Health Affairs, 22, no. 1 (2003): 190-198 doi: 10.1377/hlthaff.22.1.190 © 2003 by Project HOPE	New Online   Pay Cuts For Medicare Docs   Access To Care Woes   Public Coverage More Efficient   Empowering Consumers

This Article Abstract Figures Only Reprint (PDF) Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal 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 (16) Citing Articles via Google Scholar Google Scholar Articles by Colwill, J. M. Articles by Cultice, J. M. Search for Related Content PubMed Articles by Colwill, J. M. Articles by Cultice, J. M. Related Collections Physicians Rural Health Care

Health Tracking

MARKETWATCH

The Future Supply Of Family Physicians: Implications For Rural America

Abstract

 
Throughout the past century rural health care has been dependent upon general practitioners (GPs) and their successors, family physicians (FPs). Only FPs and GPs have practiced in rural areas in proportion to the population, then and now. As specialization occurred, numbers of GPs declined and physician shortages developed in rural areas. The creation of family practice residencies in the 1970s halted this decline, but rural shortages persist today. During the 1990s the number of allopathic and osteopathic FP residency graduates rose 54 percent. At the same time, the percentage of women enrolled in these residencies increased to 46 percent, and women have been less likely than men to select rural practice. We project that if current numbers of graduates continue, the nonmetropolitan FP/GP-to-population ratio will increase 17 percent by the year 2020. However, today, medical students’ interest in primary care residencies (including family practice) is declining precipitously. If numbers of FP graduates return to 1993 levels, the density of FPs in rural America and in the nation as a whole will decline after 2010.

The shortage of rural physicians has been attributed to the relative social and professional isolation of rural communities, the availability of hospitals and technology in cities, and the flight to urban affluence.³ However, almost certainly the decline of the GP has been the most important cause of the shortage of rural physicians. As more and more physicians specialized, numbers of GPs—those most likely to serve rural areas—gradually declined.

Two overarching trends in the physician-to-population ratio for all physicians and for GPs and their successors, family practitioners (FPs), during the twentieth century are apparent (Exhibit 1). First, the overall physician-to-population ratio remained relatively constant throughout the first two-thirds of the century and then dramatically increased in the last third as a result of a doubling of the number of U.S. medical graduates and recruitment of international medical graduates (IMGs). Second, the ratio of FPs/GPs to population fell continuously to approximately one-fifth of the 1900 level. This occurred as practicing GPs limited their practices to a specialty in the early 1900s and then as almost all medical school graduates entered specialty training following World War II.

View larger version (24K): [in this window] [in a new window]   EXHIBIT 1 Total Physicians And Family/General Practitioners Per 100,000 U.S. Residents During The Twentieth Century

Looking to the future, two major forces will influence the future supply of rural FPs: recent increases in the number of residency graduates, and the increasing feminization of the FP workforce. The managed care revolution of the 1990s sparked a tremendous increase in medical students’ interest in the primary care specialties. This was fostered by a belief that there would be a shortage of generalists and an increasing surplus of specialists as managed care dominated the medical marketplace. Hospitals expanded FP residencies and initiated ninety-three new allopathic FP residencies.⁶ Between 1992 and 2000, graduates of allopathic FP programs increased by 49 percent, and osteopathic FP graduates increased by 111 percent (Exhibit 2).⁷ More than 4,000 physicians completed FP residency programs in 2000—an increase of 54 percent since 1993. This supply of new FPs, if it continues, will increase the generalist physician supply and will have a sizable impact on the rural physician workforce.

The second major workforce trend is the increasing proportion of women among family practice graduates. The proportion of women in allopathic FP residencies, only 19 percent in 1980, has blossomed today to 46 percent of allopathic FP graduates and 37 percent of osteopathic graduates.⁸ This is highly significant, since women have been less likely to select rural practice. Data from the American Medical Association (AMA) Masterfile demonstrate that 24 percent of men and 16 percent of women under age forty-five in family practice are located in nonmetropolitan counties. An increasing proportion of women will reduce the proportion of FP graduates entering rural practice. In this paper we examine the interplay of increasing numbers of FP graduates and an increasing proportion of women on the future supply of both rural and urban FPs. We also model the impact of a decline of FP graduates to 1993 levels on future FP supply.

Study Methods And Limitations

Top Study Methods And Limitations Study Findings Discussion NOTES

 
This study projects the rural and national supply of practicing family physicians to the year 2020 using the Bureau of Health Professions (BHPr) Physician Supply Model.⁹ This model adds numbers of new entrants to the workforce and subtracts losses from the workforce. Our projections are based upon two scenarios: (1) a continuation of the year 2000 FP output of 4,015 graduates annually, and (2) a decrease over five years to 2,602, the number of 1993 FP graduates, a scenario that might occur if current declines continue. Our projections are for active patient care FPs. We exclude resident physicians and those in administration and academic settings.

Our workforce data were obtained from the AMA Masterfile and from the masterfile of the American Osteopathic Association (AOA). Using metropolitan statistical area (MSA) definitions, we identified numbers of non-metropolitan and metropolitan patient care FPs/GPs, excluding residents, and stratified them by age and sex. Numbers and sexes of recent FP graduates were obtained from the American Academy of Family Physicians (AAFP) and the AOA.

Attrition from the patient care FP workforce may occur in three ways: movement to non–patient care activities such as administration, switching to another specialty, and separation from medicine through death or retirement. About 4.4 percent of active FPs/GPs, excluding residents, indicated that they are in academic or administrative positions. We assume that half of this 4.4 percent loss occurs soon after residency and half over the next twenty years. Two percent of FP graduates plan training in another specialty following residency. Therefore, we reduced FP graduates by 2 percent over six years. Subsequent specialty changes were calculated from self- reported specialty data from the AMA Masterfile for 1992–1996, and stratified by age and sex.¹⁰ Overall, about 11 percent of FPs in patient care are expected to leave patient care through either specialty shifting or moving to a non–patient care position over a professional lifetime. Attrition from the workforce through retirement and death was calculated from the BHPr Physician Supply Model.

AMA Masterfile data indicate that 11 percent of active patient care FPs/GPs are older than age seventy. Although careful studies have not been performed, it is likely that many of these physicians have very limited practices and that many others have retired but not notified the AMA. Consequently, physicians over age seventy were excluded from our analyses.

The projection of rural FP numbers must take into account the percentage of women and the lower probability that women will enter rural practice. Currently, 45 percent of allopathic and osteopathic graduates are women. Since 16 percent of female and 24 percent of male FPs practice in nonmetropolitan counties, we calculate that 20 percent of all graduates will practice in these settings. Thus, curent FP graduates will enter nonmetropolitan practice in proportion to the general population inasmuch as 20 percent of the population also resides in rural non-MSA counties.

We use Bureau of the Census population data for the year 2000 and prorate future population based upon earlier Census Bureau projections because newer projections are not yet available. These projections produce an anticipated U.S. population of 334 million in 2020. Rural population projections are provided by Standard and Poor’s DRI.¹¹ They predict that the rural population, while continuing to increase, will fall from 20 percent of the total population to 19 percent in 2020. Therefore, the proportion of FP graduates entering rural practice is assumed to decrease gradually from 20 percent today to 19 percent in 2020.

The methodological assumptions on which our projections are based must be recognized. The AMA data on professionally active physicians are in essence a "head count" of all physicians working more than twenty hours per week. These physicians average fifty-five hours of professional activity weekly, so the head count approximates the number of full-time-equivalent (FTE) physicians. No productivity adjustments are made for age or sex. The projection model is based on historical rates of retirement from practice. An increase in retirement rates, a realistic possibility in the current environment, could greatly reduce total FP numbers. An increase in the proportion of FPs/GPs engaged in non-patient care activities could lead to lower-than-expected numbers of patient care FPs. Our population projections are dependent on many assumptions such as rates of birth and death and of immigration and assumptions about demographic trends in urban/rural distribution. As a result, short-term projections are likely to be more accurate than long-term ones are.

The masterfiles of the AMA and the AOA are the only comprehensive sources of data on the physician workforce. The accuracy of the data is dependent on the self-report of physicians. A recent study by Thomas Konrad suggests that numbers of rural practitioners may actually be 20 percent below AMA Masterfile figures.¹² If these findings are verified, our projections may greatly overestimate numbers of rural FPs.

Study Findings

Top Study Methods And Limitations Study Findings Discussion NOTES

 
If current numbers of family physicians continue to graduate from FP residencies, the total number of practicing family physicians, excluding residents, will increase 49 percent to 112,000 by the year 2020 (Exhibit 3). However, if the number of FP graduates declines over five years to the 1993 level of 2,602, the total projected number of FPs nationally will be 89,000. The proportion of women in the FP workforce will double to 40 percent in 2020.

View larger version (17K): [in this window] [in a new window]   EXHIBIT 3 Total Practicing Family/General Practitioners, 1997–2020

View larger version (16K): [in this window] [in a new window]   EXHIBIT 4 Practicing Family/General Practitioners Per 100,000 Population, 1997–2020

View larger version (16K): [in this window] [in a new window]   EXHIBIT 5 Rural Practicing Family/General Practitioners Per 100,000 Population, 1997–2020

Top Study Methods And Limitations Study Findings Discussion NOTES

View larger version (19K): [in this window] [in a new window]   EXHIBIT 6 Number Of Physicians By Specialty Per 100,000 Population, By County Population

We are now completing projections for numbers of general internists and general pediatricians. If current numbers of graduates continue, we expect the rural generalist supply to increase to 59 per 100,000 in 2020, with thirty-six FPs, sixteen general internists, and seven general pediatricians.¹⁴ However, trends toward increasing subspecialization may be returning to both internal medicine and pediatrics and may moderate generalist increases in these specialties as well. If numbers of generalist graduates decline to 1993 levels, total rural generalists will fall to 47 per 100,000 in 2020, with twenty-nine FPs, twelve general internists, and six general pediatricians. Numbers of general internist and general pediatrics graduates were estimated by subtracting those entering subspecialty fellowships and those switching to another specialty from residency graduates in each specialty.

Numbers of nurse practitioners (NPs) and physician assistants (PAs) are rising; these providers play important roles in many primary care settings. However, their role in office-based primary care practice may be less than many believe.¹⁵ The National Ambulatory Medical Care Survey (NAMCS) provides data on the practice of nonfederal office-based physicians. In analyzing these data, Roderick Hooker found that NPs/PAs were involved in only 3 percent of 462 million visits to office-based primary care physicians annually between 1995 and 1999.¹⁶ He noted that 23 percent of urban and 28 percent of rural primary care physicians work with NPs/PAs and that these NPs/PAs accounted for 11 percent of patient encounters in these practices. Independent nursing practice also appears to be small. Data from the 1998 Medical Expenditure Panel Survey (MEPS), provided by Fryer, indicate that eight million visits were made to nurses (including NPs) at locations where a physician did not work. The percentage of NP visits among these visits is not known, but fewer than half were for diagnosis and treatment, which an NP might do. These eight million visits can be compared with 871 million physician visits in 1998. The NAMCS data probably underestimate the total number of NP/PA encounters in primary care practice because the NAMCS samples physicians and the NPs/PAs they supervise. Figures would have been higher had NPs/PAs been directly sampled. Nevertheless, using similar survey questions, the National Hospital Ambulatory Medical Care Survey documents a greater number of NP/PA visits in hospital outpatient departments and emergency rooms than occurred in office-based primary care practice.¹⁷ NPs/PAs clearly provide invaluable services in many settings, including care for the underserved. Although the data are inadequate to define the extent of these providers’ activity in office-based primary care, it appears that the bulk of their activity occurs outside of this setting.

FPs will continue to be essential for health care for rural Americans, one of the nation’s largest underserved populations. Demand for their services is great as indicated by the fact that rural FPs average 20 percent more office visits and six hours more work per week than their counterparts in large metropolitan areas.¹⁸ Unfortunately, their net income does not exceed that of their urban colleagues because of lower reimbursement rates and greater numbers of uninsured patients in rural areas.¹⁹ They are also central to health care for the nation as a whole, providing 22.5 percent of ambulatory visits to all physicians and 45 percent of visits to generalist physicians while totaling only 11 percent of all physicians.²⁰

For the first time in almost a century, the nation now appears to be educating adequate numbers of FPs. If current numbers of graduates continue, the supply of rural family physicians will almost reach the range of 60–80 generalists per 100,000 suggested by the Council on Graduate Medical Education (COGME).²¹

Unfortunately, however, multiple forces threaten the nation’s ability to maintain an output of 4,000 FPs annually. Numbers of U.S. medical school graduates matching in family practice in 2002 have returned to 1993 levels. Open positions will have to be filled by increasing numbers of IMGs. The percentage of IMGs in FP residencies has increased from 12.6 percent in 1998 year to 23.4 percent in 2001 and will increase further, an issue of great concern to program directors. Further, as demand increases for subspecialists, hospitals face pressure to reduce FP positions to allow for more residents in other specialties. Funding of FP residency programs has been highly dependent upon Title VII training grants. The Bush administration is recommending the elimination of these funds, just as the number of graduates has increased to necessary levels.

From a policy perspective, FP workforce needs should be considered separately from those in other generalist specialties because family practice is unique in its ability to meet the primary care needs of smaller rural counties. Numbers of FP graduates should be maintained at approximately current levels.

Many studies demonstrate that physicians are more likely to select rural practice if they have a rural background, enter a medical school with a commitment to rural medicine, select family practice, and have rural experience as part of their educational program.²² Medical schools and residency programs should continue to emphasize recruitment of candidates with rural backgrounds and should provide rural educational experience if they hope to increase numbers of rural physicians.

Recent growth in the number of family practice residency graduates is good news for rural America as well as for the nation as a whole. For the first time in a century, the ratio of FPs to population is increasing in both rural and urban areas. But will these increases be sustained, or are they a "flash in the pan"? Today, interest in family practice by current U.S. medical students is diminishing, and teaching hospitals may have less motivation to maintain FP residency positions as a result of competing demands for residency positions in other specialties and budgetary constraints. As a matter of sound health policy, particularly for rural America, it is vital that we not lose the ground that has been gained. Adequate numbers of FPs have been, and will continue to be, essential for the health care needs of rural America as well as for the nation as a whole.

Editor's Notes

 
Jack Colwill, a physician, is emeritus professor of family and community medicine, University of Missouri-Columbia School of Medicine. James Cultice is an operations research analyst and technical director for the National Center for Health Workforce Analysis, Bureau of Health Professions, Health Resources and Services Administration, in Rockville, Maryland.

This work was supported in part through Health Resources and Services Administration (HRSA) Contract no. 98-BHPR-B32877.

NOTES

Top Study Methods And Limitations Study Findings Discussion NOTES

 

1. M.M. Davis, Public Medical Services: A Survey of Tax-Supported Medical Care in the United States (Chicago: University of Chicago Press, 1937).
2. G.A. Silver, "Family Practice: Resuscitation or Reform?" Journal of the American Medical Association 185, no. 3 (1963): 188–191.[Medline]
3. J.K. Cooper et al., "Rural or Urban Practice: Factors Influencing the Location Decision of Primary Care Physicians," Inquiry 12, no. 1 (1975): 18–25; [Medline]R.A. Rosenblatt and I.S. Moscovice, Rural Health Care (New York: John Wiley and Sons, 1982); F.D. Mott and M.I. Roemer, Rural Health and Medical Care (New York: McGraw-Hill, 1948); and M.I. Roemer, "Historic Development of the Current Crisis of Rural Medicine in the United States," in Victor Robinson Memorial Volume: Essays on History of Medicine, in Honor of Victor Robinson on His Sixtieth Birthday, August 16, 1946 (New York: Froben Press, 1948), 333–342.
4. M.D. Overpeck, "Physicians in Family Practice, 1931–67," Public Health Reports 85, no. 6 (1970): 485–494.[Medline]
5. Mott and Roemer, Rural Health and Medical Care.
6. American Academy of Family Physicians, Family Practice Residency Programs, Reprint no. 150 (Kansas City, Mo.: AAFP, 1998).
7. Ibid.
8. Ibid.
9. L. Greenberg and J. Cultice, Forecasting the Future Supply of Physicians: Logic and Operation of the BHPr Physician Supply Model, Final Report (Rockville, Md.: Health Resources and Services Administration, Bureau of Health Professions, Office of Health Professions Analysis and Research, 1992).
10. Ibid.
11. Standard & Poor’s DRI, Total Population by Metropolitan Statistical Areas Projected 1999–2020 (Rockville, Md.: BHPr, 1999).
12. T.R. Konrad et al., "Using the American Medical Association Physician Masterfile to Measure Physician Supply in Small Towns," Journal of Rural Health 16, no. 2 (2000): 162–167.[Medline]
13. G.E. Fryer et al., "The United States Relies on Family Physicians Unlike Any Other Specialty,"American Family Physician 63, no. 9 (2001): 1669.[Medline]
14. J.M. Colwill and J.M. Cultice, The Generalist Physician Supply: Yesterday, Today, and Tomorrow, Report of HRSA Contract no. 98-0679P (Rockville, Md.: BHPr, 2002).
15. R.S. Hooker and L.F. McCaig, "Use of Physician Assistants and Nurse Practitioners in Primary Care, 1995–1999," Health Affairs (July/Aug 2001): 231–238; and R.R. Aparasu and M. Hegge, "Autonomous Ambulatory Care by Nurse Practitioners and Physician Assistants in Office-Based Settings," Journal of Allied Health 30, no. 3 (2001): 153–159.[Medline]
16. Hooker and McCaig, "Use of Physician Assistants and Nurse Practitioners."
17. Ibid.
18. M.L. Gonzalez and P. Zhang, eds., Socioeconomic Characteristics of Medical Practice, 1997/1998 (Chicago: Center for Health Policy Research, American Medical Association, 1998).
19. L.G. Hart et al., "Rural Health Care Providers in the United States," Journal of Rural Health 18, no. 5 supp. (2002): 211–232.[Medline]
20. D.K. Cherry et al., National Ambulatory Medical Care Survey: 1999 Summary (Hyattsville, Md.: National Center for Health Statistics, 2001), 1–35.
21. Council on Graduate Medical Education, Patient Care Physician Supply and Requirements: Testing COGME Recommendations, Eighth Report (Rockville, Md.: Health Resources and Services Administration, 1996).
22. J.M. Colwill, "Where Have All the Primary Care Applicants Gone?" New England Journal of Medicine 326, no. 6 (1992): 387–393; [Medline]J.E. Verby et al., "Changing the Medical School Curriculum to Improve Patient Access to Primary Care," Journal of the American Medical Association 266, no. 1 (1991): 110–113; [Abstract]R.A. Rosenblatt et al., "Which Medical Schools Produce Rural Physicians?" Journal of the American Medical Association 268, no. 12 (1992): 1559–1565; [Abstract]D.E. Pathman et al., "Preparing and Retaining Rural Physicians through Medical Education," Academic Medicine 74, no. 7 (1999): 810–820; and [Medline]H.K. Rabinowitz et al., "Critical Factors for Designing Programs to Increase the Supply and Retention of Rural Primary Care Physicians," Journal of the American Medical Association 286, no. 9 (2001): 1041–1048.[Abstract/Free Full Text]

                      What's this?

This article has been cited by other articles:

				R. L. Ferrer Pursuing Equity: Contact With Primary Care and Specialist Clinicians by Demographics, Insurance, and Health Status Ann. Fam. Med, November 1, 2007; 5(6): 492 - 502. [Abstract] [Full Text] [PDF]

				D. S. Meyers, R. Mishori, J. McCann, J. Delgado, A. S. O'Malley, and E. Fryer Primary care physicians' perceptions of the effect of insurance status on clinical decision making. Ann. Fam. Med, September 1, 2006; 4(5): 399 - 402. [Abstract] [Full Text] [PDF]

				S. S. Mick The Physician "Surplus" and the Decline of Professional Dominance Journal of Health Politics Policy and Law, August 1, 2004; 29(4-5): 907 - 924. [PDF]

				S. Gottlieb One in three doctors don't tell patients about services they can't have BMJ, July 17, 2003; 327(7407): 123. [Full Text]