If we want health workers to go to underserved areas, we should quit training so many specialists and concentrate on educating medical generalists.
A new study about national distributions of physician workforce appeared in Rural and Remote Health Journal. It is interesting comparing the distributions of physicians to concentrations of people, income, medical schools, and development.
I was struck by what researchers found in Turkey.
In Turkey, the article tells me, generalist physicians are much more likely to spread out around the country. There is little variation in the distribution of generalists in Turkey. No matter if the community is rural or urban, high income or low income, educated more or less, higher or lower in property value — generalists are found at a relatively constant number compared to the population.
In the United States, it turns out, the same thing holds. There are about 30 family physicians per 100,000 people across a wide range of communities in most of the nation.
Generalist medical workers tend to spread out. They go where they are needed, in Turkey and the U.S.
I’m counting family practice employed MDs, osteopaths (Dos), nurse practitioners (NPs), and physician assistants (PAs). This even distribution does not hold true if the MD, DO, NP, or PA is employed in something other than family practice, as is true with 88% of MDs and about 75% of DOs, NPs, and PAs.
The United States continues to graduate just 3,000 family physicians a year. There have been increases in the number of NPs and PAs. But declines in the proportions of NPs and PAs who go into family practice have negated expansions in the number of graduates.
Training for family practice medicine works to distribute health care professionals into rural areas — to serve generally the 65 percent of the U.S. population that lives in the 30,000 zip codes where physicians are not concentrated.
Between 24% and 30% of family practice physicians can be found in rural communities. This compares to 9% to 10% for the physician average. Non-family practice physicians have less than 3% to 8% rural distribution, except for the general types of specialties (general surgery, general Ob-Gyn, general orthopedics). These general specialties have managed 11% to 14% rural distribution, but they are in the process of being minimized, marginalized, or eliminated by designs that shape workforce and reimbursement.
A Failure by Design
Other nations train generalist physicians by design. The U.S. does not.
The design for medical training forces family physicians to go to specific types of colleges, through the most expensive education and preparation, and to medical schools where fewer than 15% of doctors go into family medicine.
Best would be training for family medicine that would begin directly out of high school. This would require schools devoted to family medicine — a route followed in other nations.
Half of all physician assistants went into family practice decades ago. Now fewer than 25% do.
Nurse practitioners train 45% in family practice, but only 25% of total NP workforce is employed in family practice, according to surveys. Surveys and studies find that about 30% of the employed family practice NPs and PAs are found in rural locations.
In the U.S. today there are more opportunities for health care workers to specialize. Those who would become generalists are depleted during training, at graduation, and across the years after graduation. Higher pay, more support and greater benefits to employers (more revenue generation, lower personnel costs) are part of an entire system that shapes health care in this country — and diminishes access to general health care. This is especially true in the flexible and versatile NP and PA workforce.
Salary and benefit and support factors all insure that specialists in top concentrations get the most support while generalists in practices outside of concentrations get the least and have the least support for what they do. The same is true for nurses, teachers, public servants, and other front line careers.
The Era of the Generalist
Generalists health care suppliers received the broadest support from 1950 to 1970. There was formal family medicine training.
From 1970 to 1980 those who were driving the family medicine train managed to gain federal support and state support. Family medicine grew to 3,000 annual graduates by 1980.
Family medicine is still at 3000 annual graduates 32 years later. All class years still send about one in four graduates to rural communities, but the number of graduates has not changed.
There have an expansions in the number of NPs and PAs, but those numbers have been negated for rural and underserved areas, and for primary care in general, as these workers are increasingly lured into specialties.
The primary care delivery capacity of the United States has been shrinking year to year, except for a brief period in the 1990s. Internal medicine is shrinking from 90,000 doctors in 2000 to fewer than 40,000 projected in 2030. After all, only 1,400 people are entering primary care medical training each year.
Family medicine is fixed at about 90,000 for a primary care broadest generalist workforce because this is all that 3,000 per year entering school can maintain. Pediatric primary care is stagnant despite a 50% increase in graduates. Pediatric rural care doctors are declining.
Solutions for Rural Workforce
• Develop Rural Workforce Measurements
Understand which sources of health workers produce the most rural doctors, nurses and PAs. Then focus on sources that deliver the most care in rural settings.
• Community Friendly Training
Support training that has doctors, nurses and PAs spending at least a year of clinical training in a rural location. Discourage rural training that lasts less than 4 months because this results in too much time and effort from underserved sites compared to the student contribution to the health care team.
• Focus On Long Term Solutions and Stop Paying for Short Term Fixes
Loan repayment, recruitment, and retention costs are increasing far too fast to be sustainable. Stop supporting policies that rearrange the deck chairs of existing primary care without increasing the primary care delivery capacity result.
• Support Rural Health Careers
Support training that results in primary care as a career. Stop supporting primary care training that results in 70% of graduates entering the non-primary care workforce. For the best result for rural communities, focus on training that produces 90% of graduates who are the broadest medical generalists — the only reliable source for rural practices.
Consider rural medical schools that are given the long term obligation to restore primary care and the general surgery doctors needed in rural locations (general surgery, general ob-gyn, general orthopedics). Create family medicine medical schools that graduate only family medicine doctors.
• Long Term Obligations
Support scholarships that require at least 6 years and preferably 8 years of service in underserved areas. Stop supporting scholarships that have short term obligations.
• Long Term In-State Retention
Identify and then support training that results in 70% or more of graduates staying inside of the state sponsoring medical education and training. Stop supporting training that sends over half of all graduates outside of a state.
Specific interventions are required to produce a rural medical workforce. The starting point is creating a system that provides a healthcare workforce for rural populations. Most Americans need more primary care. With that as a specific goal, the design is obvious.
What we don’t need is what we have now — a system that funds individuals, institutions, and groups that have failed for 30 years to deliver what is needed.
Dr. Robert C. Bowman, M.D., founder of the Rural Medical Educators Group of the National Rural Health Association, is a physician and long time health education policy advocate. He is professor in Family Medicine at A.T. Still University School of Osteopathic Medicine in Arizona, a school that partners with the National Association of Community Health Centers.