Oversimplifying Primary Care Supply and Shortages

Recent reports have warned of a crisis related to a shortage of primary care physicians.¹ However, much of the current concern seems to have stemmed from articles in the medical literature specifically reporting that fewer internal medicine residents are choosing to pursue primary care and that fewer medical students are choosing family medicine residencies. Hauer et al² found, among a national sample of fourth-year medical students, that only 22% planned careers in internal medicine and just 2% intended to practice general internal medicine. Ebell³ hypothesized that fewer medical students are choosing primary care specialties as a result of lower salaries relative to other specialties. It is important to note that the results of his trend-line analysis showed this was true only for family physicians.

There has not been an appreciable change in the proportion of pediatric residents who choose primary care pediatrics. Thus, the concerns expressed regarding primary care should really only focus on the care of adults. While this is a large and important patient population, adults are not the only group served by primary care physicians. Historically, primary care has encompassed internal medicine, pediatrics, and family medicine; in some instances, obstetrics/gynecology is included. These groups are similar with respect to their primary care roles, but they are markedly different in many other ways, including workforce projections.

For these reasons, it is disappointing that national discussions about the “crisis in primary care” have moved beyond the valid concerns regarding a shortage of physicians providing primary care to adults and have somehow come to imply a shortage of all primary care physicians.

Multiple government reports also fail to distinguish among the primary care specialties when reporting on the crisis in primary care. For example, in its February 2008 report to Congress, the Government Accounting Office reported findings predicting a shortage of primary care physicians in 2018, without reference to specific specialties.⁴ Assessments by those tasked with determining sufficiency of training the physician workforce are no more specific. A 2005 report by the Council on Graduate Medical Education did not distinguish among the primary care specialties when assessing the future training needs of the country in this important area.⁵

The need to understand primary care workforce issues in a specialty-specific manner is critical, because any policy changes or economic incentives that affect the proportion of physicians pursuing primary care will likely have a differential effect across the primary care specialties. For example, it might be viewed as positive if the number of internal medicine residents who pursue primary care were to double. Indeed, a November 2008 American Medical Association news release reported a new policy aimed at “increasing the ranks of primary care physicians” without any reference to specific primary care specialties.⁶ However, the most recent published data regarding pediatric residents completing training in 2008 demonstrated that 40% were planning to pursue a career in primary care, with 10% still undecided.⁷ Similar findings were seen among senior residents scheduled to complete their training in 2009 (G.L.F. and J.A.S., unpublished data, 2008). Thus in pediatrics, a doubling or substantial increase of the proportion entering primary care could result in a possible oversupply of primary care pediatricians, likely at the expense of needed pediatric subspecialists. Markedly increasing the number of primary care pediatricians in the pipeline could create more such physicians than jobs available, resulting in the opposite effect of any intended legislation or other policy solution aimed at primary care.

In contrast, the idea that there is a shortage of primary care pediatricians—or that, relative to the past, there are fewer primary care pediatricians to care for the current population of children—is contrary to published workforce and population data.^{8 - 9} While the absolute number of children in the United States has remained relatively stable, the number of pediatricians has increased substantially (Figure). This has resulted in an increase in the number of primary care pediatricians, from 32 to 78 per 100 000 children in the period 1975 to 2005. Similarly, there has been an increase in the number of pediatric subspecialists during this same period.^{8 - 9} Part of this increase is the result of an increase in the number of recognized subspecialties and the continuing need to populate those fields with fellowship-trained pediatricians.

It is important to realize that this increase in the per capita supply of generalist and subspecialist pediatricians does not imply an oversupply. Although the metrics by which need is assessed are variable and controversial, the pediatric workforce has easily assimilated the increase without untoward consequences and, by all reasonable accounts, with a positive effect on children's health.

Some authors, while expressing concern over the shortage of physicians to care for adults, have presented data contending that the supply of general pediatricians will be in balance with the child population over the next several years.¹⁰ However, the prominence of such findings and the resultant need to temper comprehensive policy actions are often lost in the remainder of such reports and subsequent commentaries. Even after acknowledging the likely adequacy of the primary care pediatric workforce, articles hypothesize the effect that an impending shortage of primary care physicians will have on racial disparities and access to care.¹¹ Additional research in the area of true population-need assessment, specific to the pediatric population, will better guide future workforce policy than will blanket calls for increased numbers of primary care physicians.

The apparent sufficient supply of primary care pediatricians is not a recent discovery. Shipman et al¹² conducted an analysis in 2004 demonstrating that a shortage of pediatricians was unlikely, despite the concerns regarding other primary care specialties. Goodman and the American Academy of Pediatrics Committee on the Pediatric Workforce reached similar conclusions in 2005.¹³ Why such findings are not given prominence in national reports of a “primary care shortage” is unclear. However, despite the seemingly apparent adequacy of the overall supply of general pediatricians, there are geographic distribution problems with the workforce and many underserved areas in the United States.

The issues raised in the literature related to primary care have often focused more on the quantity of clinicians rather than on the quality of care they provide. In addition to addressing the needs of specific specialties to increase their numbers of primary care physicians, financial support is needed to improve the training in the quality of care that all primary care physicians provide. This is an area in which all primary care specialties and the public can benefit. Increased investment in how primary care is delivered across all age ranges is vital. This is an especially important issue in pediatrics, because residency programs at freestanding children's hospitals receive only limited support from Medicare relative to training programs based in adult hospitals. Furthermore, this limited support is tenuous at best and relies on an annual appropriation separate from the federal funding of internal medicine and family medicine residency training.

Other critical issues extend across all primary care specialties. Each specialty struggles in its own way with concerns regarding physician job satisfaction, and primary care specialties have legitimate concerns regarding reimbursement for care provision relative to procedural-based specialties. Even these financial concerns have important specialty-specific differences. For internists and family physicians, one of the greatest issues is the reimbursement rates set by Medicare. However, the only children covered by Medicare are those with end-stage renal disease—a very small proportion. For pediatricians, Medicaid reimbursement, rather than Medicare, warrants significant attention among the publicly insured population. Indeed, Medicaid reimbursement is far less than that provided by Medicare, creating a more acute issue for pediatricians and for the minority of family physicians who care for a substantial number of indigent children.¹⁴

These differences among primary care specialties are also important from a policy standpoint and highlight distinctions with regard to the priorities of each specialty in addressing reimbursement issues. While internists and most family physicians focus on Medicare rates and look to national solutions for payment issues, pediatricians are often forced to address reimbursement on a state-by-state basis, because that is where Medicaid reimbursement rates are established. From a policy standpoint, the danger in assuming that “all primary care physicians are the same” is that the public as well as some policy makers may assume they have addressed an issue in a comprehensive fashion (ie, “fixed” primary care reimbursement), when in fact that has not occurred. This was the case in 2008, when Congress addressed reimbursement for Medicare but not for other (ie, Medicaid) physician payments.

Attempting to ensure an appropriate primary care workforce to meet the needs of the United States is a complex and daunting task that is vital to the ultimate health of the nation. Oversimplifying the nuances of the primary care workforce may result in policies and priorities at odds with needs. For the specialty of pediatrics, it appears that a close to appropriate proportion of trainees continues to enter the primary care arena. Certainly, an erosion of that proportion or in the absolute number of physicians entering pediatrics would require appropriate action to ensure a continued capacity to provide general and subspecialty care to children. Those concerned with workforce adequacy should continue to monitor the situation closely.