Current Issue Pic Top Harvard Health Policy Review Archives Top
Current Issue Pic Middle About Us Fill Archives Bottom
Contact Us Fall 2000; Volume 1, Number 1
In Focus

Nurse Practitioners and Primary Care Physicians: Complements, Substitutes and the Impact of Managed Care
David Auerbach
page 1 | page 2 | page 3 | references

Non-physician clinicians (NPCs) is a general term given to many categories of professionals who perform duties in medical care that have traditionally been relegated to physicians. Some NPCs perform duties most similar to the traditional scope of primary care physicians (Nurse Practitioners (NPs), Physician Assistants (PAs), others are trained in specialty care such as Nurse Midwives, Certified Nurse Anesthetists (CRNAs), and Optometrists, and still others perform what might be deemed alternative care (Chiropractors, Acupuncturists, Naturopaths). State regulations concerning the practice of these caregivers vary widely in terms of licensing (whether licensing is required to practice as NPC), with respect to autonomy (whether NPCs are permitted to practice independently or only under the supervision of a physician), with respect to reimbursement from private and public insurance programs, and with respect to prescriptive authority (whether NPCs can prescribe drugs independently, or at all).

In two recent articles appearing in the Journal of The American Medical Association, Richard Cooper et al., discuss recent workforce trends of 10 categories of NPCs whose roles consist of at least some overlap with various physician specialties. The greatest wealth of available information and apparent interest is for the pair, Nurse Practitioners and Primary Care Physicians (henceforth, NPs and PCPs, respectively). At the current time, the numbers of NPs licensed to practice primary care in the U.S. exceeds 80,000 (Moses, 1996), is increasing by more than 10% per year and will probably overtake the number of practicing PCPs over the next decade, depending on how one categorizes physicians (Cooper et al., 1998). These groups of practitioners have roles which overlap considerably despite large differences in training and salary. Thus, one is inevitably led to questions of substitution - Can money be saved by private health care organizations and public budgets in substituting NPs for PCPs without sacrificing quality of care? Is this substitution happening now?

Toward answering the first question, a pioneering study was performed in Colorado among a largely ethnic Latino population in which patients were randomized to either NPs or PCPs as their primary care providers over a 1.5 year period and physiologic and subjective outcomes were assessed at a 6-month follow-up period (JAMA, 2000). No significant differences were found among the groups on any of the measures. While the study was criticized in a series of letters (JAMA, 2000) claiming that, e.g., the study population was a healthy one unlikely to suffer enough adverse events in 6 months to show a difference among practitioner types even if there were a real difference, there have been other studies also showing no difference in outcomes by practitioner type (e.g. Safriet, 1992).

Complements or Substitutes?
With the question of the implication of substitution of NPs for PCPs on quality of care necessarily left open, in this paper, I will focus on whether the two groups actually do, in fact, seem to act as substitutes. The question is an important one. If substitution is occurring on a wide scale, those involved in workforce planning for physicians in the future would well want to take this substitution into account. Also, since NPs provide primary care at a lower cost than PCPs do, one would expect more cost-conscious health care organizations (such as managed care organizations) to capitalize more strongly on this substitution (quality considerations aside). The result would be a different mix of care across institutions, and potentially a subsequently different health care experience for patients across institutions if there were any differences in the quality or scope of care provide from the two groups. Further, if increased use of NPs were associated with managed care, as managed care becomes more prevalent in the U.S., trends associated with more prevalent NPs would be expected to continue into the future. Finally, if it can be established that states can influence the numbers and roles of their NPs through myriad practice regulations, then any change in numbers of NPs would also have an effect on PCPs if there exists an interaction between the two. I first discuss some past thinking on the issue of complementation vs. substitution before embarking on a brief empirical investigation of this question using data from all 50 states from the years 1987 to 1996.

In a companion letter to his trend study noted above (JAMA, 1998), Richard Cooper addresses the complement-substitute question mainly out of concern for a possible decreaseddemand for physicians given the possibility of substitution of non-physician clinicians such as NPs.

Displacement vs. Supplementation—The growing numbers, increasing prerogatives, and expanding participation of NPCs will surely affect the demand for physicians, particularly those involved in simple licensed general care and routine licensed specialty care. However, the interrelationships between the demand for physicians and the availability of NPCs are complex. For example, while the need for physicians will be directly affected when NPs or naturopaths provide primary care, when CNMs perform deliveries, CRNAs administer anesthesia, services such as acupuncture, spinal manipulation, and herbal therapy may supplement rather than supplant the care provided by physicians. Counseling, patient education, and care management may also be adjunctive. However, it is likely that even these groups of services will decrease the demand for physicians, although they may not directly overlap the services that physicians provide.
He hints at the notion that the two have overlapping roles, and thus it may not be clear whether the substitution or complement effect is stronger. In commentary on the roles of NPs and primary care physicians, Cooper states, "the practices of NPCs are largely limited to wellness care and the treatment of uncomplicated acute and chronic conditions, a range of care that encompasses approximately 50% to 75% of the office visits to primary care physicians."

Diane Mahoney visited this issue 10 years earlier in Nurse Practitioner (Mahoney, 1988), noting that while nurses traditionally are complements to physicians, nurse practitioners can function as an attractive substitute to physicians, especially in a world of increasing managed care. She cites several studies showing the cost-savings potential of NPs vs. primary care physicians, based on their lower salaries and their differing practice emphases, specifically, that NPs are more oriented toward wellness care, treatment compliance, healthy lifestyles and attention to chronic conditions. NPs were also found to favor non-drug therapies over drug prescriptions in another cited study. These qualities all would be likely to reduce long-run costs via readmission and hospitalization, and such cost savings would be more attractive to managed care organizations which would benefit directly from such savings.

The complement-substitution issue is also discussed recently in a letter to JAMA, "Physicians and Nonphysician Clinicians: Complements or Competitors?" appearing in 1998 (Grumbach and Coffman, 1998). Noting the expected 10% per capita rise in physicians between 1995 and 2005 along with Cooper's projected 60% rise in NPCs in the same period, they ask, "Will there be room for this proliferation of physicians and NPCs? Will physicians and NPCs complement each other, fostering pluralism and collaborative teamwork, or will these groups primarily be competitors, tussling within the confines of a budget-limited system that cannot financially accommodate this many practitioners?" Speaking mainly of the Nurse Practitioner-Primary Care Physician groups, they again describe complementary scenarios, with NPs proving more 'wellness' care and physicians providing more 'illness' care, and with NPs providing more care in rural, 'underserved' areas while physicians stay in the larger markets.

Supporting the complement model is evidence from an exploratory study by Jacobson et al., (1998) who conducted interviews of NPs, PAs, and primary care physicians at 8 sites representing differently organized health care provision settings where the types of providers worked in close proximity. He notes that, "Regardless of whether the management style was independent NP and PA patient panels or the more traditional delegatory style seen in MSCs [multi-specialty clinics], the vast majority of NPs and PAs described their interactions with physicians as collaborative and collegial. The vast majority of physicians described professional interactions with NPs and PAs similarly." They do caution, however, that "Our finding the NPs and PAs had a largely collegial relationship with physicians may reflect our selection of institutions that have integrated NPs and PAs into their clinical practice." Further, they emphasize that managed care organizations could tip the balance toward NPs over PCPs, noting: our sample had greater autonomy and a wider scope of practice at institutions with a larger managed care population, where the tendency was to create teams of primary care practitioners (including physicians, NPs and PAs) or to assign patients to NP or PA panels. In the HMOs we visited, NPs and PAs tended to practice primary care as equal team members, whether or not they had their own patient panels. This appeared to be much less true at the MSCs we visited, where NPs and PAs performed functions delegated by supervising physicians, or at the hybrids, where the NP or PA scope of practice and autonomy varied considerably within the organization. One possible explanation for these results is that an organization with a larger percentage of managed care patients focuses on managing available resources to treat the needs of the patient population, which compels a different mix of providers than traditional fee-for-service practice. In either case, it is reasonable to expect that an organization with an extensive managed care (especially capitated) patient population will have greater incentives to incorporate NPs and PAs into an expanded primary care role. [emphasis added]
The potential for enhanced roles of NPs relative to PCPs in managed care organizations seems enormous.

page 1 | page 2 | page 3 | references

Fall 2000, Volume 1, Number 1
Table of Contents
Editor's Note
Features: Election 2000
Health Highlights
In Focus
Glossary of Health Care Terms

about us | links | contact us | subscribe | epihc