Skip to Main Content

Nearly three years ago, we began training as primary care doctors in two residency programs at a Harvard-affiliated teaching hospital. We understood the value of longitudinal patient-doctor relationships and wagered that primary care would be the bedrock of this nation’s health care system.

That was even after hearing the warnings: predictions of a national shortage of more than 44,000 primary care physicians (PCPs) by 2035, rampant physician burnout, and a workforce saddled with two hours of required documentation for every hour of patient care.

Nevertheless, we felt inspired to join the front lines of health care.


Yet when we finish our residencies on June 28, neither of us will be practicing traditional primary care. We are not alone in turning away from this field: Approximately 80% of internal medicine residents, including nearly two-thirds of those who specifically chose primary care tracks, do not plan to pursue careers in primary care.

We get it. The day-to-day practice of primary care feels daunting and unsatisfying. Primary care physicians need an estimated 18-hour workday to complete their various responsibilities. The much-maligned electronic health record inhibits PCPs from fully connecting with patients and creating the very patient-doctor relationships that once brought meaning to the work — the relationships that lured us into this type of medicine. Instead, our interactions with patients now feel transactional.


And despite the long hours and increasing demands, compensation in primary care trails behind that of specialty care. After four years of drowning in medical school debt and three years barely keeping afloat on a resident’s salary, the preference for becoming a specialist is entirely rational. The prospect of scrounging for reimbursement for our services from insurers and clicking through cumbersome electronic health records all day isn’t a tenable way to offload heavy student debt.

The practice and financial realities tell only part of the story, however. Primary care training is also complicit.

Primary care residencies based in academic medical centers do little to promote or incentivize careers in primary care. These residencies uphold the hospital-centric health care system. Outpatient training is often an afterthought, and dysfunctional experiences in the clinic taint trainees’ impressions of primary care.

Internal data from our hospital show that residents have a higher percentage of Medicaid patients in their panels than their preceptors do. Not only is it unfair for these patients, who often have complex needs, to be mainly seen by new physicians, but such clinical experiences can feel impossible and overwhelming for trainees. Because most primary care programs demand many more than the mandated 12 months of inpatient training, primary care residents struggle to fully integrate into their clinics, to build confidence with outpatient medicine, and to create longitudinal relationships with patients.

Our experiences belie a deeper existential crisis in primary care. The pressure on primary care is mounting: to coordinate care for an aging population beset by chronic disease, to improve the overall wellness of the population, and to control costs and eliminate waste, all the while ensuring a satisfying patient (and supposedly clinician) experience.

Is it reasonable to expect primary care to do all of this? To be all things to all patients? It seems like today’s primary care is being defined less by what it is and more by what the rest of the health care system isn’t — or doesn’t want to do.

As the demographics and cultural values of the U.S. change, traditional primary care is struggling to define its value proposition. Office visits to primary care doctors declined 18% between 2012 and 2016. Patients are increasingly choosing urgent care centers, smartphone apps, telemedicine, and workplace and retail clinics that are often staffed by nurse practitioners and physician assistants for their immediate health needs.

Millennials are choosing solutions that circumnavigate physicians, accessing direct-to-consumer diagnoses and prescriptions for conditions like hair loss, erectile dysfunction, and skin care.

Or consider frail elders, a growing subset of patients who require interventions to prevent falls, manage multiple comorbidities, and plan end-of-life care. Much of their care is being outsourced to geriatricians. Even when they are hospitalized and are at their most vulnerable, they are treated by anonymous hospitalists rather than their familiar primary care physicians.

And then there are patients with complex behavioral needs, ranging from mild anxiety to full-blown substance use disorders, often complicated by trauma, lack of housing, and food insecurity. Caring for this fragile population demands a high-touch collaborative effort by a team of social workers, community health workers, nurses, and doctors. Organizations like Cityblock Health are placing the community health worker — not the primary care physician — at the center of the care team as it works to address the social determinants that are driving health outcomes.

Solutions that exploit the inefficiencies of traditional primary care and fill in the gaps leave us uncertain about the future role of primary care physicians. We wonder whether it’s right or realistic to insist that primary care be the same thing for people of all ages across all demographics, geographies, and incomes. While we still believe that coordinated care that is longitudinal and based on relationships must be the foundation of a healthy America, traditional primary care is simply stretched too thin to provide it and is being eroded, circumvented, and replaced.

It is time we reimagine the role of the primary care physician.

While retaining its core value proposition — longitudinal, relationship-based, coordinated care — primary care must differentiate into narrower, more manageable scopes of practice. The doctor managing opioid use disorder alongside a community health worker is just as much a primary care physician as the one doing intensive behavior counseling patients with prediabetes, the one coordinating care and preventing hospitalizations for high-risk patients, or the one designing balance training programs for the elderly. Given the variety of expertise required in such care models, the primary care physician may no longer be the sole — or even the most valuable — player on the team. This counterintuitive “specialization” within primary care will reduce the current burden on current PCPs and better achieve primary care goals like prevention, efficient resource utilization, and population health.

How primary care physicians are trained must also change. That needs to start by ensuring that future PCPs are entering a profession with reasonable demands and loan forgiveness. During training, they should rotate through high-functioning patient-centered medical homes, hospital-at-home programs, telemedicine units, Teaching Health Center programs, and concierge care practices to experience different models and explore various versions of primary care.

Training in management, leadership, and advocacy skills is also a must — we can no longer separate the clinical practice of primary care from the practice model and the community in which care is delivered. Senior trainees should have access to scribes and technology-enabled support tools to maximize learning and patient interactions. And clinical experiences in cutting-edge primary care models should receive graduate medical education accreditation.

These innovations, while disruptive, are essential to address the dysfunction that plagues traditional primary care. Nostalgia for tradition is thwarting progress, driving burnout, compromising patient satisfaction, and adding stress to a fractured health care system.

Instead of waiting for payment models and policies to dictate how primary care evolves, primary care should focus — now — on making itself invaluable for both patients and physicians alike.

Richard Joseph, M.D., and Sohan Japa, M.D., are senior residents in primary care-internal medicine at Brigham and Women’s Hospital in Boston. Following the completion of their residencies, Joseph will be working in weight management and obesity medicine and Japa will be working as a hospitalist.

  • The design fails for half of Americans. A new design, the opposite of the traditional MD DO NP and PA design is needed. In the MD and DO area – Long ago there should have been family medicine only medical schools. There were some attempts in the 1960s and 1970s, but the financial design eventually took out even the most dedicated. Even some of the best at FM production like Mercer fell from 30% FM to 5% in about 5 years – and during the time of an FM dean. No matter how much the emphasis on primary care, rural, and underserved – these do not matter compared to more lines of revenue and the top reimbursement across research, graduate medical education, corporate ventures, foundations, subspecialized care, hospital care, and more.

    Health professional training is more about itself and little about health access – but it uses health access claims to further its needs. So the only solution is to part company and focus on the needs of most Americans and their basic health access. For the good of most Americans, for the good of their health access, and for the good of those doing the training put in ridiculous bondage by the training debt, there is a better way.

    Primary care only MD DO NP and PA schools with local only training are the remaining solution – as the financial design is so bad.

    How bad is the financial design stacked against most Americans???

    About 45% – 50% of MD DO NP and PA are concentrated together in 1% of the land area in 1100 zip codes with just 10% of the population. They are about as far away from most Americans as they can be. They are about as exclusive, academic and distant and unaware as they can be. They don’t know Americans or their basic health needs. Those who become leaders are filtered through institutions, associations, corporations, and foundations who think alike and formulate the same designs – without understanding the consequences to most Americans most behind already.

    In contrast, 45 – 50% of the US population has only 25 – 30% of the workforce. They have 1 or 2 lines of revenue which are most abused by payments 15 – 30% for the same services. They cannot cherry pick change their location for better patients. They cannot cherry pick the best health plans. They cannot change the worthless public and private insurance plans that abuse them and their patients. The states tolerate this and don’t listen to the abuses.

    Those behind get the worst of every design – shaped by those in higher concentrations. Even the last decade of reforms managed to extract another 10 – 20% of the dollars that they used to get – but now goes to consultants, corporations, CEOs, and others via HITECH, MACRA, PCMH, and value based designs. This squeezes their time with patients, their time with team members, and their time with families. They have had enough – 10 years ago.

    They get penalized more because they cannot afford certified EHR. They get penalized more also because they care for patients with inherently the worse outcomes. They have the worst collection rates. Some do additional side jobs to keep their practices going.

    Not surprisingly, their practices are closing, compromised, and burned out. The impacts are all worse on their team members asked to do more with fewer despite higher complexity. Patients do not see this, but their experience suffers as does their access.

    The local workforoce where most Americans most need care is predominantly generalist and general specialty, and this is 90% of local services. Any true reform should address these patients and this specific workforce.

    As you can see there are way too few by half because they and their hospitals and their patients receive the worst financial and other treatment by CMS and “payers.”

    Payments are a myth because they don’t pay or low pay – resulting in no access or low access by design. The payers not only don’t pay, they make life miserable for those in most need of care and those attempting to do their care.

    CMS and the states tolerate this abuse by design. CMS heaps on even more with their innovative plans and programs – read cost cutting.

    CMS is the main reason for shortages because they set the example with payments too low for each of the following – office services, cognitive services, person to person interactive services, primary care, mental health, women’s health (which is predominantly basic and annual care not about controversial areas), basic surgical services, rural health care, care in 2621 counties lowest in workforce concentrations.

    But even in the counties most concentrated in workforce, American designers have found a way to worsen access. Even that 10% of the population in top concentrations of health care workforce has Medicaid populations that have difficulty getting care – because Medicaid payers pay too little and so providers avoid Medicaid. Expansions of Medicaid are not that helpful, because the payments for basic services are still below the cost of delivering care.

    So of course there is a special federal shortage designation for such populations – which does not fix the problem because Medicaid pays below cost. You actually have to fix Medicaid and Medicare to pay above cost of delivery – to improve health access, care where needed, generalists, general specialists, and distributions of health care dollars and workforce.

    No one wants to fix the problems. Not fixing the problem works better. They are in the business of training more graduates. Unlike medical school leaders of 100 years ago, they do not care if they create a massive glut of workforce – and they are, all of them MD DO NP and PA. They are better off claiming to be able to fix primary care or rural health – so they can get even more dollars to train even more massive numbers of graduates which are already a glut of workforce.

    And despite the massive glut and expansions of annual graduates at 6 – 14 times the annual population growth rate – there is not even a discussion.

    As for Primary Care, there is a pecking order from most underserved and worst financed to most concentrated and best paid with lowest complexity.
    1. most underserved
    2. less underserved
    3. faculty
    4. administrative positions
    5. primary care with best patients, plans, and finances

    Beginning team members new to health care that do well can move up from 1 to 5 – or their can do even better outside of primary care in subspecialty or hospital positions – and they do.

    This also segments the patients, the team member numbers, the training of team members, and more.

    Reflect on about every health article that you see every day – and see all of the grand distractions away from true reform – real help for most Americans and those who serve them.

  • Yes, the AMA Masterfile does indicate that there are some class years where Harvard and other elite school graduates do the back door to family medicine – enter outside the match later. There are the same number entering via the match at graduation 0 – 3 as go into family medicine residencies a year or two later.

    Baylor’s mistreatments of family medicine included a department of FM and chair put together without working with the existing division of FM or director. Then they did it again by setting up a Chair and Department of Community and Family Medicine displacing the old department. Often the elite schools have bucked regulations and have not established FM departments. After all, they are elite and can do what they want. They also get special research pathways to get priority funding from NIH.

    Baylor promised Chairman Rakel 21 FTE of faculty and the number 3 job head of promotion and tenure. For decades the incoming FM chair had one chance to negotiate on behalf of the department – at the start. Rakel was successful in negotiations. I was the 15th FTE hired, but Camelot came to a rapid halt. About 4 months into my stay, I was told that my rural health efforts were being replaced along with most faculty and department plans. We were displaced as St. Lukes hospital decided to keep a million in GME dollars rather than giving it to the residency to care for residents. Baylor was apparently unable or unwilling to force St Lukes to ante up. The faculty melted away as Baylor dictated. It was a great time all too short.

    In my experience, two different family medicine departments violated Equal Opportunity Employment regulations because they found out their funding was being cut. Others made promises if we did special duties requiring travel or extra efforts. They did not honor their promises.

    Family medicine faculty have had to accept lower than market salaries, more ways for their job to be stretched, and God forbid if they had any passion such as getting doctors into small towns or various research areas. Promotion to assistant, associate, or full professor did not help salaries and neither did experience. Eventually you find out that you must leave to get a better paycheck, only to find out that the same situation exists at other places too.

    They are desperate to find faculty, but low on support – and essentially because primary care finances are worthless and primary care people are treated similarly. Of course they are the most valuable for health access – but no one treats them that way. FM moves more from no access to some and from some to better and is most important in counties without a hospital – a population growing at one of the fastest rates in the nation. But who cares about the future or health access for most Americans.

    And for those of you who think that academic faculty do little – I did studies of their activities and you would be blown away by all that they were a part of or made to be a part of.

    As bad as the salaries as a faculty, I doubled my income in my first full time faculty position as compared to the rural practice I was driven away from – by the financial design.

    The state of Texas as with other states continue to marginalize FM and steal special funding.

    Years ago, family medicine had to go direct to the legislature so that they could get funding designed for them – but stolen by the medical school.

    Long ago there should have been family medicine only medical schools. Even some of the best at FM production like Mercer fell from 30% FM to 5% – and during the time of an FM dean. No matter how much the emphasis on primary care, rural, and underserved – these do not matter compared to more lines of revenue and the top reimbursement across research, graduate medical education, corporate ventures, foundations, subspecialized care, hospital care, and more.

    There are numerous similar stories at other institutions, elite and more normal.

  • Ironies abound. Its interesting the authors are mentioning their choice to avoid primary care but fail to indict their own institution.

    Large academic centers have long embraced pcps as referral ‘monkeys.’ Hence the frequent interchanging between doctor and np.

    At harvard and elsewhere pcps are not rewarded for taking the time to navigate complicated medical management. Instead patients are referred to cardio, nephro etc where an np manages them and only procedurally by a cardiologist.

    If harvard wants to reward primary care then pay them so. An internist should be able to see 18 pt a day, in 20-40 appt slots and make over 200k comfortably. Without this effort by employers why would anyone choose primary care when crnas easily break 180 with 36 hour work weeks?

  • From this great article; ” The much-maligned electronic health record inhibits PCPs from fully connecting with patients and creating the very patient-doctor relationships that once brought meaning to the work — the relationships that lured us into this type of medicine. Instead, our interactions with patients now feel transactional” I think we have identified the root cause of the problem in attracting and retaining the much needed medical professionals.

  • I’m just a patient but I have to manage my chronic illness and have seen a lot of doctors – primary care and specialists over the years. I have come to highly value primary care over and above specialty care. I really want my primary care doctor to be well supported in every way, from financial, to all aspects of the work setting. Frankly, I am fed up with specialists and think that as more and more chronic care is being relegated to them, patients well-being is suffering. I don’t know what I, as a patient, can do about the situation, but I am concerned about it.

  • Salaries have some impact, but a focus on salaries as driving specialty choice is wrong – especially in primary care where you know there is lower pay.

    It is the financial design that shapes
    1. Salaries
    2. Benefits
    3. Numbers of team members
    4. Past training and experience of team members
    5. Team member turnover
    6. Future expectations of team members
    7. What team members say about primary care
    8. What students and residents and fellows see about primary care
    9. What we talk about, and more

    The business world is different as salaries are more attractive as a key factor in choice of that job. Primary care is different. Salaries have been lower and fixed.

    But even savvy business types realize that to go anywhere, you go with the firm that has a firm financial design. But the business types look at salary as shaping choice.

    Do yourself a favor – look beyond salary to the financial design to see what shapes attraction, retention, recruitment, productivity, more team members, better team members, better morale, and much more.

  • Once again, the financial design for primary care hurts any current types of primary care (MD DO NP PA) and will hurt any new or proposed types (assistant physician, pharmacist, others). The financial design for generalists and general specialists relative to hospital and specialized care drives people away from primary care of all types.

    Everyone has been compromised including the team members that deliver the care beyond MD DO NP and PA. RN and other costly and higher trained professionals have been moved away from primary care for some time.

    NP and PA have followed the financial design steadily away from generalist and general specialty positions where the finances are worse, the team members are fewest, and the team members are a mix of dedicated, burnt out, or not very productive – worsened by high turnover.

    All of the above, all, have been worse where half of Americans have half enough generalists and general specialists who are all that remains that provide 90% of local services. Yes, 70% of the health care services of the nation overall have been compromised for decades.

    Nurse practitioner proportions of primary care have fallen most dramatically.
    1. The funding for primary care minus higher costs of delivery does not support more team members of all types
    2. NP are only 60% active (HRSA Nursing workforce), much the same as RN
    3. NP and PA have had highest turnover rates between primary care practices and departing primary care for other specialties.
    4. Massive expansions with shrinking primary care finances drive even lower proportions in primary care.

    Now factor in that 40 – 50% of NP have 0 to 5 years of experience as a practicing NP plus the above. This is fact, not fiction. It is entirely about the massive expansions of NP graduates from 10000 to 35000 and still increasing by 2000 to 3000 a year. When you expand at 14 times annual population growth rates for decades, you will eventually inflate a bubble that will pop – hurting all associated with that bubble. When the bubble involves massive expansions of NP PA DO MD and minor sources, the explosion will be bad – for newer graduates especially.

    This contrasts with the 1970s and 1980s primary care residency program graduates with 30 year careers of higher volume, scope, and intensity.

    We are changing from the most to the least experienced workforce in all of the high turnover areas – primary care, urgent, and retail and possibly ER.

    Front line is least supported and most complex.

  • If we think rationally we can see that the best (least costly and most efficient) practice situation is one where all the expensive and most useful apparatus, as well as those who can use it, is concentrated in what you term the hospitalcentric model. Primary care is dead or dying as the number of physicians in this type of practice continues to shrink. This is a job which is being left to nurse practitioners. I also think the EMR is not maligned, it just stinks and shoud be tossed and redone.

    • Your ignorant commentary smacks of a total lack of understanding.

      Primary care value is highest when led by a physician. How many can manage dm ii with the variety of complicated medications available, the patients with ckd iii, the cancer patients the variety of chronic pain patients, afib, chf, pulm htn, etc etc

      In no way am i going to sit here and say my job is replaceable by an np.

      If you think it is, it is because your generation of doctors in primary care lack basic medical knowledge and have failed in their care of patients.

      I cannot mention the number of times I have had to make drastic changes in the treatment plan for my patients whose previous pcp left or retired and is over the age of 55.

      Its attitudes like yours that are bankrupting medicine.

    • Hi, I co-founded HealthCrew in 2013; we shut it down in 2015. I was also part of another company called DropletLife that shut down in 2016. Many failures, lots of learning. Those experiences are what got me interested in primary care.

  • In my experience, this accurate but incomplete summary is more a reflection of Harvard medical system bias against Family Medicine and describes more of the Internal Medicine primary care tract, with its continued emphasis on hospital style health care, depersonalization through worship of science, and lack of understanding of community needs. Sorry, but as a family physician I agree with many of the problems mentioned but little with the solutions that continue to subordinate primary care to “specialization” and lack of continuity we see today.

Comments are closed.