Office Procedures for Older Adults by Physician Associates and Nurse Practitioners

ABSTRACT

Objective: We hypothesized that physician associate (PA) and nurse practitioner (NP) procedural roles are expanding. We sought to describe ambulatory procedures these professionals performed in 2021 for older adults.

Study Design: Retrospective observational cohort study of Medicare Part B data. US Bureau of Labor Statistics data were used to provide overall PA and NP employment context.

Methods: Medicare Part B databases were probed for outpatient events by PAs and NPs using a modified list of the Council of Academic Family Medicine’s recommended clinical procedures that focused on 29 procedures organized into 9 categories called procedure clusters. These procedures were linked to Current Procedural Terminology codes and PA and NP National Provider Identifier codes in Medicare Part B and then tabulated and analyzed for 2021. The Bureau of Labor Statistics provided NP and PA employment trends for context. The trend of the procedures and providers spanning 2014-2021 was analyzed.

Results: In 2021, 23,581 NPs and PAs filed 9.6 million Medicare Part B enrollee procedure claims. Most procedures (96%) involved skin or the musculoskeletal system. PAs filed more than twice as many claims for skin and musculoskeletal procedures as NPs, and NPs filed 1.25 times as many as PAs for the eye, ear, nose, and throat; pulmonary; genitourinary; gastrointestinal-colorectal; and women’s health categories. From 2014 through 2021, the number of PAs and NPs in clinical practice increased by 72%, and the number of those who filed procedure claims increased by 74%.

Conclusions: Overall, PAs performed more skin and musculoskeletal procedures than NPs, and NPs performed more procedures in the other 7 procedure clusters than PAs. PA and NP employment growth does not fully explain these observations. We suggest that outpatient procedural task-shifting activity presents an area for further research.

Am J Manag Care. 2024;30(4):e109-e115. https://doi.org/10.37765/ajmc.2024.89532

Takeaway Points

Understanding clinical procedures commonly performed by physician associates (PAs) and nurse practitioners (NPs) for older patients may help inform staffing and education decisions relevant to the optimal utilization of these professionals in outpatient settings.

• In 2021, 96% of procedures performed by PAs and NPs for Medicare patients were in the skin or musculoskeletal categories.
• Nearly twice as many PAs performed the most common skin and musculoskeletal procedures compared with NPs.
• The 74% growth in the number of PAs and NPs filing procedure claims from 2014 through 2021 is consistent with, but not fully explained by, the 72% growth of these professions during that period.

Clinical procedures are the mental and motor skills a health professional needs to execute a patient-care task in medical and surgical practice. Although these procedures are an undertaking in most office settings, what is known about the type and range of procedures performed by physician associates (PAs) and nurse practitioners (NPs) is limited. In 1982, Larsen and Kirkwood compared office procedures by PAs vs their employing physicians (general internists) and found the range of procedures reported was similar.¹ When investigators in another study compared office-based care between Iowa primary care PAs and physicians, little difference was seen between 62 primary care PAs and associated physicians in the type and rate of procedures recorded.² In another study, responses from 102 surveyed NPs produced a list of the most frequently reported clinical procedures in their settings. The authors noted that 7 “clinical procedure skills” were completed at least once a week.³

There are several reasons to examine procedures performed at the clinical level. Office-based procedures are considered essential elements in outpatient care and service delivery by the National Center for Health Statistics.^4,5 Minor surgical procedures such as excisions, suturing, and musculoskeletal joint injections are good examples of medical activities that may be performed within hospitals or by community-based specialists but are typically outpatient undertakings.⁶

Ambulatory-based care and clinical procedures have emerged as important for several reasons. The value of office-based procedures is that they can save the patient an unnecessary referral, which may result in the loss of care continuity.⁷ Certain procedures also may be relegated to individual clinicians with greater interest and experience, thus freeing up other team members for different roles.^8,9

The type and nature of office-based procedures are evolving, such as using skin glue instead of sutures. Knowing how and when to perform a procedure is essential to clinical medicine. However, the limited literature on procedural activities involving PAs and NPs suggests this topic needs to be addressed in modeling primary care capabilities. The reduction of outpatient ambulatory procedures performed by physicians in older adults has affected clinical outcomes.¹⁰ The primary care outpatient procedures commonly performed by PAs and NPs in office settings have yet to be noted, particularly for older adults. Our objective in investigating the range of procedures reported by NPs and PAs is to set the stage for further work on how they can be optimally utilized to address outpatient service delivery.^11,12

Nearly 17% of Americans are 65 years or older, and adults in this age range are projected to be 20% of the US population by 2030.¹³ Concurrently, the number of physicians per capita is declining. As the number of PAs and NPs increases, we hypothesize that the outpatient procedures they perform for older patients will rise accordingly. Our undertaking intends to identify the current skills of a diversifying medical workforce and inform educators about which procedures to emphasize for the optimal employment of new clinicians. The growing health care needs of an older population make this topic germane.

METHODS

We used clinical tasks identified by the Council of Academic Family Medicine (CAFM) as essential primary care procedures.¹⁴ In 2014, the CAFM recommended a list of 68 core procedures that family physicians should be able to perform upon completing their residency.⁷ This list was organized into 13 therapeutic categories called procedure clusters.¹⁴

We matched the CAFM list of recommended procedures with the appropriate Current Procedural Terminology (CPT) codes.¹⁵ To do this, the CAFM list was modified to remove procedures not relevant to patients 65 years or older (eg, intrauterine device insertion, neonatal umbilical vein cannulation), procedures lacking a corresponding CPT code (eg, digital rectal examination), and procedures unlikely to be performed in an outpatient setting (eg, central venous cannulation). Next, we merged some CAFM-recommended procedures to better match how CPT codes are organized. For example, the CAFM listed laceration repair with tissue glue or skin staples as one procedure and simple laceration repair with sutures as another. The CPT codes related to wound repair do not describe how the wound was repaired but instead categorize these procedures by complexity, location, and size. Thus, we combined these CAFM-recommended procedures into a modified procedure called “simple, intermediate, and complex laceration repair with sutures, tissue glue, or staples.” The CPT codes were matched to the modified CAFM (mCAFM) list and linked with National Provider Identifiers (NPIs) to probe the Medicare database for outpatient procedure claims filed by PAs and NPs.

The Medicare Part B database was selected because it provides information on outpatient services and procedures provided to its beneficiaries by physicians, PAs, NPs, and other health care professionals. Data are available only for those covered under the original fee-for-service Medicare Part B and do not include beneficiaries of Medicare Advantage plans. The data set contains use, payments, and submitted charges organized by NPI, Healthcare Common Procedure Coding System code (which mirrors CPT codes), type of setting, and geography.¹⁶ Medicare policy limits public data to providers who have performed more than 10 of any specific procedures that year. Furthermore, Medicare does not identify the medical or surgical specialty or employment role of PAs and NPs as it does for physicians; thus, the analysis could not stratify PAs and NPs by their functional therapeutic specialty.

The year 2021 was purposely selected for comparing procedures between providers because it was the first full year after the beginning of the COVID-19 pandemic, a period that dampened national outpatient office visits and reporting.¹⁷ Descriptive statistics were applied, and comparisons were made between PAs and NPs using Medicare Part B procedure claims and PA and NP employment numbers from the US Bureau of Labor Statistics.¹⁸ Trends on PAs and NPs submitting Medicare Part B claims were calculated for 2014-2022.

All analyses were conducted with Excel, Python, and Tableau analysis systems. Because deidentified publicly available data were used, the Marshall B. Ketchum University Institutional Review Board classified this cross-sectional, retrospective cohort study as deidentified participant research and exempt from further review.

RESULTS

After the adjustments described in the Methods section were completed, the mCAFM list contained 29 procedure types organized into 9 categories linked to 140 CPT codes (Table 1). When the mCAFM codes linked to the 2021 Medicare Part B database for outpatient settings were stratified by PAs and NPs, 9,649,285 mCAFM procedure claims (approximately 6.7 million procedure claims by PAs and 2.9 million by NPs) were filed (Table 2). These claims ranged from 0 cardiovascular procedures by PAs and NPs to 8.0 million skin procedures by 4651 PAs and 2557 NPs. The procedure claims filed most frequently—making up 99% of all mCAFM procedure claims—were in the skin (84%); musculoskeletal (12%); and eye, ear, nose, and throat (EENT) (3%) categories. The procedures that the largest number of PAs and NPs claimed differed from the types of procedures performed most frequently. Although the most commonly performed procedures were in the skin category, these claims were submitted by 3261 fewer PAs and NPs than the number who submitted musculoskeletal claims (Figure 1). In the musculoskeletal category, 7756 PAs and 2713 NPs performed more than 10 procedures compared with 4651 PAs and 2557 NPs who submitted claims for skin procedures.

We compared the most frequently performed procedure categories by PAs or NPs in 2021. In the skin category, PAs filed 5.6 million and NPs filed 2.4 million procedure claims; in the musculoskeletal category, PAs and NPs filed 877,046 and 314,536 procedure claims, respectively. In the third most common category, EENT procedures, PAs filed 132,152 and NPs filed 123,428 procedure claims. The number of procedures in the 6 remaining categories (anesthesia, cardiovascular, gastrointestinal-colorectal, genitourinary, pulmonary, and women’s health) comprised 1.26% of the total. The median number of all Medicare mCAFM procedure claims filed in 2021 was 60 per PA (range, 11-7611) and 48 per NP (range, 11-8160).

We conducted a trend analysis of Medicare Part B mCAFM procedures by NPs and PAs from 2014 through 2021. The number of procedures in each category increased annually from 2014 to 2019, and then a downtick occurred in 2020, followed by an uptick in 2021. From 2014 through 2021, the number of PAs who filed an mCAFM claim increased by 67%, and the number of NPs who did so increased by 87% (Figure 2).

To better understand the context of office-based mCAFM procedures, we assessed the total number of PAs and NPs filing a claim for any CPT code. In 2021, a total of 82,399 individual PAs and 149,911 NPs filed more than 10 claims in Medicare Part B. The percentage of PAs and NPs filing mCAFM procedure claims was 18% and 6%, respectively, based on the total number of PAs and NPs filing a claim for any CPT code (Figure 3). Combined, of all PAs and NPs who submitted any Medicare claims, 10% of both professions filed mCAFM claims.

DISCUSSION

Medicare served more than 58 million beneficiaries in 2021.¹⁹ Of these, 53% were enrolled in the traditional fee-for-service Part B.¹⁹ At the same time, an estimated 132,940 PAs and 234,690 NPs were clinically employed.²⁰ To assess the office-based procedural activities of PAs and NPs within the increasing older population, our cohort study used Medicare Part B data for 2021. We identified the range of mCAFM procedure claims filed by PAs and NPs, and we found that 99% of the claims in 2021 were related to the skin, musculoskeletal, and EENT categories and were filed by 21,461 NPs and PAs. The number of PAs filing claims for mCAFM codes was 1.7 times greater than that of NPs, although there were 1.7 times as many NPs in clinical practice as PAs. Overall, 6% of all clinically active members of both professions combined, including those who did not submit any Medicare claims under their own NPI number, filed claims for mCAFM codes. Between 2014 and 2021, the 87% increase in NPs filing claims for mCAFM codes was consistent with the 92% growth of NPs in clinical practice. For PAs, the 67% increase in mCAFM filings exceeded the 45% growth of the profession. This suggests that both professions are increasingly providing procedural care for older adults in the outpatient setting.

The use of PAs and NPs in outpatient clinics has been underway since their introduction to American society in the mid-1960s.²¹ Most importantly, their ability to competently perform in-office procedures, such as suturing and casting, has been only marginally described in contemporary literature.^9,22 We found that the increase in the number of procedure claims filed and the number of PAs and NPs filing them is consistent with, but not fully explained by, the growth of both professions. Thus, the growth in the number of NPs submitting mCAFM claims is consistent with the increase in their number in clinical practice, whereas the growth of PAs doing so was greater. This could be due to more NPs and PAs submitting claims to Medicare under their own NPI number, increased procedural activity by these clinicians, or a combination of the two.

Although there were 1.7 times as many NPs as PAs in clinical practice in 2021, the number of PAs filing mCAFM claims and the number of claims filed in the 2 most common categories (skin and musculoskeletal) were at least 2.35 times more than the number of NPs who did so. In the remaining categories, which comprised approximately 4% of all procedures, 1.25 times more NPs filed claims than PAs, consistent with their relative numbers in clinical practice. This may reflect differences in how each profession is initially trained. Another interpretation is that PAs may be more procedurally oriented than NPs and/or are more likely to submit Medicare claims under their own NPI, or some blend of these 2 dynamics. The growth we found of NPs and PAs who independently billed Medicare for services is consistent with similar observations of common dermatologic services by Adamson et al.²³

Compared with NPs and PAs filing any CPT claim, the proportion filing mCAFM procedure claims was 18% for PAs, 6% for NPs, and 10% for both professions combined. These percentages likely reflect the reality that office-based procedures comprise a much smaller proportion of all outpatient CPT codes, which include the myriad of evaluation and management procedures and other tasks not identified by the CAFM as primary care clinical procedures (eg, 15-minute office visits, venipuncture, vaccine administration).

The primary care outpatient procedural skills that clinically active NPs and PAs perform are revealed in this report. The American and Canadian educators of family medicine physicians have recommended that select procedure skills be learned by graduation.^14,24 The Accreditation Review Commission on Education for the Physician Assistant requires that PA school curricula “…must include instruction in technical skills and procedures based on current clinical practice.”²⁵ Understanding what procedures are needed in outpatient clinical settings is essential because PAs and most NPs are schooled in general or family medicine.²⁶

These findings have several implications. The first is the growing importance of NPs and PAs in caring for all patients, especially older ones.^27,28 Because team-based care is important in primary care, the findings of this study suggest that a division of labor may exist.^29,30 Another inference is the accelerating number of NPs and PAs needing to take on more complex roles in patient management. The expansion of PA and NP roles is based on the forecast of physician and surgeon shortages in the next decade.³¹

Because outpatient procedural services are growing among the older population, as are the number of PAs and NPs who perform them, the stage is set for a closer examination of the types of procedures, their appropriateness and value, and the characteristics of patients and proceduralists. As observers of this phenomenon, we would like to know whether a division of labor is underway that could improve service delivery outcomes. Questions about relative value units, the range and appropriateness of procedures in various settings, and liability issues are growing topics for exploration. These questions are particularly important given the emergence of low-value care as a focus in the literature.^32,33

Limitations

There are several limitations to this analysis. First, the historical “incident to” reimbursement policy, for which a physician submits a claim in their name for procedures done by PAs and NPs (under their direction), may mask the complete assignment of the proceduralist, suggesting that the magnitudes of NP and PA claims reported here are conservative. Conversely, Medicare Part B does not permit split or shared billing.

A second limitation is that the publicly available Medicare claims data exclude beneficiaries covered under Medicare Advantage plans, which in 2021 accounted for nearly 47% of all Medicare beneficiaries.¹⁹ It is possible this omission could have skewed our results, but that would presume clinicians alter their clinical practice based on the Medicare insurance coverage of each patient.

Third, publicly available Medicare data do not report the practice site where the encounter occurred. Instead, the data indicate only whether the patient encounter was in an outpatient or inpatient setting.

Finally, because CMS does not identify the functional specialties of PAs and NPs, we could not determine whether they were in primary care practices. Instead, we captured all PAs and NPs who submitted more than 10 mCAFM claims. Thus, this is an analysis of primary care procedures reported by NPs and PAs, not activities in primary care practices. In 2022, Medicare policy for PA and NP reimbursement changed and now includes specialty assignments, which may mitigate this limitation for future studies. That improved categorization of the clinician’s role will sharpen the comparison of encounters in primary care and other settings along with patient demographics.¹⁶

We suggest that the stage is set for the triad assessment of procedures by PAs, NPs, and physicians across populations, specialties, and settings. There are clear trends underway in Medicare claims involving PAs and NPs. The next step in health workforce research will be to explore ambulatory procedural activity across all age groups and populations.

CONCLUSIONS

In 2021, nearly 133,000 PAs and 235,000 NPs were clinically employed. At least 10% of them filed nearly 9.6 million Medicare Part B CPT claims for mCAFM procedures in ambulatory care settings under their NPI. The types of procedures were primarily in the skin and musculoskeletal categories, accounting for 96% of all procedures. The number of PAs filing claims for those procedures and the number of claims filed were at least 2.35 times greater than those for NPs. The number of procedure claims filed by NPs and PAs is consistent with or greater than the growth of both professions. Furthermore, the workforce participation of NPs and PAs is growing, consistent with population growth. With a surge in service demand for Medicare patients, PAs and NPs appear to be involved with ever greater visibility. Ambulatory care procedures is an understudied area of health services research.

Acknowledgments

The authors acknowledge Jason Culotta, a health policy analyst at CMS, for his assistance with this manuscript. Mr Culotta is the director of the Division of Outcomes Measurement within the Data Analytics and Systems Group in the CMS Center for Program Integrity.

Author Affiliations: School of Physician Assistant Studies, Marshall B. Ketchum University (REM), Fullerton, CA (ORCID: 0009-0008-9392-5091); retired health policy analyst (RSH), Ridgefield, WA (ORCID: 0000-0002-8416-8673); independent analyst (RC), Portland, OR.

Source of Funding: This work was supported by a small research grant to Dr McKenna from the American Academy of Physician Associates in 2022.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (REM, RSH, RC); acquisition of data (RC); analysis and interpretation of data (REM, RSH, RC); drafting of the manuscript (REM, RSH, RC); critical revision of the manuscript for important intellectual content (REM, RC); statistical analysis (RSH, RC); provision of patients or study materials (RC); obtaining funding (REM, RC); administrative, technical, or logistic support (REM, RSH, RC); and supervision (REM, RC).

Address Correspondence to: Robert E. McKenna, DMSc, MPH, PA-C, School of Physician Assistant Studies, Marshall B. Ketchum University, 2575 Yorba Linda Blvd, Fullerton, CA 92831. Email: remckenna01@gmail.com.

REFERENCES

1. Larsen KM, Kirkwood CR. Procedures performed by family physicians, internists, and a Medex in a small group practice. J Fam Pract. 1982;15(2):285-292.

2. Dehn RW, Hooker RS. Implications for practice and education: clinical activities of Iowa family practice PAs. JAAPA. 1999;12(4):63-64, 67-69, 73-74 passim.

3. Want DR, Goodman P, Selway J. Primary care nurse practitioner clinical procedural skills. J Nurse Pract. 2022;18(2):200-204. doi:10.1016/j.nurpra.2021.10.004

4. Lau DT, McCaig LF, Hing E. Toward a more complete picture of outpatient, office-based health care in the U.S. Am J Prev Med. 2016;51(3):403-409. doi:10.1016/j.amepre.2016.02.028

5. Urman RD, Punwani N, Shapiro FE. Office-based surgical and medical procedures: educational gaps. Ochsner J. 2012;12(4):383-388.

6. Clebak KT, Reedy-Cooper A, eds. Office-Based Procedures: Part II, An Issue of Primary Care: Clinics in Office Practice. Vol 49. Elsevier Health Sciences; 2022.

7. Nothnagle M, Sicilia JM, Forman S, et al; STFM Group on Hospital Medicine and Procedural Training. Required procedural training in family medicine residency: a consensus statement. Fam Med. 2008;40(4):248-252.

8. Poulin EA, Swartz AW, O’Grady JS, Kersten MPM, Angstman KB. Essential office procedures for Medicare patients in primary care: comparison with family medicine residency training. Fam Med. 2019;51(7):574-577. doi:10.22454/FamMed.2019.659478

9. Heikel TE, Lighthall JG. Utilizing physician assistants for in-office procedures in outpatient otolaryngology clinic. J Otolaryngol Rhinol. 2021;7(4):111. doi:10.23937/2572-4193.1510111

10. Moonesinghe SR, Lowery J, Shahi N, Millen A, Beard JD. Impact of reduction in working hours for doctors in training on postgraduate medical education and patients’ outcomes: systematic review. BMJ. 2011;342:d1580. doi:10.1136/bmj.d1580

11. National Academies of Sciences, Engineering, and Medicine. Implementing High-Quality Primary Care: Rebuilding the Foundation of Health Care. The National Academies Press; 2021. doi:10.17226/25983

12. Grumbach K. Forging a social movement to dismantle entrenched power and liberate primary care as a common good. Ann Fam Med. 2023;21(2):180-184. doi:10.1370/afm.2950

13. America’s Health Rankings Annual Report 2022. United Health Foundation; 2023. Accessed October 10, 2023. https://assets.americashealthrankings.org/app/uploads/ahr_2022annualreport.pdf

14. CAFM consensus statement for procedural training in family medicine residency. Association of Family Medicine Residency Directors. 2014. Accessed June 21, 2022. https://www.afmrd.org/page/cafm

15. CPT 2019 Professional Edition. 5th ed. American Medical Association; 2018.

16. Browse by category. CMS. Accessed July 14, 2023. https://data.cms.gov/browse-data-categories

17. Alexander GC, Tajanlangit M, Heyward J, Mansour O, Qato DM, Stafford RS. Use and content of primary care office-based vs telemedicine care visits during the COVID-19 pandemic in the US. JAMA Netw Open. 2020;3(10):e2021476. doi:10.1001/jamanetworkopen.2020.21476

18. Occupational employment projections data. US Bureau of Labor Statistics. Updated January 26, 2024. Accessed February 8, 2024. https://www.bls.gov/emp/data/occupational-data.htm

19. Ochieng N, Clerveau G, Cubanski J, Neuman T. A snapshot of sources of coverage among Medicare beneficiaries. KFF. August 14, 2023. Updated December 13, 2023. Accessed February 8, 2024.
https://www.kff.org/medicare/issue-brief/a-snapshot-of-sources-of-coverage-among-medicare-beneficiaries

20. Tables created by BLS. US Bureau of Labor Statistics. Accessed February 12, 2024. https://www.bls.gov/oes/tables.htm

21. Cawley JF, Cawthon E, Hooker RS. Origins of the physician assistant movement in the United States. JAAPA. 2012;25(12):36-40, 42. doi:10.1097/01720610-201212000-00008

22. Hart AM, Bowen A. New nurse practitioners’ perceptions of preparedness for and transition into practice. J Nurse Pract. 2016;12(8):545-552. doi:10.1016/j.nurpra.2016.04.018

23. Adamson AS, Suarez EA, McDaniel P, Leiphart PA, Zeitany A, Kirby JS. Geographic distribution of nonphysician clinicians who independently billed Medicare for common dermatologic services in 2014. JAMA Dermatol. 2018;154(1):30-36. doi:10.1001/jamadermatol.2017.5039

24. Battaglia F, Sayed C, Merlano M, et al. Identifying essential procedural skills in Canadian undergraduate medical education. Can Med Educ J. 2021;12(5):e17-e23. doi:10.36834/cmej.68494

25. Accreditation Standards for Physician Assistant Education. 5th ed. Accreditation Review Commission on Education for the Physician Assistant Inc; 2019. Updated September 2023. Accessed October 10, 2023. https://www.arc-pa.org/wp-content/uploads/2023/10/Standards-5th-Ed-September-2023.pdf

26. Mabee J, Tramel J, Lie D. Current status of procedural skills training in physician assistant programs in the United States. J Physician Assist Educ. 2014;25(4):4-11. doi:10.1097/01367895-201425040-00001

27. Everett CM, Docherty SL, Matheson E, et al. Teaming up in primary care: membership boundaries, interdependence, and coordination. JAAPA. 2022;35(2):1-10. doi:10.1097/01.JAA.0000805840.00477.58

28. Li Y, Cimiotti JP, Evans KA, Clevenger CK. The characteristics and practice proficiency of nurse practitioners who care for older adults. Geriatr Nurs. 2022;46:213-217. doi:10.1016/j.gerinurse.2022.01.016

29. Pany MJ, Chen L, Sheridan B, Huckman RS. Provider teams outperform solo providers in managing chronic diseases and could improve the value of care. Health Aff (Millwood). 2021;40(3):435-444. doi:10.1377/hlthaff.2020.01580

30. Morgan PA, Smith VA, Berkowitz TSZ, et al. Impact of physicians, nurse practitioners, and physician assistants on utilization and costs for complex patients. Health Aff (Millwood). 2019;38(6):1028-1036. doi:10.1377/hlthaff.2019.00014

31. IHS Markit Ltd. The Complexities of Physician Supply and Demand: Projections From 2019 to 2034. Association of American Medical Colleges; 2021. Accessed March 2, 2022. https://www.aamc.org/media/54681/download

32. Razavi M, O’Reilly-Jacob M, Perloff J, Buerhaus P. Drivers of cost differences between nurse practitioner and physician attributed Medicare beneficiaries. Med Care. 2021;59(2):177-184. doi:10.1097/MLR.0000000000001477

33. O’Reilly-Jacob M, Perloff J, Buerhaus P. Low-value back imaging in the care of Medicare beneficiaries: a comparison of nurse practitioners and physician assistants. Med Care Res Rev. 2021;78(3):197-207. doi:10.1177/1077558719877796