The physician-workforce pipeline is broken. Here’s how to fix it.

Anyone who has recently visited an emergency department would agree that the U.S. health system is broken. After frequently waiting for hours in overflowing waiting rooms, patients increasingly find themselves receiving care in overflow hallway beds that are understaffed by overworked physicians and nurses burnt out by grueling workplace conditions exacerbated by the pandemic. 

Even worse, according to the U.S. Centers for Disease Control and Prevention, more than 80 percent of U.S. inpatient hospital beds are currently occupied.  When hospital admission is needed, patients frequently end up boarding in the ER for days. This is massively expensive but yields public health outcomes that rank abysmally compared to other countries.  For all but the wealthiest Americans, our healthcare system has passed the breaking point. 

The toll this cycle is taking on the physician workforce is quite pronounced. Out of 1.02 million licensed physicians in the U.S. in 2020, 117,000 gave up and quit the profession between 2020 and 2021.  Put another way, more than 11 percent of the physician workforce decided to leave a job they spent decades training to perform, rather than continue practicing under those conditions. 

No profession or industry can support this level of attrition. Without an adequate system to replenish the physician workforce, we are on track to decimate this critical resource beyond recovery in the coming years.  

The downstream issues, unfortunately, will not be able to compensate for the supply side problems currently existing further upstream in the pipeline, specifically in Graduate Medical Education.  Graduate medical education, which includes residency and fellowship training programs, sits immediately upstream of our broader physician workforce. It is the necessary path to be eligible for medical license, specialization and board certification in the U.S.  

The major challenge for graduate medical education is that our current educational and training systems were not designed to replenish a rapidly diminishing physician workforce. The time it takes to train a physician is extensive — four years of medical school, plus three to seven years of residency, depending on specialty. Under current conditions, each state produces only hundreds or thousands of new physicians per year, on the order of about 50,000 nationally.

There is a solution, but it will require coordination, planning and collaboration to provide more real-time data and analysis into graduate medical education recruitment processes.

Currently, the federal government takes many months to aggregate outcomes data, which limits and slows potential interventions, rendering them too late to be meaningful. Real-time information, on the other hand, could provide insight into key demographics of residency applicants and programs and changes in specialty-wide competitiveness and market demand. It could be used to determine whether a given training program will fill its federally-funded slots. 

Studies have shown that a majority of physicians establish their practices where they complete their residency training. Therefore, building physician supply in medically underserved areas requires decision to possess information that accurately reflects market conditions and can be used to shift federal graduate medical education funding more flexibly to align more closely with need.

While the pandemic certainly exposed critical weaknesses in the physician-workforce training pipeline, the problem is not a new one.  In 2019 a U.S. Health Services and Resources Administration report estimated that as many as 13,750 additional primary care physicians would be needed simply to remove Health Professional Shortage Area designations for areas with primary care shortages.  Since that time, the problem has only worsened. 

A 2021 report by the Association of American Medical Colleges found that the U.S. could suffer a shortfall of between 37,800 and 124,000 physicians by 2034, with deficiencies in both primary and specialty care.

For its part, Congress has tried to address this issue.  The 2020 CARES Act specifically directed the Secretary of HHS to develop a “comprehensive and coordinated plan with respect to HHS health workforce development programs, including education and training programs.”  In response to that requirement, HHS published the Health Workforce Strategic Plan 2021 which identified four key goals, including the use of data and evidence to strengthen and forecast health workforce needs. 

An updated 2022 report from HHS to Congress found that “historical methodologies used to determine workforce supply and demand may no longer be appropriate and may lead to inaccurate projections.” It concluded that “dedicated investments in data collection and analysis, research, evaluation, and other evidence building activities…are essential to supporting future health workforce decision making.”

In other words, HHS has acknowledged its data problems and the need to modernize its collection efforts.

Congress has also tried boosting the number of residency training slots.   The 2021 Consolidated Appropriations Act directed CMS to begin funding 1,000 new graduate medical education positions targeted toward medically underserved areas, with the first allocation of funding already awarded this year. 

However, the HPSA scores used in such determinations — the objective numerical measure of health professional shortages — do not account for human behavior and the difficulties encountered in trying to recruit physicians across different specialties to these areas. HPSA scores may be the best measure currently available, but they are limited in determining the likelihood of the funded position being filled and producing a fully trained physician. Nor can they reliably predict that the resulting physician will remain in that area to practice.

Despite the broad recognition of the problem, bureaucratic inertia is preventing action.  Since the release of HHS’s Health Workforce Strategic Plan in 2021, no systematic efforts have been made to improve data collection efforts on the physician-workforce training pipeline.  Specifically, as the 118th Congress looks to address health workforce issues, it should:

1. Hold hearings with the broad range of federal stakeholders who play a role in graduate medical education to determine a single point of leadership for workforce planning efforts;
2. Require that HHS expeditiously update its data systems in line with its previously stated goal to do so; and
3. Require an annual report from HHS on real-time graduate medical education program performance and whether efforts to boost physician supply across specialties is actually working as intended.

The physician-workforce pipeline is a nuanced and complex issue that frequently does not receive the attention it deserves.  But as the data shows, the U.S. faces a critical tipping point in physician supply. If we fail to address this issue now, then it will be impossible to develop the diverse and representative physician workforce that can deliver the high quality health care that all Americans deserve.

Mario Ramirez is an emergency physician and consultant at Akin Gump. Jason Reminick is CEO and founder at Thalamus.