Controlling the growth rate of health care spending is central to the success of the Affordable Care Act or any subsequent reform. Because labor represents more than 50 percent of health care costs and the clinical workforce drives use and prices, the size and composition of the health care workforce has important ramifications for spending growth. We set out to understand the trends underlying the growth in the clinical workforce and their potential implications for health care spending, health policy, and health system design.

A large literature establishes a link between primary care–oriented health systems and lower spending. Areas with a higher concentration of primary care physicians have much lower spending per beneficiary, higher-quality care, better patient satisfaction, and lower mortality rates. Given this, many existing payment reform strategies prioritize primary care, and the success of these reforms will require a vibrant—and likely growing—primary care workforce.

How The Physician Workforce Has Changed

To observe the evolution of the clinical workforce, we used the Bureau of Labor Statistics’ Occupational Employment Statistics files between 2005 and 2015. This data set is released in May of each year and records the number of jobs (not the number of full-time–equivalent employees) by industry, occupational type, and geography. Using the North American Industry Classification System (NAICS), we limited our analysis to NAICS 621 (ambulatory health care services), 622 (hospitals), and 623 (nursing and residential care facilities). We defined “primary care physicians” as family and general practitioners, general internists, obstetricians and gynecologists, and general pediatricians, and categorized all other physician categories as “specialists” (Note 1).

Overall, there was a net increase of 2.6 million jobs in the health care sector between 2005 and 2015, accounting for 35 percent of total job growth in the United States during that period. Six percent of these jobs were for physicians. The number of primary care physician jobs grew by approximately 8 percent, while the number of jobs for specialists grew about six times faster (see Exhibit 1). In an era when we might have expected (and hoped for) rapid primary care physician growth, the share of the physician workforce devoted to primary care actually decreased from 44 percent to 37 percent, and the number of primary care physicians per capita has remained roughly flat.

Exhibit 1: Growth In Physician Workforce, 2005–15

Source: McKinsey and Company

What The Workforce Trends Mean

Given the aging of the population and expanded coverage, these findings raise concerns about access to care. Many have suggested expanding the role of non-physician primary care providers to fill the gap between the need for primary care and the supply of primary care physicians. When we broadened our definition of primary care to include the physician assistants and nurse practitioners working in primary care, the total primary care workforce grew considerably faster (17 percent between 2005 and 2015), although still much slower than specialists (Note 2). It seems we are addressing our increasing primary care needs with non-physician labor, but more research is needed to understand the clinical and economic ramifications of that trend.

Under the right conditions, the rapid growth in specialists would not necessarily be negative for health care spending. If health care markets were competitive, one might expect a greater supply of specialists to lead to lower prices for specialist care and greater competition for referrals. With the right incentives in place, this increased competition could lead to lower spending and better outcomes.

Yet, there are reasons to be skeptical of this competitive model. Fees from public payers are set administratively and unlikely to be responsive to competitive pressures. Integration between hospitals and physicians, strong patient preferences for particular specialty groups or affiliated hospitals, and the numerous information problems in health care may dampen the ability of competition to drive down specialist prices.

Moreover, it is likely that the greater number of specialists working within health systems that charge facilities fees on top of expensive specialty care will lead to more expensive care. Furthermore, specialists are paid a larger salary; a recent salary survey found the four highest-paying occupations in the United States were physician specialists. These factors will work in opposition to efforts to control health care spending growth.

Possible Policy Responses

The data raise concerns in light of the belief that we need to increase the share of primary care providers (both physicians and non-physicians) to reduce the rate of growth in health care spending. They also add urgency to recommendations made by the Medicare Payment Advisory Commission (MedPAC), the Health Resources and Service Administration (HRSA), and the Association of American Medical Colleges (AAMC) to support the growth of primary care.

MedPAC suggested in both its 2016 and 2017 reports that the disparities in physician payment resulting from the Medicare fee schedule undervalue primary care and over-compensate certain specialists, and that the fee schedule ought to be amended to reflect the value generated by primary care physicians. In 2013, the HRSA recommended that graduate medical education funding be directed more toward students who will work in family medicine, geriatrics, general internal medicine, general surgery, pediatrics, and psychiatry. In 2012, the AAMC recommended that half of newly created residency positions should be for primary care and generalist disciplines.

While these recommendations are consistent with the goal of reorienting the health care system toward primary care, efforts to expand the primary care workforce are not new. As our data suggest, past initiatives such as low interest loan programs, training grants, or service programs such as the National Health Service Corps, which provides students with loan forgiveness in exchange for a commitment to practice primary care in underserved areas, have met limited success. The workforce continues to shift toward specialists. If we are to bend the cost curve, we likely need to move more aggressively on fee schedule changes, payment reform, and workforce policies.

Note 1

This included anesthesiologists, psychiatrists, surgeons, and the Bureau of Labor Statistics (BLS) group physicians, all other. This final group accounts for “all physicians not listed separately.” Ophthalmologists, dermatologists, gastroenterologists, and cardiologists are given by the BLS as representative occupations. The Occupational Information Network includes a few more detailed occupations under this heading: allergists and immunologists, dermatologists, neurologists, nuclear medicine physicians, ophthalmologists, pathologists, radiologists, preventative medicine physicians, sports medicine physicians, urologists, and preventive medicine physicians.

Note 2

The Bureau of Labor Statistics did not track nurse practitioners separately before 2012. We constructed this statistic using published numbers from the American Association of Nurse Practitioners, the Government Accountability Office, and the Agency for Healthcare Research and Quality.