This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (30) Citing Articles via Google Scholar Google Scholar Articles by Hooker, R. S. Articles by McCaig, L. F. Search for Related Content PubMed PubMed Citation Articles by Hooker, R. S. Articles by McCaig, L. F. Related Collections Workforce Issues

DataWatch

Use Of Physician Assistants And Nurse Practitioners In Primary Care, 1995–1999

Abstract

 
Federal policies and state legislation encourage the use of physician assistants (PAs) and nurse practitioners (NPs) in primary care, although the nature of their work has not been fully analyzed. In this paper we analyze primary care physician office encounter data from the 1995–1999 National Ambulatory Medical Care Surveys. About one-quarter of primary care office based physicians used PAs and/or NPs for an average of 11 percent of visits. The mean age of patients seen by physicians was greater than that for PAs or NPs. NPs provided counseling/education during a higher proportion of visits than did PAs or physicians. Overall, this study suggests that PAs and NPs are providing primary care in a way that is similar to physician care.

In 1999 an estimated 757 million visits were made to physicians’ offices in the United States. PAs and NPs were seen at 1.8 and 1.1 percent of these visits, respectively.⁴ Approximately 90 percent of all NP graduates (48,000) and 50 percent of all PA graduates (21,000) work in primary care (defined here as internal medicine, general and family practice, pediatrics, and obstetrics/gynecology).⁵ Differences exist in the training of PAs and NPs and in the state regulations regarding their use—PAs are dependent on physicians as supervisors in all states, and NPs have the option of practicing independently in some states. However, the basic clinical tasks that they perform do not differ greatly in most primary care settings.⁶

We examined data from the 1995–1999 National Ambulatory Medical Care Surveys (NAMCS) to describe visits to primary care office-based physicians in which the patient was seen by a PA, an NP, or a physician. Since the NAMCS is the only national probability sample survey that collects data on the provider seen, it offers a unique opportunity to evaluate the role of PAs and NPs in physicians’ offices nationwide.

Methods

Top Methods Study Results Discussion NOTES

 
The NAMCS is conducted by the Centers for Disease Control and Prevention’s (CDC’s)National Center for Health Statistics. The survey’s target universe includes visits made in the United States to nonfederally employed physicians who are in office-based patient care. The NAMCS uses a three-stage probability sample design involving geographic primary sampling units (PSUs), physician practices within PSUs, and patient visits within physician practices. Sample physicians, selected from the American Medical Association and the American Osteopathic Association Master files, were asked to complete approximately thirty patient record forms for a systematic random sample of office visits occurring during a randomly assigned one-week period. The number of physicians eligible to participate in the survey during the five-year study period was 10,064; they completed 135,494 patient record forms. The average physician response rate was 69 percent. Five categories of visits based on the provider(s) seen were defined as follows: PA, NP, or physician only; PA orNP without a physician; and PA orNP with a physician.

Data from the NAMCS were weighted to produce average annual national estimates for both physicians and office visits. Five years of data were combined to provide more reliable estimates. Three years of data (1997–1999) were combined to produce estimates for the total number of diagnostic/screening services and therapeutic and preventive services, since these variables were not on the 1995 or 1996 data files. The 95 percent confidence intervals around the means and proportions were computed using Taylor approximations with SUDAAN software, which takes into account the complex sample design of the NAMCS.⁷ Estimates are not presented if they are based on fewer than thirty cases in the sample data. The determination of statistical significance was based on a two-tailed t-test. The Bonferroni inequality was used to establish the critical value for statistically significant differences at the .05 level of significance, based on the number of possible comparisons within a particular variable or combination of variables of interest. Significance of trends was based on a weighted least squares regression analysis.

Study Results

Top Methods Study Results Discussion NOTES

 
During 1995–1999 an average of 760.8 million physician office visits were made annually in the United States, 60.8 percent of these to primary care physicians. Approximately 25 percent of primary care office-based physicians used PAs and/or NPs. Among this group, about 82 percent used them for fewer than 20 percent of visits. When PAs orNPs were used, they saw patients for an average of 10.8 percent of visits. At 43 percent of all PA visits, a PA was seen without a physician; the corresponding figure for NPs was 46.9 percent (Exhibit 1). Practice characteristics of physicians who used a PA or NP were the same as for those who did not (Exhibit 2). However, examination of the physician specialty distribution for visits in which only a PA was seen showed that PA-only visits were more likely to be associated with general and family practice than were physician-only visits (68.5 percent versus 40.3 percent, respectively).

View larger version (13K): [in this window] [in a new window] EXHIBIT 3 Percentage Of Primary Care Visits, By Type Of Provider And Patient Age Group, 1995–1999

Intensity of services. The physician-only group was more likely to order two or more tests than were PAs alone (Exhibit 4). Therapeutic and preventive services, which include counseling/ education and other nonmedication therapy, were ordered or provided at a higher proportion of NP-only visits than of PA-only and physician-only visits. There was no difference in the mean number of diagnostic or screening services, therapeutic and preventive services, or number of medications ordered or provided by provider group. Mean duration of visit could only be calculated for visits in which a physician was seen, since the instructions on the form are to enter zero if no physician was seen. The mean duration for visits in which only a physician was seen (17.3 minutes) was lower than for visits in which a physician and a PA or an NP were seen (21.3 minutes).⁸

Top Methods Study Results Discussion NOTES

Reasons for survey disparities. The disparity between NAMCS and other survey results may be explained by several factors. The NAMCS only samples physicians; therefore, data are only collected for PAs and NPs whose supervising physician is selected for the survey. If PAs and NPs were sampled independently of physicians, the resulting visit estimates for them would be much greater. Visits to PAs and NPs who work in hospitals or federal government settings or NPs who have their own practices are not captured by the NAMCS. However, data from the 1999 National Hospital Ambulatory Medical Care Survey showed that there were 4.7 million PA and 1.5 million NP hospital emergency department visits, and 4.4 million PA and 5.0 million NP hospital outpatient department visits.¹⁰ Finally, while the wide variation in physicians’ use of PAs and NPs may suggest errors in reporting, our cognitive studies found that data abstractors had no difficulty completing this item.

Practice similarities and differences. The results indicate a similarity in the characteristics of physician practices that use and do not use PAs and/or NPs and of visits in which patients are seen by these providers and those in which only a physician is seen. This suggests that PAs and NPs are practicing in ways that are similar to each other and to physicians. One explanation for this observation may be the wide confidence intervals surrounding the estimates for the PA- and NP-only groups. The NAMCS findings support the results of other studies where few differences were found between PA and NP visits combined compared with total visits to hospital emergency and outpatient departments and where patients were randomized to NPs and physicians in the same primary care clinics and patient outcomes were found to be comparable.¹¹

Some differences did exist. When a physician was seen either alone or with a PA or an NP, the patients were more likely to be older than those seen by a PA or an NP without a physician. Also, the mean patient age was highest for the physician-only group. In addition, this group appeared to be attending the more complicated cases—patients who were older and required more diagnostic tests. PAs and NPs working with a physician spent more time with the patient than did physicians practicing alone, which may indicate that PAs or NPs see the patient first and then ask the physician for a consultation. In the case of NPs, it may imply that NPs are spending more time counseling patients. The finding that the NP-only group provided counseling services more often than the PA-only or physician-only groups did is supported by a study of Tennessee NPs, which found that NPs were oriented toward a nursing paradigm and were most interested in health promotion and education, while PAs were more directed toward a medical model of practice that focuses on disease, similar to that used to train physicians.¹²

Underestimation. Because the NAMCS was designed to be representative of the amount of ambulatory care being delivered by nonfederally employed physicians in office-based practice, it underestimates the full degree of primary care services that PAs and NPs deliver nationally. The federal government is the largest employer of PAs, followed by state governments.¹³ In some branches of the military, PAs make up more than one-fourth of the medical officer cadre.¹⁴ Employment of PAs in government agencies and large managed care organizations is rising, possibly because they are able to see the same number of patients in outpatient settings at approximately half the cost of a physician.¹⁵ Comparable employer information for NPs does not exist.

The results of this study may encourage health planners to use PAs and NPs in medically underserved areas. The presence of a physician in a rural area has been shown to help maintain the economy of the community.¹⁶ PAs and NPs may serve a similar role in producing employment multiplier effects, as they are more likely than physicians are to reside in communities of less than 10,000 population.¹⁷ Estimating the full extent of health services delivered by PAs and NPs is difficult, although the findings reported here may improve projections of the nonphysician clinician workforce.

Editor's Notes

 
Roderick Hooker is an associate professor and chief of the Division of Health Services Research at the University of Texas Southwestern Medical Center in Dallas. Linda McCaig is a survey statistician for the Ambulatory Care Statistics Branch, Division of Health Care Statistics, National Center for Health Statistics, in Hyattsville, Maryland.

The authors thank Irma Arispe, Jennifer Madans, Fitzhugh Mullan, and Richard Cooper for valuable comments. Only the authors are responsible for the contents of this paper.

NOTES

Top Methods Study Results Discussion NOTES

 

1. J.F. Cawley and P.E. Jones, "Nonphysician Clinicians in the Health Care Workforce," Journal of the American Medical Association 281, no. 6 (1999): 509–510[Medline]; and S.A. Schroeder, "How Can We Tell Whether There Are Too Many or Too Few Physicians? The Case for Benchmarking," Journal of the American Medical Association 276, no. 22 (1999): 1841–1843.
2. R.A. Cooper, P. Laud, and C.L. Dietrich, "Current and Projected Workforce of Nonphysician Clinicians," Journal of the American Medical Association 80, no. 9 (1998): 788–794; D.C. Goodman et al., "Benchmarking the U.S. Physician Workforce: An Alternative to Needs-Based or Demand-Based Planning," Journal of the American Medical Association 276, no. 22 (1996): 1811–1817[Abstract/Free Full Text]; J.P. Weiner, "Forecasting the Effects of Health Reform on U.S. Physician Work force Staffing Patterns: Evidence from HMO Staffing Patterns," Journal of the American Medical Association 272, no. 3 (1994): 222–230[Abstract/Free Full Text]; and D.A. Kindig, "Strategic Issues for Managing the Future Physician Workforce," in Strategic Choices for a Changing Health Care System, ed. S.H. Altman and U.E. Reinhardt (Chicago: Health Administration Press, 1996), 149–182.
3. American Academy of Physician Assistants, General Census Report, 1998 (Alexandria, Va.: AAPA, 1998); and A. Running et al., "A Survey of Nurse Practitioners across the United States," Nurse Practitioner 25, no. 7 (2000): 15, 16, 110–116.
4. D. Cherry, C.W. Burt, and D.A. Woodwell, National Ambulatory Medical Care Survey: 1999 Summary, Advance Data from Vital and Health Statistics no. 324 (Washington: National Center for Health Statistics, 2001).
5. AAPA, General Census Report, 1998; and Running et al., "A Survey of Nurse Practitioners."
6. J.F. Cawley, Letter to the Editor, Journal of Pediatric Health Care 6 (1994): 132.
7. B.V. Shah, B.G. Barnwell, and G.S. Bieler, SUDAAN User’s Manual, Release 7.0 (Research Triangle Park, N.C.: Research Triangle Institute, 1996).
8. Visit durations were as follows: PA with physician, mean 23.4 minutes (95 percent confidence interval, 19.9, 26.8); NP with physician, mean 18.6 minutes (17.1, 20.0); physician only, mean 17.3 minutes (16.9, 17.7); and PA or NP with physician, mean 21.3 minutes (19.1, 23.6).
9. AAPA, Number of Patient Visits Made to Physician Assistants and Number of Medications Prescribed or Recommended by Physician Assistants in 1999 (Alexandria, Va.: AAPA, 1999); and Running et al., "A Survey of Nurse Practitioners."
10. L.F. McCaig and C.W. Burt, National Hospital Ambulatory Medical Care Survey: 1999 Emergency Department Summary, Advance Data from Vital and Health Statistics no. 320 (Washington: NCHS, 2001); and N. Ly, L.F. McCaig, and C.W. Burt, National Hospital Ambulatory Medical Care Survey: 1999 Outpatient Department Summary, Advance Data from Vital and Health Statistics no. 321 (Washington: NCHS, 2001).
11. R.S. Hooker and L.F. McCaig, "Emergency Department Uses of Physician Assistants and Nurse Practitioners: A National Survey," American Journal of Emergency Medicine 14, no. 3 (1996): 245–249[Medline]; L.F. McCaig and R.S. Hooker, "Physician Assistants and Nurse Practitioners in Hospital Outpatient Departments, 1993–1994," Public Health Reports 113, no. 1 (1998): 75–82; and M.O. Mundinger et al., "Primary Care Outcomes in Patients Treated by Nurse Practitioners or Physicians: A Randomized Trial," Journal of the American Medical Association 283, no. 1 (2000): 59–68.[Abstract/Free Full Text]
12. N.B. Moody, P.L. Smith, and L.L. Glenn, "Client Characteristics and Practice Patterns of Nurse Practitioners and Physicians," Nurse Practitioner 24, no. 3 (1999): 94–96, 99–100, 102–103; and P.D. Jacobson, L.E. Parker, and I.D. Coulter, "Nurse Practitioners and Physician Assistants as Primary Care Providers in Institutional Settings," Inquiry (Winter 1998/99): 432–446.
13. R.S. Hooker, "The Economics of Physician Assistant Employment," Physician Assistant 24, no. 4 (2000): 67–85.
14. R.S. Hooker, "The Military Physician Assistant," Military Medicine 156, no. 12 (1991): 657–660.[Medline]
15. Hooker, "The Economics of Physician Assistant Employment."
16. G.A. Doekson,T. Johnson, and C. Willoughby, Measuring the Economic Importance of the Health Sector on a Local Economy: A Brief Literature Review and Procedure to Measure Local Impact, Report no. 202 (Mississippi State: Southern Rural Development Center, 1997); and S. Kleinholz and G.A. Doekson, "The Impact of Rural Hospitals and Rural Physicians on the Economy of a Rural Community" (Paper presented at the 1991 Western Multi-Conference Simulation, Anaheim, California, 1 March 1991).
17. Ibid.

                      What's this?

This article has been cited by other articles:

				J. M. Colwill, J. M. Cultice, and R. L. Kruse Will Generalist Physician Supply Meet Demands Of An Increasing And Aging Population? Health Aff., May 1, 2008; 27(3): w232 - w241. [Abstract] [Full Text] [PDF]

				P. Rosenfeld, H. Kim, G. Londono, C. Kovner, and M. Mezey Adult Ambulatory Care Visits to Nurses and Physicians: Methodological Limitations of the Medical Expenditure Panel Survey Data Policy Politics Nursing Practice, August 1, 2005; 6(3): 221 - 228. [Abstract] [PDF]

				N. P. Hanrahan and E. M. Sullivan-Marx Practice Patterns and Potential Solutions to the Shortage of Providers of Older Adult Mental Health Services Policy Politics Nursing Practice, August 1, 2005; 6(3): 236 - 245. [Abstract] [PDF]

				D. C. Goodman and the Committee on Pediatric Workforce The Pediatrician Workforce: Current Status and Future Prospects Pediatrics, July 1, 2005; 116(1): e156 - e173. [Abstract] [Full Text] [PDF]

				E. R. Lenz, M. O. Mundinger, R. L. Kane, S. C. Hopkins, and S. X. Lin Primary Care Outcomes in Patients treated by Nurse Practitioners or Physicians: Two-Year Follow-Up Med Care Res Rev, September 1, 2004; 61(3): 332 - 351. [Abstract] [PDF]

				R. Town, D. R. Wholey, J. Kralewski, and B. Dowd Assessing the Influence of Incentives on Physicians and Medical Groups Med Care Res Rev, September 1, 2004; 61(3_suppl): 80S - 118S. [Abstract] [PDF]

				E. R. Lenz, M. O. Mundinger, S. C. Hopkins, S. X. Lin, and J. L. Smolowitz Diabetes Care Processes and Outcomes in Patients Treated by Nurse Practitioners or Physicians The Diabetes Educator, July 1, 2002; 28(4): 590 - 598. [Abstract] [PDF]