Strengthening Primary Care to Bolster the Health Care Safety Net

The future of primary care is in jeopardy, and nowhere is this more evident than in the current debate on physician workforce. Reversing its projections that the United States would be faced with a physician surplus by 2000,^{1 - 2} the Council on Graduate Medical Education in early 2005 projected a physician shortage by 2020.³ This conclusion was reached despite the council's estimate that the number of physicians per 100 000 population would actually increase 5% between 2000 and 2020. The projected shortage is supported by forecasts that the US population's demand for specialized services will increase more rapidly than the growth in physician supply.⁴

Just as the national consensus in the early 1990s of a specialty physician surplus was associated with a subsequent increase in the number of medical graduates choosing a primary care specialty (family medicine, internal medicine, or pediatrics), it appears that recent claims of a specialist shortage are having the opposite effects. In at least 1 state, residents completing specialist training receive more job offers than those trained as generalists,⁵ and medical graduates' interest in family medicine is on the decline.⁶

The proclamations of a looming specialist shortage are made even though there is no evidence that more specialty care improves population health. Nations with a strong primary care infrastructure have better health outcomes than those such as the United States that emphasize specialty medicine.⁷ One reason for this phenomenon may be that primary care is much more important than specialty care in providing services to those most in need (ie, vulnerable populations), which serves to narrow health disparities associated with ethnic group, socioeconomic, and geographic residence status.

Amid the physician workforce banter on whether there are too many or too few physicians,⁸ consensus can be found regarding one claim: physicians in the United States are not equitably distributed.⁹ Geographic maldistribution of physicians creates pockets of medically underserved communities while others have excess supply. Market forces in the US health sector have failed to supply physicians where they are needed.⁹ Large numbers of uninsured Americans lack access to care, and as many as 50 million live in communities that the federal government designates as health professional shortage areas.¹⁰

The government has long recognized the inability of the market to rationally distribute physicians and has created a variety of programs to improve access to primary care in underserved communities. The most prominent of these is the now 40-year-old health center program. Nearly half of all uninsured persons in the United States live within 5 miles of a health center.¹¹ Closer proximity to health centers is associated with decreased levels of unmet need and lower likelihood of emergency department use and hospitalization for the uninsured.¹¹ Expansion of the health center safety net holds the potential for improving access while promoting cost efficiency for low-income and uninsured persons.

The study by Rosenblatt and colleagues¹² in this issue of JAMA provides sobering evidence that recent physician workforce trends may be adversely affecting access to primary care for the nation's most vulnerable populations and may put at risk the planned expansion of the health center safety net. The authors conducted a nationally representative survey of health center chief executive officers (CEOs) to examine physician staffing, job vacancies, and recruitment barriers. The largest number of vacancies was for family physicians, although recruitment was most difficult for obstetrics/gynecology and psychiatry, both with more than 20% of positions vacant. Some physician vacancy is to be expected, and the optimal or expected level is not known. Nonetheless, 42% of rural centers reported that recruiting family physicians was “very difficult” vs 21% of urban centers. The message of the study is clear by any criterion: health centers, particularly those in rural settings, are having a hard time recruiting and retaining physicians.

Difficulties attracting physicians to health centers may pose significant challenges to the federal government's planned expansion of the health center safety net. In fiscal year 2002, the federal Health Center Initiative was passed into law establishing a 5-year, $780 million program to create new or expand existing health center access points in medically underserved communities. Between 2001 and 2005 the number of health center users increased from 10.3 million to 14.0 million, which is about two thirds of the number of new users anticipated by the initiative.¹³ This expansion has not been accompanied by a higher workload for health center physicians. Between 1996 and 2004, the average annual number of primary care encounters per physician employed by health centers was stable at 4000 (personal communication, Leiyu Shi, DrPH, January 19, 2006). So, if the health center safety net is successfully expanding and primary care physicians are not seeing more patients, why are CEOs reporting such difficulty attracting physicians?

One explanation may be that the CEO survey was a single snapshot of recruitment and staffing patterns in 2004. The data do not permit drawing the conclusion that physician recruitment is more or less difficult than it always has been. However, health centers are located in communities that are underserved because of low physician supply, which makes it highly likely that recruitment has always been challenging. Another important trend is that most of the centers in the survey were actively expanding their capacity, and this new recruitment burden may be straining their ability to attract qualified physicians, particularly family physicians who comprise half of the health center workforce and whose numbers are in decline.

For every primary care specialty, rural centers reported substantially more difficulty filling physician vacancies than did urban centers.¹² To overcome these barriers, rural centers make greater use than urban counterparts of 3 government physician workforce programs: National Health Service Corps (NHSC) medical school scholarships; federal and state medical school loan repayment; and the J-1 visa waiver program for international medical graduates. Without this assistance from government, medically underserved rural populations would experience severe barriers to primary care and poorer health status.

The greatest burden of ill health and morbidity in the United States is experienced by its vulnerable populations. To reach the goals of Healthy People 2010,¹⁴ much greater attention must be given to achieving a strong and expanded primary care workforce. Bolstering this safety net is one of the best strategies for improving the health of the nation. Several practical options are immediately available to policy makers for improving access to physicians in underserved areas.

The cornerstone of a national strategy for strengthening the safety net begins with ensuring a well-trained cadre of primary care professionals. Title VII of the Health Professions Educational Assistance Act provides funds specifically for primary care training and is associated with an increase in the number of family physicians working in medically underserved communities.¹⁵ Its political support, however, is unstable, and several federal budgets in both Democratic and Republican administrations have attempted to eliminate it. Without strong advocacy from such organizations as the American Academy of Family Physicians and the Association of American Medical Colleges, Title VII may not be funded today. Ensuring sustained and enhanced financial support for Title VII is critically important and must be a legislative priority.

Title VII, section 747 program funds for primary care training, adjusted for inflation, have been decreasing since 1977 and amount to about $100 million annually.¹⁶ The government provides substantially larger subsidies for hospital-based physician training, most of which is for future specialists. Medicare and Medicaid provide about $10 billion in direct and indirect graduate medical education (GME) payments to teaching hospitals.¹⁷ GME funds are paid directly to hospitals, which are free to allocate these resources at their discretion. Some academic medical centers have tried to move more training into outpatient settings, but complex Medicare regulations concerning ambulatory training have greatly hindered this transition.¹⁷ The Centers for Medicare & Medicaid Services should give more attention to facilitating the transfer of direct GME payments to the physician organizations actually providing the medical training. This accountability could lead to increased support of generalist physician training programs.

A limited number of NHSC scholarships are given each year to medical students, who in return commit to practice primary care in medically underserved areas. This program is popular and highly competitive among medical trainees: for every applicant awarded an NHSC scholarship there are 7 applications.¹⁰ The NHSC has a 3-decades-long history of staffing the primary care safety net, and with increased funding it could readily be expanded given its existing administrative capacity. Increasing the number of NHSC scholars and recipients of NHSC and state loan repayments is perhaps the most direct and immediately effective way to expand the health center physician workforce.

However, if the NHSC program were expanded, more attention would need to be given to retaining NHSC scholars beyond their service commitment. One study has shown that NHSC physicians leave health centers sooner than their colleagues who are not NHSC scholarship recipients, most commonly exiting at the conclusion of their service obligation.¹⁸ As Rosenblatt et al suggest, one of the most effective strategies for physician recruitment may be to get more resources for health centers to provide better compensation packages. One simple strategy would be to add retention bonuses from the Health Services and Resources Administration for NHSC physicians remaining in their placement.

Health centers have been on the leading edge in the use of advanced practice nurses. Recruitment of nurse practitioners, according to Rosenblatt et al, is not as difficult as recruiting physicians. Long-term solutions to staffing the primary care safety net in both rural and urban areas will involve a greater reliance on nurse practitioners and other nonphysician clinicians. Further research is needed to better understand the most effective and efficient staffing patterns of physicians and nurses, particularly for patients with multiple morbidities and high health care needs.

Finally, overcoming the fragmentation that characterizes primary care in the United States must become a priority. The absence of a professional organization for all primary care professionals, linking family medicine, internal medicine, and pediatrics, and advance practice nursing renders coherent primary care workforce planning substantially more difficult.