We need more generalist physicians. And a new health system so that they can be effective.

We don’t have enough primary care physicians in the US to properly (or, often and in many places, even improperly) care for our people. We all need a primary care doctor, and some of us need specialists. But rarely does a specialist have the time, interest, or expertise to manage all of your problems, and also do the recommended preventive work. Unfortunately, primary care physicians rarely do either, especially when it involves managing the care delivered by many different specialists, who can be unaware of what treatments the other is recommending and of how it might conflict with treatment they are giving. Doing this properly takes a lot of work and time.

For example, you mention to your cardiologist that you have knee pain. They say “I don’t do knees; here’s a referral to an orthopedist”. They’re done. If you go to your primary care doctor, they will examine you, try to figure it out and maybe treat it. But if they refer you to an orthopedist, they will follow up, read the report, look at the x-rays, read the recommendation, and try to explain to you what it all means. In short, a specialist referring to another specialist saves time and work; a primary care doctor referring to a specialist generates more work for themselves.

But there are not enough primary care doctors to do this, and those that exist are not given sufficient time by their employers (most are employed by hospital systems or large groups, often owned by private equity) to do it right. Because – and this is the most important thing – these hospital systems and especially private equity are, like insurance and pharma companies, in the business of making money, and only incidentally providing health care. Yes, many doctors are also interested in making money, but this is not their only or primary concern.

So, because most doctors are employed, and because primary care doctors may work much more for much less money, we have a dramatic shortage of primary care doctors in the US. A recent issue of Your Local Epidemiologist describes this in some depth.

One approach to primary care is that which Sweden has adopted, as documented here by the Commonwealth Fund.

Sweden is aligning primary care access goals with physician workforce policy more explicitly. Rather than treating physician training as separate from care delivery, Sweden is using workforce policy as a tool to strengthen continuity, accessibility, and person-centered primary care. To support this goal, the Swedish National Board of Health and Welfare analyzed primary care workload and physician capacity, established a benchmark of one primary care physician per 1,100 residents, and issued recommendations to guide regional workforce planning toward that target.

The Commonwealth Fund notes that

The United States faces a similar problem. Although physician training strongly influences where physicians ultimately practice and what specialties they choose, federal graduate medical education (GME) financing remains largely tied to algorithms based on hospital funding patterns established decades ago. The Centers for Medicare and Medicaid Services lacks authority to measure workforce outcomes associated with the nearly $21 million in Medicare funding that supports physician training. The result is a system that often produces workforce outcomes misaligned with patient needs, specifically not enough primary care physicians, general surgeons, and psychiatrists.

Commonwealth adds “Sweden’s recent reforms suggest a different model: define national access goals first, then align workforce policy around them.”

What a novel idea! Imagine figuring out what you are trying to achieve and then designing a system of achieve it! This is completely different from how health policy, including (or perhaps especially) physician supply and makeup is done in the US. (Well, except for for-profit companies who know what their goal is – make more money by any means necessary, which is never providing you with the best, most comprehensive and effective, or accessible, healthcare.) The system we use is to … do the same old thing. Pay subspecialists more, allow students to choose which specialty they want, and expect that somehow enough will choose primary care, despite the fact that they will make far less and often (not always, depending on the other specialty) works more, will almost be always employed by a large hospital system or group practice owned by private equity, and be on a hamster wheel of “productivity”, which means they cannot see enough patients, cannot deliver adequate care to the people they do see, and for themselves achieve limited professional satisfaction in having the control necessary to influence positively the health of their patients. 

As I have written before (e.g., Changes in the RUC: None.. How come we let a bunch of self-interested doctors decide what they get paid?, July 21, 2013; Pay primary care more: Kennedy may be getting this one right!, July 23, 2025 ), there is no mystery on how to change the income levels of primary care physicians relative to subspecialists. The Medicare payment “pie” is divided up between specialties mostly on the recommendations of an AMA-appointed group of physicians called the “RUC”, which is dominated by subspecialists. CMS, whichadministers Medicare, doesn’t have to follow the RUC’s recommendations, but usually does. And the RUC’s makeup of about 15% primary care physicians guarantees that primary care will be underpaid by Medicare. And, since almost all insurance company reimbursements are based on multiples of Medicare payment, this translates to the whole system. CMS could set up, or require the AMA to set up, a RUC that was, say, half primary care doctors!     

A word on the term “primary care”. While it has become the standard phrase in the US, and we can’t get away from it, it is not a great one. It was popularized by internists (adult medicine doctors) to distinguish those more properly called “general internists” from their internal medicine brethren who did mostly hospital work (“secondary care”) including intensive care (“tertiary care”), and then expanded to include general pediatricians and family physicians. Of course, family physicians are mostly all (90%) in what is now called “primary care”, and they (or the also commonly used “general practitioners”) are the mainstay of outpatient care physicians for the populations of most countries. Family physicians differ from other “primary care” physicians in that they are trained to provide comprehensive care for people of all ages, with all problems (referring when necessary) across the lifespan, including caring for children, old people, delivering babies, caring for folks in the hospital, in nursing homes, and doing lots of different procedures.

Unsurprisingly, it is hard to find even family physicians doing all these things now, except sometimes in rural areas where there are fewer physicians in other specialties. It has always been true that the privileges granted to family physicians have been largely determined by how much other physicians can make doing that work; many will be “first assistant” surgeons (fairly well paid) except when their communities have a glut of surgeons who “need” the work = income. Many years ago, it was discovered that only about 10-15% of graduating family medicine residents in Texas were interested in delivering babies in rural Texas; however, this was compared to 0% of OB-Gyn residents!

So let’s start with identifying what we want to achieve: enough primary care physicians, distributed broadly enough across the country in rural and urban areas, for every single person to have one. For those primary care (or family, or generalist) physicians to be paid enough relative to subspecialists that students are attracted to the field. For them to have enough time budgeted to manage complex problems and even more complex interactions between patients and other doctors. This is especially a need for older adults, who are most likely to be seeing multiple specialists for multiple problems. Relying on ERs or Urgent Care is a very poor and inadequate alternative.

That requires a complete reworking of how our health system functions. Covering everyone, such as by the improved and expanded Medicare for All system called for by the S. 1506 and HR 3069 bills, is an absolutely necessary part of this, but is not alone sufficient. We also need to restructure physician payment to encourage enough primary care that people can easily access it and it can be effective, implementing completely Starfield Four Pillars of Primary Care: [first] Contact, Continuity, Comprehensiveness, and Coordination (Starfield B. Primary Care: Concept, Evaluation, and Policy. New York, NY: Oxford University Press; 1992). If ANY of these are missing or incomplete, we have an inadequate health system – as we in fact do.

No messing about at the edges. No "reforms" that reform little. No saying we cannot afford universal health care, something every other developed country does, because our insurance companies need to make huge profits. No denying the need for an immediate steps to ensure an adequate primary care workforce because we don’t want to pay subspecialists less.

This must happen now, happen yesterday. There is no morally acceptable alternative.

 

                                       

We need more primary care to serve our people: Why do the medical schools lie? (reposted from April 5, 2021)

I recently re-read this blog post from 2-1/2 years ago, and decided that it was still important as well as relevant and accurate. So, in a "first" I am reposting it, since some folks may have missed it:

Every year the nation’s medical schools graduate thousands of people with MD and DO degrees. But this is just the start of becoming a practicing physician; they now need to complete residency programs in a specialty area, ranging from 3 to as many as 8 years, to become family physicians, surgeons, radiologists, dermatologists, orthopedists, etc. Indeed, for many physicians this “postgraduate” training (meaning post-medical school, since medical school itself is post-graduate, requiring a bachelor’s degree for entrance) can have two components as well. First there is the primary residency program, say an internal medicine residency of 3 years, and then there is subspecialty training, usually called “fellowship”, where that internist becomes a cardiologist, or endocrinologist, or pulmonary medicine physician. While the internist who completes 3-year residency may practice general internal medicine and thus become a primary care physician for adults, those subspecialists do not. A similar process exists for pediatrics. Family physicians completing their 3 year residencies can also do fellowships in a limited number of areas, and some limit their practices to sports medicine or geriatrics or adolescent medicine, but most add these skills to their primary care practice. And, of course, geriatrics and adolescent medicine are, like general internal medicine or general pediatrics, primary care for a particular population.

This is important. Primary care doctors provide care for their patients that is comprehensive and unrestricted, other than by age for pediatrics, internal medicine, and geriatrics. They meet the World Health Organization (WHO) criteria for primary care, providing continuous, comprehensive, community-and-family-centered care. Distilled down, this means that primary care physicians see their patients for everything, whatever concerns them, referring when needed. They are the doctors for their people, not for a particular disease or set of diseases. The lack of sufficient numbers of primary care doctors has significant negative impact on the health of our people. Of course, it falls hardest on those who are always most disadvantaged – the poor, members of minority groups, and rural residents. But it also has negative impact upon the health of privileged people who see lots of subspecialists, in two ways. One is that the specialist may be expert in their field, but miss appropriate treatments, and especially preventive measures, outside it. The other is that many specialties and subspecialties rely on and extensively use care that is very high-tech and expensive, which can lead to people getting tests and treatments that are not only costly but may not be of any benefit, and indeed may lead to harm.

 So, when a medical school claims that it is good at producing primary care physicians, this is serious, and should be accurate. But it usually is not, because schools want to look as good as possible so establish criteria that make them look good, counting a wide variety of specialties that their graduates might enter as “primary care”. The biggest “offender” in this regard is counting all graduates entering internal medicine residency programs as entering primary care. As described above, some of these end up doing fellowships to become subspecialists and do not practice primary care; indeed, “some” is an understatement as it is about 80%. In addition, about half the rest end up practicing as “hospitalists”, taking care of hospitalized patients only, rather than practicing primary care. So an approximation would be to assume about 10% of those entering internal medicine residencies will practice primary care. In pediatrics, continuing as a general pediatrician is much more common; the appropriate multiplier is probably 60%, and for family medicine as much as 95%. There are also residency programs in a combination of medicine and pediatrics (Med/Peds) which can produce primary care doctors, and whose graduates are less likely to pursue subspecialty training; however, they are very likely to choose only one of those areas (adult medicine or pediatrics) and also to become hospitalists.

In addition, some (or many) schools include in the primary care numbers specialties that are simply not primary care at all. Most commonly, they include emergency medicine and obstetrics/gynecology. Emergency medicine does indeed provide first-contact care, but it does not provide continuity. Obstetrics/gynecology can provide some aspects of primary care (and indeed OBGyns may be the only doctors some young women see) but it is limited in that it is not comprehensive; women are more than their reproductive tracts, and they can have a variety of conditions OBGYN does not care for (diabetes, hypertension, heart disease, depression, arthritis, asthma and other lung problems, substance abuse, etc., to name a few). Perhaps the most egregious abuse is counting all students who enter internal medicine “transitional” or “preliminary” years. Such one-year programs, which have replaced the old “rotating internships”, are required for many specialties such as neurology, anesthesiology, radiology, ophthalmology, dermatology, and others, whose practitioners do not do primary care at all.

If we want to know how well a school is doing in graduating students who actually practice primary care at the end of their residency and fellowship training, these inflated numbers do not inform us. Fortunately, one of the most popular sources of information on medical (and other) schools, US News, has worked with the Robert Graham Center, the policy center of the American Academy of Family Physicians (AAFP) to develop and publish a metric that does show which schools actually produce primary care physicians, available at https://www.usnews.com/best-graduate-schools/top-medical-schools/graduates-practicing-primary-care-rankings. The top of this list is dominated by schools of osteopathic medicine, which consistently graduate higher numbers of primary care physicians, and, among the allopathic schools, the mainly public schools who have been doing well in this area for a long time. The private, largely northeastern, schools that usually top rank lists are nowhere to be found.

It is important to look at this list, not the list of “Top Primary Care Schools”, to get accurate data on production of primary care physicians. The metric on percent of students going into primary care has also been fixed in the “Top Primary Care” rankings, so it is better, but it still only accounts for 40% of that ranking. “Peer Assessment” (subjective rankings) account for 30%, half from medical school deans and other leaders, and half from residency directors. The other 30% is half “faculty resources” (largely faculty ratio) which may be skewed to the advantage of research-intensive schools, because it includes faculty who are mostly in laboratories and not teaching, and half “student selectivity” (based on student grades and MCAT scores), which is actually negatively associated with entry into primary care. This doesn’t mean the students that enter primary care are not as smart; it means that the cachet of attending a research-intensive school makes the competition greater. Unsurprisingly, adding these other criteria does affect the rankings; Harvard, for example, is now #8 in “best primary care schools”, although it ranks #141 of 159 schools in percent of graduates practicing primary care. (In contrast, the University of Kansas, which ranks #9 in primary care, below Harvard, ranks #17 in graduates practicing primary care, at 37.8%). Reputation affects peer assessments in at least 3 ways. One is spillover effect -- well, it’s Harvard, and good in everything so it must be good in primary care. A second is the ignorance of non-primary care deans and residency directors about what kinds of doctors the school produces. Finally, the fact that “good in primary care” can mean things other than what specialties the graduates enter can have an effect; there are schools in which the family medicine and other primary care faculty are well-known for their research and leadership in national organizations, but which do not graduate very many students into primary care disciplines.

The fact remains, though, that the US very short of the primary care doctors it needs to provide quality health care to the American people. The way to begin to change that is to stop deceiving ourselves. Then we can start the process of producing a higher percentage, in every school.

What is the problem with Primary Care? The US health system!

What is wrong with primary care in the US? Shall I count the ways? Medscape details a number of them in its recent article, citing much of the data provided in the Commonwealth Fund report “Mirror, Mirror on the Wall” which I discussed in my last blog post, Our health system: Not equitable, not effective, and not even efficient. Bad business!, (March 4, 2022). The spoiler answer is: what is wrong with primary care in the US is the US health care system – how it is designed, how it is implemented, the purposes for which it is intended, and the intrinsic corruption of it. If the primary care portion of the US health care system is in particular disarray, it is because it is the (relatively) poor stepchild of a system that is all about making lots of money for corporations, particularly large health systems, insurance companies, and the vendors of drugs, devices, and equipment. The way our health system is currently structured is to feature those parts of it that generate this money, rather than those that maximize the people’s health, and under our current reimbursement system primary care is not in the game. Thus, it is unsurprising – if incredibly depressing – that our primary care sector performs poorly on the metrics assessed by Commonwealth (and reported on by Medscape), because they are looking at different markers, that is, how it meets the health needs of our people, and a robust and effective primary care capacity is critical to that. If only they would look at corporate profit they would see how well the health system, by neglecting primary care, is doing!

What is primary care and what are primary care doctors? They care for all the issues that a person has, not limited to disease, organ system, procedure, etc. They care for people with as-yet undiagnosed problems, with undifferentiated conditions. They provide care over time, and consider the physical, mental, and social conditions affecting a person. They provide care in the context of a person’s family and community. Any issue that is affecting a person’s health, or that they think is, is fair game to bring to a primary care doctor, who will try to diagnose and treat it, referring if necessary. In a coherent and effective health system, they continue to be involved with the person, even after referral or hospitalization. The characteristics of primary care, and the reasons for its benefit to people and to society are discussed most clearly by Barbara Starfield, MD, in many papers including this one. I like to think that while the relationship between primary care doctors and their patient is defined by the relationship, not the disease, or procedure involved. Family medicine, unlike even other specialties in primary care does not even limit its practice to certain age groups. But even these doctors are being relegated to practice only part of what they could; few deliver babies, most don’t do hospital work, and a large number do not care for children.

But few of us have seen such a physician lately, still less with a “full scope” practice. There are not enough family physicians or other primary care doctors in the US. There are not enough to meet the primary care needs of our people, nor to adequately perform the role that primary care should play in regard to specialists – that is, assessing a patient, determining if they can be treated by the primary care doctor, and if not referring. Otherwise subspecialists spend a lot of time caring for things that could have been done by a primary care physician. Or missing problems that are outside their specialty focus when people directly self-refer. And it is not only in the US; in parts of Canada, there are such shortages of primary care doctors (there they are virtually all family physicians) that consideration is being given to a new profession, possibly called associate physicians. In the US, much primary care is delivered by nurse practitioners and physician’s assistants. Some of them do excellent work, but they are also hampered by the same constraints as those primary care physicians face – excessive workload, assembly line production, (relative) underpayment, and a perverted reimbursement system.

To the extent that the move to non-physicians is driven by the fact that they cost less because they earn less money, any such effort is doomed. Nurse practitioners are increasingly being recruited by hospitals and subspecialty physician groups where they can earn, as do the doctors in those specialties, more money. This has overwhelmingly already happened in the case of physician assistants. The answer to the need for more primary care is simple: PAY MORE MONEY. Pay them as much as, or almost as much as (70% would probably do it) other specialists. There are a lot of students, residents, doctors, nurse practitioner and physician’s assistant trainees who would like to do primary care, and would be good at it, but are dissuaded because they can earn WAY more in another specialty. It is not that complicated; virtually all reimbursement for health care in the US is based on Medicare rates; private insurers pay some multiple of what Medicare pays. So all that has to happen is for Medicare to completely revise its reimbursement schedule so that primary care is paid a lot more, and interventive procedural specialty care less. Don’t increase the size of the pie; reallocate!

Sadly, the reallocation (under both Republican and Democratic administrations) has been instead to increase the privatization of Medicare, effectively enhancing corporate profits rather than quality health care. The Medicare Advantage program, while it can be good for some seniors, is being touted as the greatest thing since sliced bread by many in Congress, although it is heavily subsidized and saps funds from Traditional Medicare (TM). MedPac (not a “political action committee”, but the official group convened by Congress to make recommendations on Medicare) has raised serious concerns about the program, which essentially cherry picks healthy seniors, gives them low cost benefits, and eschews sick people while getting more money from Medicare. As I have written before (Direct Contracting Entities: Scamming Medicare and you and bad for your health!, Feb 7, 2022), a program called Direct Contracting Entities (DCEs) was developed to push even those who have chosen TM into corporate controlled profit centers. And now, after DCEs have received criticism in Congress, they haven’t been abandoned, but re-branded as REACH, essentially the same model.

The problems with primary care are not with the clinicians. The problem is with the corporate model that seeks to limit the practices of the clinicians and speed up their work so they cannot provide the benefits of primary care. The key part here is being the core person who knows about you and your family and manages directly or in conjunction with others all your care. It cannot effectively happen if you are seeing different doctors in every setting, and no one is responsible for YOU. This is much different from being the person who orders the tests or prior authorizations. Family physicians and other primary care doctors and clinicians need to have the time to spend with the patients, getting to know them, getting to know them well enough that they are trusted by their patients, who may then reveal the Pandora’s box of complicated, difficult-or-impossible to solve problems that physicians dread to hear about and corporate employers hate to pay for. You can’t get to these, not to mention begin to solve them, in 15 minute visits. Often you can’t really begin to solve them at all, since they are based in the overall circumstances of life that people find themselves in, what are often referred to as the "social determinants of health" -- their income, jobs, education, housing, food, safety, and discrimination for starters. But they need to be revealed.

This is scary to corporate types, who want to continue to do what they do – generate big bucks by hiring procedural specialists to care for well-insured or rich people for big reimbursement.