussain & Mahmud, 2019). For continuous variables, we assessed differences between time points using a two-sample t-test for significance. For categorical variables, the chi-squared test was used. All significance tests were performed with a p-value cut-off of .05. For both NPs and PAs, we identified changes in the gender, number of years since graduation, geographic region, and rurality from 2017 to 2025.

We created choropleth graphs to identify state-level and county-level trends. We assessed the percentage change in the number of NPs and PAs for each state from 2017 to 2025. Local Moran’s I was used to identify whether counties and their geographical neighbors were statistically significantly different from the national average for NP and PA density per 100,000 population in 2025. This clustering technique classified counties into four groups: high-high, low-low, low-high, and high-low. The first label represented whether a particular county had a higher or lower NP or PA density than the national average. The second label represented whether the neighbors of a particular county had a higher or lower NP or PA density than the national average. High-high classifications indicated regions with higher NP or PA density than average, and low-low classifications indicated regions with lower NP or PA density than average. The libpysal Python package was used for computation of spatial analysis statistics and cluster identification (Rey & Anselin, 2009).

To assess whether current workforce density differed across regulatory environments, we conducted one-way analysis of variance (ANOVA) tests comparing mean NP and PA density across practice restriction categories. Separate ANOVA models were estimated for NPs and PAs. When ANOVA results were significant, post hoc pairwise comparisons with Tukey honestly significant difference (HSD) correction were performed to identify specific group differences. All statistical analyses were run using the SciPy package in Python (Gommers et al., 2025).

Results

Demographic characteristics of the full cohort stratified by year (biennial data for simplicity) are shown in Table 1 with annual data available in Supplemental Digital Content 3–4, http://links.lww.com/JAANP/A456.

Table 1.

Analysis of physician associate and nurse practitioner Medicare workforce trends from 2017 to 2025

Category20172019202120232025p-ValueaTotal NP (N)109,100138,409170,936204,997252,755<.001Total PA (N)65,50073,93887,360102,244123,419<.001Mean years since graduation NP (SD)9.0 (8.0)8.5 (7.7)8.4 (7.4)8.5 (7.2)8.6 (7.1).91Mean years since graduation PA (SD)11.1 (8.8)11.0 (8.7)11.0 (8.6)11.0 (8.6)10.8 (8.7).45NP gender Female98,489 (90.3%)123,296 (89.1%)150,957 (88.3%)180,133 (87.9%)220,700 (87.3%)<.001 Male10,611 (9.7%)15,113 (10.9%)19,979 (11.7%)24,864 (12.1%)31,783 (12.6%)<.001 Unknown0 (0.0%)0 (0.0%)0 (0.0%)0 (0.0%)272 (0.1%).18PA gender Female43,487 (66.4%)49,995 (67.6%)59,870 (68.5%)71,319 (69.8%)87,629 (71.0%)<.001 Male22,013 (33.6%)23,943 (32.4%)27,490 (31.5%)30,925 (30.2%)35,648 (28.9%)<.001 Unknown0 (0.0%)0 (0.0%)0 (0.0%)0 (0.0%)142 (0.1%).04NP region Northeast19,589 (18.0%)24,686 (17.8%)29,622 (17.3%)35,229 (17.2%)43,957 (17.4%)<.001 Midwest29,562 (27.1%)37,089 (26.8%)44,617 (26.1%)52,065 (25.4%)61,422 (24.3%)<.001 South43,158 (39.6%)55,705 (40.2%)70,445 (41.2%)85,642 (41.8%)106,641 (42.2%)<.001 West16,777 (15.4%)20,904 (15.1%)26,208 (15.3%)32,015 (15.6%)40,658 (16.1%)<.001PA region Northeast14,742 (22.5%)17,304 (23.4%)20,612 (23.6%)24,261 (23.7%)29,307 (23.7%)<.001 Midwest14,806 (22.6%)17,039 (23.0%)19,633 (22.5%)22,413 (21.9%)26,757 (21.7%)<.001 South22,164 (33.8%)24,446 (33.1%)28,930 (33.1%)34,010 (33.3%)41,454 (33.6%)<.001 West13,767 (21.0%)15,118 (20.4%)18,149 (20.8%)21,528 (21.1%)25,858 (21.0%)<.001NP rurality Rural21,439 (19.7%)28,597 (20.7%)36,619 (21.4%)43,314 (21.1%)49,200 (19.5%)<.001 Urban87,661 (80.3%)109,812 (79.3%)134,317 (78.6%)161,683 (78.9%)203,555 (80.5%)<.001PA rurality Rural10,927 (16.7%)13,960 (18.9%)16,766 (19.2%)19,008 (18.6%)21,747 (17.6%)<.001 Urban54,573 (83.3%)59,978 (81.1%)70,594 (80.8%)83,236 (81.4%)101,672 (82.4%)<.001

Note: PA = physician associate; NP = nurse practitioner.

a

p-Values were calculated using the Mann Kendall Trend test to assess longitudinal monotonic trends with yearly data from 2017 to 2025.

The number of NPs increased from 109,100 in 2017 to 252,755 in 2025 (p-value <.001), while the number of PAs increased from 65,500 in 2017 to 123,419 in 2025 (p-value <.001). This corresponds to an overall 131.7% increase in NPs from 2017 to 2015 and an 88.4% increase in PAs from 2017 to 2025.

The number of NPs increased in all four major regions from 2017 to 2025: Northeast 19,589 (18.0%) to 43,957 (17.4%), Midwest 29,562 (27.1%) to 61,422 (24.3%), South 43,158 (39.6%) to 106,641 (42.2%), and West 16,777 (15.4%) to 40,658 (16.1%); all these regions demonstrated significant upward trends. The number of PAs increased in all four major regions from 2017 to 2025: Northeast 14,742 (22.5%) to 29,307 (23.7%), Midwest 14,806 (22.6%) to 26,757 (21.7%), South 22,164 (33.8%) to 41,454 (33.6%), and West 13,767 (21.0%) to 25,858 (21.0%); all these regions demonstrated significant upward trends. Some U.S. NPs and PAs, such as those in Puerto Rico, were not counted in this regional analysis.

The mean number of years since graduation for NPs decreased from 9.0 in 2017 to 8.6 in 2025, and the mean number of years since graduation for PAs decreased from 11.1 in 2017 to 10.8 in 2025, although neither trend was significant. The number of urban NPs increased from 87,661 (80.3%) in 2017 to 203,555 (80.5%) in 2025; the number of rural NPs increased from 21,439 (19.7%) in 2017 to 49,200 (19.5%). The number of urban PAs increased from 54,573 (83.3%) in 2017 to 101,672 (82.4%) in 2025; the number of rural PAs increased from 10,927 (16.7%) in 2017 to 21,747 in 2025 (17.6%). A chi-square test demonstrated a significant association between clinician type (e.g., PA versus NP) and urban versus rural practice location in 2025 (p-value <.001), between clinician type and gender in 2025 (p-value <.001), and between clinician type and region in 2025 (p-value <.001).

A state-level assessment of the percentage change from 2017 to 2025 in the number of NPs is shown in Figure 1, and a state-level assessment of the percentage change in number of PAs is shown in Figure 2. These choropleth graphs are centered on the same value of 100%, the average of the median percentage increase of NPs and the median percentage increase of PAs from 2017 to 2025. Individual state data for the percentage change in number of NPs and PAs can be found in Supplemental Digital Content 5, http://links.lww.com/JAANP/A456.

The median per-state percentage increase in the number of NPs was 117.9%, and the median per-state percentage increase in the number of PAs was 81.4%. The three states with the largest percentage increase in number of NPs were Nevada (230.1%), Florida (211.3%), and California (206.7%); the three states with the smallest percentage increase in number of NPs were Washington (79.5%), Wisconsin (79.6%), and Minnesota (81.3%). The three states with the largest percentage increase in number of PAs were New Jersey (180.9%), Missouri (158.2%), and Indiana (144.9%); the three states with the smallest percentage increase in number of PAs were New Mexico (28.0%), Kansas (37.3%), and South Dakota (38.8%).

Figure 3 shows the county-level clustering of high and low NP density in 2025, and Figure 4 shows the county-level clustering of high and low PA density in 2025.

Clusters of high NP density in 2025 are identified in Mississippi, Kentucky, and Arkansas. Clusters of low NP density in 2025 are identified in California, Nevada, Oregon, and Pennsylvania. Some states, such as Texas, have regions of both high and low NP density. Of the 331 counties in clusters of low NP density, 176 (53.2%) were rural. Of the 113 counties in clusters of high NP density, 99 (87.6%) were rural. Clusters of high PA density in 2025 are identified in Montana, Pennsylvania, and North Carolina. Clusters of low PA density in 2025 are identified in Alabama, Mississippi, Arkansas, Missouri, and Texas. Some states, such as Nebraska, have both regions of high and low PA density. Of the 457 counties in clusters of low PA density, 312 (68.3%) were rural. Of the 104 counties in clusters of high PA density, 56 (53.8%) were rural.

The mean 2025 PA density in states with “optimal” policies was 53.8, in states with “advanced” policies was 49.4, in states with “moderate” policies was 35.1, and in states with “reduced” policies was 28.4 (p-value <.001). The mean 2025 NP density in states with “full practice” policies was 87.4, in states with “reduced practice” policies was 90.2, and in states with “restricted practice” policies was 77.4 (p-value = .08).

Post hoc pairwise comparisons using Tukey HSD indicated that states with “optimal” policies had significantly higher PA density than states with “moderate” policies (mean difference = −18.8; 95% CI: −35.9 to −1.6; p-value = .03), and significantly higher PA density than states with “reduced” policies (mean difference = −25.4; 95% CI: −42.9 to −7.9; p-value = .002). States with “advanced” policies also had significantly higher PA density than states with “moderate” policies (mean difference = −14.3; 95% CI: −27.4 to −1.2; p-value = .03), and significantly higher PA density than states with “reduced” policies (mean difference = −21.0; 95% CI: −34.5 to −7.4; p-value = .001). The differences between “optimal” and “advanced” states were not statistically significant (p-value = .87), and the differences between “moderate” and “reduced” states were not statistically significant (p-value = .67).

Discussion

Our national analysis of NPs and PAs spanning 2017–2025 demonstrates a substantial upward trend in the number of Medicare NPs and PAs across the United States. For both Medicare NPs and PAs, the number of providers increased for both male and female genders, in both urban and rural areas, across all four major geographic regions, and in each of the 50 states in the United States. These findings reinforce literature documenting the rapid increases in the NP and PA workforce in recent years (Auerbach et al., 2018; Hooker et al., 2022; Xue et al., 2017, 2019). These providers could play an integral role in expanding health care access throughout the United States. However, we also noted substantial variations in the growth and distribution of NPs and PAs across different degrees of rurality, geographic regions, and demographic characteristics.

First, there were significant differences in the urban-rural distribution of NPs and PAs. For every year assessed, NPs were more likely to practice in rural areas than PAs. Furthermore, high-density NP clusters were more concentrated in rural counties than high-density PA clusters. These distinctions may reflect differences in NP versus PA program training pipelines, variations in scope of practice laws, and the higher prevalence of primary care versus specialty care in rural areas (Xue et al., 2020; D. Zhang, Lin, et al., 2020). Notably, the percentage of Medicare NPs and PAs that practice in rural areas is higher than the percentage of Medicare physicians that practice in rural areas reported in the literature (Crowley et al., 2025a, 2025b,2026). Thus, given the existing rural physician workforce shortages (Skinner et al., 2019) and substantial barriers to care that rural patient populations face (Douthit et al., 2015; Kozhimannil & Henning-Smith, 2021; Maganty et al., 2023), NPs and PAs could play a significant role in supplementing the physician population in these areas and improving health care access for this vulnerable population.

Second, our county-level mapping identifies substantial regional variation in NP and PA distribution. Regionally, NPs were more likely to practice in the South and Midwest than PAs, whereas PAs were more likely to practice in the West and Northeast than NPs. These regional differences persisted over the study period. Further, low-density NP clusters in 2025 were predominantly in states in the West, whereas low-density PA clusters in 2025 were predominantly in states in the South. These differences in regional distribution may also stem from differences in locations of NP versus PA program training pipelines, with many PA training programs in the Northeast (Forister & Stilp, 2017). The development of new training programs for PAs could focus on increasing training sites in the South, whereas the development of new training programs for NPs could focus on increasing training sites in the West to address the geospatial maldistribution of providers. We also found a statistically significant relationship between state-level restrictions for PA practice and state-level PA density, with states with fewer practice restrictions showing a trend toward higher PA densities. This could indicate that PAs are more inclined or find more opportunities to practice in states with fewer restrictions. NPs, in contrast, did not show a statistically significant relationship between state-level practice restrictions and state-level NP density. States with large rural populations such as Mississippi and Alabama could consider loosening their practice restrictions for NPs/PAs as one portion of an overall effort to encourage providers to practice in rural regions in the state. Future work should delve more deeply into how state-level and county-level factors influence practice patterns of NPs and PAs.

Third, there were demographic differences between NPs and PAs. NPs were more likely to be female than PAs at all time points assessed. However, the gap has been closing over the past decade as the percentage of female PAs increases, whereas the percentage of female NPs decreases.

This increase in female PA representation mirrors similar developments in female representation of physicians, where women now make up a majority of medical school students and an increasing percentage of the overall physician workforce (Lally et al., 2026). Physician associates were also found to have had a greater number of mean years since graduation than NPs, indicating that PAs in the United States on average may have more experience in their current provider role.

This study offers novel insights into current workforce demographic shifts by using descriptive and spatial methods. However, our approach is not without limitations. First, our datasets were derived from Medicare claims, which contains data on many NPs and PAs but exclude providers who do not care for Medicare patients. The AANP estimates the NP workforce as of 2025 comprises approximately 461,000 NPs (A Behind-the-Scenes Look at the 2025 Nurse Practitioner Count, 2025). The National Commission on Certification of Physician Assistants estimates that as of December 2024 there were 189,907 board-certified PAs (Statistical Profile of Board Certified Physician Assistants, 2025). Thus, we are likely capturing only approximately 55% and 56% of the current NP and PA workforce, respectively. Hence, our findings are not necessarily generalizable to the entire NP and PA workforce. This limitation, however, is also itself policy relevant: the gap between total workforce estimates and Medicare-participating providers suggests structural barriers to Medicare participation that warrant further investigation and may represent an actionable target for expanding Medicare beneficiary access. We also do not have access to data on the type of care that NPs and PAs are providing (e.g., primary care, oncology, etc.) or the scope of practice of these providers, which limits granular analysis of care patterns. Finally, we can describe existing differences in geographic distribution but cannot draw conclusions on why these differences persist. Thus, future work focused on gaining a greater understanding of the underlying factors associated with the geographic preferences of NPs and PAs will be essential to ensure that recruitment strategies help improve the accessibility of providers in underserved areas.

Our findings offer several concrete directions for federal and state policy. First, the percentage of Medicare NPs and PAs practicing in rural areas exceeds that of Medicare physicians reported in prior work (Crowley et al., 2025a, 2025b, 2026) indicating that these providers are already filling a critical access role in underserved areas. The substantial increase in the NP and PA workforce projected over the upcoming years, if distributed equitably towards the areas in greatest need of health care providers, has the potential to greatly expand access to care. To help foster improved access to care in rural or underserved areas, existing programs such as Medicare Health Professional Shortage Area physician bonuses could be expanded to include NPs and PAs (Physician Bonuses in Health Professional Shortage Areas. CMS, n.d.). In addition, our ANOVA findings demonstrated that PA density in states with “optimal” or “advanced” practice environments was approximately 18–25 providers per 100,000 higher than in states with “moderate” or “reduced” locations. The underlying reason for this geographic pattern, such as provider preference, employer hiring practices, state-level health care infrastructure, or regulatory differences, warrant further investigation, particularly in states with documented rural access gaps such as Mississippi, Alabama, and Arkansas.

Beyond policy changes, health care organizations and training institutions have complementary roles in addressing geographic workforce gaps. The development of additional rural training programs and creating supportive rural work environments with strong mentorship for NPs and PAs could help with recruiting and retaining these providers to practice alongside physicians in underserved areas (Coombs et al., 2011; Kaplan et al., 2020, 2023). Our county-level clustering analysis identifies specific regions where such training investment may have the greatest impact on workforce reach: low-density PA clusters were concentrated in Southern states including Alabama, Mississippi, Arkansas, Missouri, and Texas, whereas low-density NP clusters were concentrated in Western states including California, Nevada, Oregon, and Washington. Identifying and adapting approaches from successful outlier regions of high NP or PA density surrounded by regions of low NP or PA density, as observed in Mississippi, Kentucky, and Montana, could also offer additional insights into potential training and retention policy changes. In addition, health care organizations may benefit from investing in recruitment pipelines that draw students from underserved rural communities, given prior evidence that rural background and rural training are associated with later rural practice. Taken together these findings frame NP and PA workforce maldistribution as a health equity issue for rural Medicare beneficiaries. The growth of these workforces represents an opportunity to expand access, but only if distributed equitably toward areas of greatest need alongside the existing physician workforce.

Authors’ Contributions

R.J. Crowley: Conceptualization, Methodology, Software, Formal Analysis, Data Curation, Writing—Original Draft, Writing—Review and Editing, Visualization. J.S. Lally: Conceptualization, Methodology, Software, Formal Analysis, Data Curation, Writing—Original Draft, Writing—Review & Editing, Visualization. D.M. Kline: Conceptualization, Methodology, Formal Analysis, Writing—Original Draft, Writing—Review & Editing, Supervision. A.M. Bunting: Conceptualization, Methodology, Formal Analysis, Writing—Original Draft, Writing—Review and Editing, Supervision. R. J. Crowley and J. S. Lally contributed equally.

Competing Interests

The authors report no conflicts of interest.

References

• A Behind-the-Scenes Look at the 2025 Nurse Practitioner Count. (2025, November 11). American Association of Nurse Practitioners. https://www.aanp.org/news-feed/a-behind-the-scenes-look-at-the-2025-nurse-practitioner-count
• Auerbach D. I., Staiger D. O., Buerhaus P. I. (2018). Growing ranks of advanced practice clinicians—implications for the physician workforce. The New England Journal of Medicine, 378(25), 2358–2360. https://doi.org/10.1056/NEJMp1801869
• Barnes H., Richards M. R., McHugh M. D., Martsolf G. (2018). Rural and nonrural primary care physician practices increasingly rely on nurse practitioners. Health Affairs, 37(6), 908–914. https://doi.org/10.1377/hlthaff.2017.1158
• Coombs J. M., Morgan P., Pedersen D. M., Koduri S., Alder S. C. (2011). Factors associated with physician assistant practice in rural and primary care in Utah. International Journal of Family Medicine, 2011, 1–6. https://doi.org/10.1155/2011/879036
• Crowley R. J., Lally J. S., Kline D. M., Bunting A. M. (2026). Geographic disparities in access to oncologists and association with cancer outcomes in the United States. Journal of Cancer Policy, 47, 100704. https://doi.org/10.1016/j.jcpo.2026.100704
• Crowley R. J., Lally J. S., Kline D. M., Bunting A. M. (2025a). Geospatial analysis of telemedicine physicians in the United States. Telemedicine Journal and E-Health, 31(11), 1393–1397. https://doi.org/10.1089/tmj.2025.0160.
• Crowley R. J., Lally J. S., Kline D. M., Bunting A. M. (2025b). Urban-rural differences in the age of US physicians. The Journal of Rural Health, 41(3), e70054. https://doi.org/10.1111/jrh.70054
• Doctors and clinicians. (2025). Provider Data Catalog. https://data.cms.gov/provider-data/topics/doctors-clinicians
• Douthit N., Kiv S., Dwolatzky T., Biswas S. (2015). Exposing some important barriers to health care access in the rural USA. Public Health, 129(6), 611–620. https://doi.org/10.1016/j.puhe.2015.04.001
• Forister J. G., Stilp C. (2017). A spatial analysis of physician assistant programs. Journal of Physician Assistant Education, 28(2), 64–71. https://doi.org/10.1097/JPA.0000000000000120
• Global Perspective on U.S. (2023). Health care | commonwealth fund. https://www.commonwealthfund.org/publications/issue-briefs/2023/jan/us-health-care-global-perspective-2022
• Gommers R., Virtanen P., Haberland M., Burovski E., Reddy T., Weckesser W., Oliphant T. E., Cournapeau D., Nelson A., Roy P., Peterson P., Polat I., Wilson J., Mayorov N., van der Walt S., Colley L., Brett M., Striega K. (2025). scipy/scipy: SciPy 1.15.0. Zenodo. https://doi.org/10.5281/zenodo.14593523
• Hacker K. (2024). The burden of chronic disease. Mayo Clinic Proceedings. Innovations, Quality & Outcomes, 8(1), 112–119. https://doi.org/10.1016/j.mayocpiqo.2023.08.005
• Hooker R. S., Christian R. L. (2023). The changing employment of physicians, nurse practitioners, and physician associates/assistants. Journal of the American Association of Nurse Practitioners, 35(8), 487–493. https://doi.org/10.1097/JXX.0000000000000917
• Hooker R. S., Kulo V., Kayingo G., Jun H.-J., Cawley J. F. (2022). Forecasting the physician assistant/associate workforce: 2020-2035. Future Healthcare Journal, 9(1), 57–63. https://doi.org/10.7861/fhj.2021-0193
• Hussain Md., Mahmud I. (2019). pyMannKendall: A python package for non parametric Mann Kendall family of trend tests. Journal of Open Source Software, 4(39), 1556. https://doi.org/10.21105/joss.01556
• Kaplan L., Pollack S., Skillman S., Patterson D. (2020). Factors that encourage and support advanced practice registered nurses to work in rural and safety-net settings. Journal of Health Care for the Poor and Underserved, 31(4S), 163–181. https://doi.org/10.1353/hpu.2020.0148
• Kaplan L., Pollack S. W., Skillman S. M., Patterson D. G. (2023). Is being there enough? Postgraduate nurse practitioner residencies in rural primary care. The Journal of Rural Health, 39(3), 529–534. https://doi.org/10.1111/jrh.12729
• Kozhimannil K. B., Henning-Smith C. (2021). Improving health among rural residents in the US. JAMA, 325(11), 1033–1034. https://doi.org/10.1001/jama.2020.26372
• Lally J. S., Crowley R. J., Mann K., Kline D. M., Bunting A. M. (2026). Representation of female physicians in the United States. Academic Medicine. doi:10.1093/acamed/wvag045.
• Machado S. R., Jayawardana S., Mossialos E., Vaduganathan M. (2021). Physician density by specialty type in urban and rural counties in the US, 2010 to 2017. JAMA Network Open, 4(1), e2033994. https://doi.org/10.1001/jamanetworkopen.2020.33994
• Maganty A., Byrnes M. E., Hamm M., Wasilko R., Sabik L. M., Davies B. J., Jacobs B. L. (2023). Barriers to rural health care from the provider perspective. Rural and Remote Health, 23(2), 7769. https://doi.org/10.22605/RRH7769
• PA State Practice Environment. (2025). AAPA. https://www.aapa.org/advocacy-central/state-advocacy/state-maps/pa-state-practice-environment/
• Parsons H. M., Enewold L. R., Banks R., Barrett M. J., Warren J. L. (2017). Creating a National Provider Identifier (NPI) to Unique Physician Identification Number (UPIN) crosswalk for medicare data. Medical Care, 55(12), e113–e119. https://doi.org/10.1097/MLR.0000000000000462
• Physician Bonuses in Health Professional Shortage Areas (n.d.). CMS. https://www.cms.gov/medicare/payment/fee-for-service-providers/physician-bonuses-health-professional-shortage-areas-hpsas
• Rey S. J., Anselin L. (2009). PySAL: A Python library of spatial analytical methods. In Handbook of applied spatial analysis: Software tools, methods and applications (pp. 175–193). Springer.
• Rhudy C., Schadler A., Talbert J. C. (2020). Rural/urban disparities in utilization of diabetes self-management training to the fee-for-service medicare population.
• Rural-Urban Continuum Codes (n.d.). Economic Research Service. Retrieved February 6, 2025, from https://www.ers.usda.gov/data-products/rural-urban-continuum-codes
• Schmitz K. H., Demanelis K., Crisafio M. E., Kennedy M. A., Schwartz A. L., Campbell A., Gorzelitz J., Wood K. C., Wilson C. M., Scalise R. L., Vincent A. (2025). Proximity to cancer rehabilitation and exercise oncology by geography, race, and socioeconomic status. Cancer, 131(1), e35515. https://doi.org/10.1002/cncr.35515
• Shen X., Kane K., Katz A. J., Usinger D., Cao Y., Chen R. C. (2024). Differences in rural versus urban patients with prostate cancer in diagnosis and treatment: An analysis of a population-based cohort. JCO Oncology Practice, 20(8), 1109–1114. https://doi.org/10.1200/OP.23.00547
• Skinner L., Staiger D. O., Auerbach D. I., Buerhaus P. I. (2019). Implications of an aging rural physician workforce. The New England Journal of Medicine, 381(4), 299–301. https://doi.org/10.1056/NEJMp1900808
• State Practice Environment. (2025). American Association of Nurse Practitioners. https://www.aanp.org/advocacy/state/state-practice-environment
• Statistical Profile of Board Certified Physician Assistants (2025). https://www.nccpa.net/wp-content/uploads/2025/05/2024-Statistical-Profile-of-Board-Certified-PAs.pdf
• The Demographic Outlook: 2025 to 2055 (2025, January 13). Congressional Budget Office. https://www.cbo.gov/publication/60875
• Walensky R. P., McCann N. C. (2025). Challenges to the future of a robust physician workforce in the United States. The New England Journal of Medicine, 392(3), 286–295. https://doi.org/10.1056/NEJMsr2412784
• Xue Y., Goodwin J. S., Adhikari D., Raji M. A., Kuo Y.-F. (2017). Trends in primary care provision to medicare beneficiaries by physicians, nurse practitioners, or physician assistants: 2008-2014. Journal of Primary Care & Community Health, 8(4), 256–263. https://doi.org/10.1177/2150131917736634
• Xue Y., Mullaney T., Smith B., Cai X., Spetz J. (2020). Scope-of-Practice regulation and nurse practitioners as usual source of care providers. Journal of Nursing Regulation, 11(3), 13–21. https://doi.org/10.1016/S2155-8256(20)30130-7
• Xue Y., Poghosyan L., Lin Q. (2024). Supply and geographic distribution of geriatric physicians and geriatric nurse practitioners. JAMA Network Open, 7(11), e2444659. https://doi.org/10.1001/jamanetworkopen.2024.44659
• Xue Y., Smith J. A., Spetz J. (2019). Primary care nurse practitioners and physicians in low-income and rural areas, 2010-2016. JAMA, 321(1), 102–105. https://doi.org/10.1001/jama.2018.17944
• Young P. A., Griffith C. F., Vasquez R. (2022). Geographic distribution and workforce diversity of dermatology physician assistants. JAMA Dermatology, 158(12), 1460. https://doi.org/10.1001/jamadermatol.2022.4453
• Zhang D., Son H., Shen Y., Chen Z., Rajbhandari-Thapa J., Li Y., Eom H., Bu D., Mu L., Li G., Pagán J. A. (2020). Assessment of changes in rural and urban primary care workforce in the United States from 2009 to 2017. JAMA Network Open, 3(10), e2022914. https://doi.org/10.1001/jamanetworkopen.2020.22914
• Zhang X., Lin D., Pforsich H., Lin V. W. (2020). Physician workforce in the United States of America: Forecasting nationwide shortages. Human Resources for Health, 18(1), 8. https://doi.org/10.1186/s12960-020-0448-3

View full references list

Keywords:

Health workforce; Medicare; nurse practitioners; physician assistants; physician associates; rural health

Supplemental Digital Content