Why is there a primary care shortage?

Health care reform is a locomotive barreling down America’s tracks. In two years, the Affordable Care Act (ACA) will cover some 30 out of 50 million of us that currently lack health insurance, provided neither the Supreme Court nor a new president overturns the law. Political beliefs aside, it would seem that supplying insurance to protect the health of more people is a societal good. Though the costs of reform will be debated for years to come, one major question remains that has not been adequately addressed:

Who will see all the new patients?

It’s no secret that there’s a looming crisis in primary care. Estimates place the shortfall of doctors at 30,000 in the next couple of years. Yet medical schools are flush with applicants. Residency slots are filling at higher rates than ever before as new medical schools have been chartered and class sizes have expanded. So where are all the new doctors?

In a word, the hospital.

“Hospital medicine is the fastest growing specialty in American medical history,” said Dr. Robert Wachter, chief of the division of hospital medicine at the University of California, San Francisco, and the man credited with coining the term “hospitalist” in 1996. According to statistics compiled by the Society of Hospital Medicine (SHM), the number of doctors practicing as hospitalists has increased 172 percent from 2003 to 2010. There are now more than 30,000 doctors nationwide that are classified as hospitalists: physicians who take care of hospitalized patients but no longer have office-based practices or do primary care.

To understand how difficult it will be to find a primary care doctor in two years, look no further than Massachusetts. In 2006 the state passed a health care law mandating that everyone obtain insurance (sound familiar?). For those unable to afford the cost, subsidies were made available.

Within weeks, the “uninsurance” rate in Massachusetts dropped precipitously. Commensurate with that was a rise in both the number of “closed” office practices and the length of time it took to get a new patient appointment. Nearly six years after the law passed, more than half of the family practice and internal medicine offices in the state are closed to new patients. According to the Massachusetts Medical Society, the average wait for a new patient to be seen by an internist is 48 days. Turns out insurance doesn’t guarantee access after all.

For young doctors just finishing residency, practicing as a hospitalist has many attractions. The most enticing aspects are financial and lifestyle considerations. A starting hospitalist (depending on what region of the country they practice in) can earn around $200,000 per year (a starting office-based internist will make in the neighborhood of $150,000). Perhaps more importantly, many hospitalist groups operate with “seven-on/seven-off” schedules. This means that a hospitalist earns that salary working seven consecutive days followed by seven days off. This option is extremely popular with doctors that are parents, as well as those that want to earn extra income or volunteer during their off time.

During the three-year internal medicine residency (like the kind I administer), doctors-in-training will spend about two-thirds of their time on hospital-based rotations. If familiarity breeds comfort, then it’s not a surprise that recent residency graduates choose to stay in an environment to which they’re well-adapted. And since hospital work is shift work, there is no on-call or after-hour responsibilities to handle. When a hospitalist leaves the hospital, they’re done — unlike office-based internists who still carry pagers and get middle of the night phone calls.

Couple the lifestyle and the training experience with the huge debt burden that U.S. medical students accrue, and deciding on a hospitalist career becomes a rational choice. Dr. Wachter of UCSF compares hospital medicine to site-based specialties that came before it: emergency medicine and critical care. All of these specialties represent a convergence of high-complexity and high-cost care in a single location, where it makes sense to have well-trained specialists who handle the specific set of problems encountered there.

Since the severity of illness seen by hospitalists tends to be high, specialization improves safety and quality, which are key metrics for hospitals as insurers now tie payment to such indices. Hospitals have almost all transitioned to hospitalist models to at least some degree. According to SHM data, the larger the hospital, the more likely it is to have hospitalists. Management likes the efficiency and improved patient satisfaction that comes with having doctors on the premises at all times. Earlier discharges and shorter lengths-of-stay for patients keep the hospital beds turning over and consequently the reimbursement dollars flowing in.

Contrast all of this to the realities of office practice: Fifteen-minute visits with patients on multiple medications, oodles of paperwork that cause office docs to run a gauntlet just to get through their day, and more documentation and regulatory burdens than ever before (e.g. new IT and compliance mandates). Students see the high pressure that primary care docs are under and are increasingly making the logical choice.

A colleague of mine recently sent shock waves through our community by leaving her internal medicine practice after 23 years to become a hospitalist. Her patients were devastated, as they had grown deeply attached to her. Yet with a child entering high school, my colleague felt that the seven-on/seven-off schedule and increased pay would dramatically improve the quality of her life and time available for her family. She was frustrated after spending all day seeing patients in an office only to come home and have at least two more hours of documentation to complete most nights.

Yet despite all of the negativity surrounding primary care, there are still holdouts.

Dr. Diane Fingold, an internist at Massachusetts General Hospital in Boston, gave voice to the downtrodden in a piece published in the New England Journal of Medicine in December. In “The Road Less Travelled,” Fingold wrote beautifully of her attachment (and the above and beyond care she provided) to a patient who’d suffered a stroke and was immobilized and unable to speak. These medical complications made it nearly impossible for the patient to advocate for herself when the pharmacy withheld her medications due to an insurance snafu. After a number of phone calls, Fingold succeeded for her patient.

It’s that deep commitment over time, all the ups and downs of her patient’s many hospitalizations that keep Fingold in the game. She writes:

I get the call and head over to the ED. As I pull back the curtain, a smile of recognition spreads over Mary’s face. She can relax now. She knows I care, that I’ll figure out her story and make sure the ED docs know all her meds, allergies, and complications; I’ll let her specialists know she’s here. She knows that if her medicines change, I’ll contact her pharmacy to ensure she gets a new blister pack. She lies back and breathes more comfortably.

Familiarity has built a fortress of trust between this patient and her doctor. Fingold concludes: “And at times like this, I recognize my deep satisfaction with the road I’ve chosen to travel.”

When I called Fingold, she told me that her hospital wants to transition as quickly as possible to a hospitalist system. But pockets of resistance remain. In her practice (which is on the hospital’s campus), all of the doctors see their own patients when they are hospitalized. In her case, proximity (and desire) allows her to be directly involved in her patients’ care. But for most of us, the luxury of having our own doctor treat us in the hospital is a thing of the past.

“Ultimately, I believe we will have to give it up,” Fingold told me. “I think it will be sad.”

John Schumann is an internal medicine physician who blogs at GlassHospital.  This article originally appeared in The Atlantic.

Comments are moderated before they are published. Please read the comment policy.

  • NewMexicoRam

    I can see that the demand for primary care will increase. Everyone wants more of what’s cheap.
    But medical students aren’t stupid–they’re medical students. That’s why very few are entering primary care. Until the AMA changes the board that helps the feds determine the reimbursement levels for different procedures and office visits, primary care will continue to get screwed and our nation’s health care costs will continue to skyrocket.

    • http://twitter.com/Cascadia Sherry Reynolds

      Perhaps what we need to do is stop having the Federal Govt (CMS) pay to train so many specialists and pay to train more primary care docs and give them a bump during training? We often hear about “medical school debt” but social workers manage to go to grad school and work for $12 an hour. (granted for 2 years less)

      FYI – What most people fail to realize is that CMS (Medicare and Medicaid) pay for almost 95% of all specialty training (9 billion a year) at a cost of about 100,000 per year per doctor. Residents are paid about 40k to 70k a year and the rest goes to the teaching hospital..

      That is the rational behind paying them less for treating medicare and medicaid patietns – they are esentially paying back the cost of their training (beyong medical school to become an internist or family practice doc). That cost to taxpayers is about 500k to 1.1 million (some surgical specialists take years) and they are paying back that grant/loan by accepting lower reimbursement rates.

  • http://www.facebook.com/robolivermd Rob Oliver


    Primary care isn’t getting “screwed” in reimbursement as compared to specialists, it’s just providing a service that can largely be done by mid levels (NP/PA) at 60% of the cost. In an era where every healthcare $ is increasingly being scrutinized, CMS & the payors have to look for value. There’s never going to be some huge salary surge for office based primary care E&M charges for that reason.

    • http://www.facebook.com/people/Steven-Reznick/100000549195050 Steven Reznick

      If you think that mid level providers are actually providing a similar service to what a well trained internal medicine or family physician can provide then I suggest you submit to a mini mental status exam or neuropsych testing along with a drug and alcohol screening.Yes when NPs and PAs were first introduced most of the candidates were experienced ER, Critical care and military war zone nurses who brought experience , depth and compassion to their training. Todays candidates coming from college straight to a graduate program really lack the clinical experience to provide the same quality of care a physician provided before the insurers kidnapped the profession and patients and sold them back to the docs at a fraction of their previous value. Now if Z Emanuel MD and Robert Wachter MD move forward with plans to reduce the length of medical school and residency training and separate docs into strictly hospital based and outpatient based categories they will both train a cheaper and dumbed down version of a doctor who may not practice at a higher level than a midlevel provider.

      • KCF

        The NPs with previous military, ER, and critical care experience are still out there. But I agree that they’re being outnumbered by those who go straight through without ever getting much valuable experience.

    • rswmd

      “a service that can largely be done by mid levels (NP/PA) at 60% of the cost”


      The cost of medical services is determined solely by the contract negociated with the insurer. It’s quite easy to see a NP in a large clinic with high fees and a facility charge, and pay three times what it would cost to see a doc in solo practice. Add in that the NP probably sees fewer patients while all the overhead remains the same, and there are no savings, either to the patient or the overall system.

      Then you factor in the quality issue . . .

    • http://onhealthtech.blogspot.com Margalit Gur-Arie

      “a service that can largely be done by mid levels (NP/PA) at 60% of the cost”

      That’s be cause the “service” has been redefined by financially interested parties to start and end at the practice door, and last no longer than 5-10 minutes at a time, with the remainder to be performed and billed by big systems. Unfortunately too many primary care docs played along either by necessity or because they were not paying attention.
      Now that the true “fragmentation” of care is taking its toll on costs and quality, the proposed solution seems to be more like a FedEx tracking system than anything else, so yeah, anybody can greet customers and take the package off their hands.
      Somehow I don’t think this will have the desired effects on people’s health. May fix the cost problems though.

      • glasshospital

        FedEx tracking–great metaphor for the PCMH.

        PCMH will greet customers, and process them through the machine. Then use bar coding to make sure we meet all quality metrics.

    • KCF

      I saw a PA at a neurosurgeon’s office. Clearly you did not do any research before writing this.

      • Close Call

        PA at a neurosurgeon’s office is much different than a NP right out of school that is allowed to practice independently and without any restrictions similar to an IM or FM doctor – which is what they are lobbying for. The PA in your example is still supervised by a physician.

        • KCF

          An unsupervised NP? That’s news to me. You might want to look into that again. You have some seriously wrong information. Plus unlike PAs, NPs can actually gain experience working as RNs before entering a higher level of practice.

          • Close Call

            “An unsupervised NP? That’s news to me.” I apologize, but I can’t tell if you’re being sarcastic or not. It is the goal of MANY state nursing associations to allow independent practicing and prescribing without physician supervision. There is much lobbying around this.


          • KCF

            I’m pretty sure the point I was making with my original comment was that all specialties have mid-level providers, so paying primary care less because they have them holds no weight. What people are lobbying for and what the future holds for NPs is a different topic. As of NOW there are no unsupervised NPs. Didn’t realize I had to spell it out for you.

          • ProudOkie

            Huh? NPs are completely unsupervised and may own independent practices in 17 states and DC. They have a very low level of supervisory requirements in many other states. Please don’t repeat your information. Any provider in the field knows NPs practice independently except for those who choose not to do so.

          • KCF

            Ok, I thought I was on a NJ blog site. So, yes, you’re right. There are other states with different laws. I still have to point out that your focus is misplaced. AGAIN I say that my argument is that it holds no weight to say primary care gets paid less because of mid level providers because every specialty has one. Here I am trying to make a small little statement in response to another comment and everyone goes off on a completely different issue. I am not trying to argue about the roles of NPs and how much autonomy they have. If you want to start a heated argument with someone about that there are TONS of other blogs, articles, etc to keep you busy. There are probably even other comments on here that have thrown that argument into the mix. Mine was not intended for that purpose. But thank you for helping me see my mistake. I will be sure to keep the blogs I follow straight in the future. State by state laws do vary. I am from New Jersey

          • ProudOkie

            KCF, My sincere apologies. I was unclear about what you were saying. Maybe should have approached more gently. Again, I’m sorry about the confusion.

  • http://www.facebook.com/jwcoppin Jonathan Coppin

    I am an MSIV and applying to family
    medicine residency in Texas. I came to medical school, in part, because I saw a
    need for good primary care doctors. All I care about in the way of money is
    being able to support my family, which I know I will be able to do with the
    smallest of doctor’s salaries. I do plan on doing outpatient primary care and I
    hope to be able to run my practice in such a way that I am still able to spend
    time with my family. I know it will be hard work, and I know there will be
    people who get paid more than me, but it’s why I’m becoming a doctor, and if I
    don’t do the type of practice that I think is most needed, that I am most
    suited for, then what’s the point of doing all this in the first place? Anyway,
    it is my hope that more students will step up to the plate and make the choice
    to be primary care docs. And, of course, a little slack for primary care
    physicians is welcome whenever they decide to get around to that.

    • buzzkillersmith

      Don’t do it man. Hear me now and believe me later. On second thought, maybe you should do it and then come the SE Washington and be my doctor since my own family doc pulled the plug. I’m laughing to keep from crying because now the hospital wants to run those depressing fam med dept meetings.

  • buzzkillersmith

    Long, eloquent article, completely off the mark. The hospitalist movement is a sideshow, although being an area of intense attention, it can fool the non-data-driven, like Dr. Schumann. If you want some actual insight, take a gander at the recent article by Grayson et al in Medical Education.
    Guess what. It’s the money! Yup, med students, no dummies, flee primary care for high-paying subspecialties, and those med students who have the most debt flee the fastest. Hospital work is a second-best for those who weren’t smart or lucky enough to jump ship earlier. Sure, hospital work is better than outpatient primary care for a lot of docs, but it is no nirvana either, kowtowing to the administrators and all that.
    All the smart young docs are not in the hospital as hospitalist scut-monkeys–they’re working as subspecialists there, those that aren’t subspecializing in the clinic.

    • http://twitter.com/Cascadia Sherry Reynolds

      you nailed it.. -we have raised a generation that wants to maximzine income and minimize personal cost. People go into medicine to heal (or to make their parents proud) but come out running a business..

  • tweets21

    Good reminder. As an American Ex Pat in Canada I have been entitled to use Ontario’s Universal Healthcare program. Have to report it runs as well as any Universal healthcare system can. Doctors however are a precious commodity. If you have a GP hang to he or she, as often you cannot get a new one if required. Physicians are not required to take new patients.

  • andymc12342003

    Primary care isn’t getting “screwed” in reimbursement as compared to
    specialists, it’s just providing a service that can largely be done by
    mid levels (NP/PA) at 60% of the cost.”

    Nonsense. At least 60 % if not more of the care in the dermatologist office where I go is done by midlevels. Should the dermatologist’s reimbursement be significantly less is because of this? Primary care is getting screwed in every way. Reimbursement, job stress, personal satisfaction and as stated already, medical students are following suit.

    • http://www.facebook.com/robolivermd Rob Oliver

      “Nonsense. At least 60 % if not more of the care in the dermatologist office where I go is done by midlevels. Should the dermatologist’s reimbursement be significantly less is because of this?”

      Yes, I do think you will have to discount patient’s not treated by the MD in these settings. You’ve got your head in the sand if you can’t see this coming. Provider reimbursement (be it MD or NP/PA) is one of the least politically painful choices CMS has to wring out savings from the system before they have to lead the charge of the light brigade into transparent rationing. While MD’s in general (and PCP’s in particular) feel they add value, you’re going to have to prove it or they’re going to look for cheaper labor. As it is going to be difficult (if not impossible) to show improved outcomes over cookbook medicine for most common medical conditions, you’re going to be be competing with independently practicing mid-levels for the healthcare dollars budgeted for these patients. You CANNOT win that fight as they will command fewer $$$. The assumption that much of the care for patients will be provided by cheaper providers is built into the models for every type of healthcare reform model.

      • rswmd

        An independently practicing mid-level has the same overhead (rent, staff, liability insurance, supplies, etc) as a physician. If they’re reimbursed at 60% of what MDs get, they’ll be out of business in 6 months.

  • glandsone

    If docs didn’t drink the EMR Kool-Aid they wouldn’t have hours of documentation to do at the end of their office work. Nothing is as fast as opening a paper chart and writing in it. EMR may bring efficiency to the healthcare system as a whole, but kills office productivity for the individual physician.

    • buzzkillersmith

      I for one did not take the Kool-Aid willingly, although my lunkhead partners did. I’d be chuckling at their foolishness but I don’t have the time because of all the documentation I have to do.

      • glandsone

        I hope you are living large with the 2% Medicare bonus. Are you the guys who are upcoding and costing them billions? Can I get a bonus for not using EMR?

        • buzzkillersmith

          Living large. Good one.

    • http://twitter.com/Cascadia Sherry Reynolds

      We implemented one of the large legacy largest EHR vendors and guess what the docs that used to stay late doing paper work were the same ones who stayed late with the EHR.. Those that understand how to use it chart during the visit are home earlier.

      The key however is changing your workflow at the same time. – otherwise is it just an electronic pencil. Change is hard but young docs won’t practice without one and it is the new standard of care. Can you honestly tell me you are ready for the malpractice lawyer that asks “so the only reason you didn’t get the lab results is because you didn’t have an EHR and that is why you missed the diagnosis”

      • Docbart

        If workflow is slowed by EHR, then that’s not really a good tool to use.

        Not using EHR is no excuse for not getting lab or imaging results. It is malpractice. If you miss results, prepare to be sued. I have 28 years of practice, none with EHR, no lawsuits. So I guess I have always been ready for that question. I am more likely to miss something important in the notes of EHR users because they have pages of useless template-driven text, with a few significant words hidden in them.

        If EHR is cutting malpractice litigation, why are my insurance rates not dropping?

  • http://twitter.com/BigIrisKC Bigiris

    An article that asks a question it never attempts to answer. I agree that more insured patients won’t mean more care for them or anyone else. But a bigger problem than whether new graduates choose to become a PCP or hospitalists is that in a hospital setting EVERYTHING COST 4-5X MORE than what it costs outside of the hospital. So, unless the taxpayer wants to subsidize hospital care at that multiple why would we want anyone going to a hospitalists for routine care? The locus of healthcare needs to shift away from this biased physician perspective to a consumer-patient driven system.

  • Bhavin Jani

    There is no continuity of care with Hospitalists and so much confusion amongst them and multiple specialists that medical errors and miscommunication will ultimately harm the patient. We need the old model where your doctor is at your service at office and the hospital

  • JannyPi

    Please remember that the “30 to 50 million of us that currently lack insurance” are already IN the system. Some were insured previously and had established relationships with primary care services.
    People get sick, regardless of their insurance status, and most of them seek treatment. The Patient isn’t NEW, they just have insurance now.

    • suger

      Exactly! 30 million people didn’t suddenly become hatched overnight. They have been here all along. They will simply have a means to pay for the services which as you mentioned some may have already had and now lost due to job loss, etc. Why does the medical community not understand this? Baffles me…

      • http://twitter.com/Cascadia Sherry Reynolds

        You got it.. Many if not most of the uninsured in Mass were only in the category for a short period of time after job losses and or the young healthy% and a small % who couldn’t get coverage because of pre-existing conditions. They weren’t however sitting on a shelf someplace. This might not be true in places like Texas or CA though.

        Mass has some of the highest costs and highest concentrations of doctors in the US and there were “shortages” before insurance reform. Doctors in fact drive almost 1/3 of the demand according to the Dartmouth Atlas studies.. The more Ortho docs in a city guess what more Ortho surgery.. No one has empty hours on their schedule..

        We know that up to 80% of all primary care visits are for time limited self resolving problems and another chunk are for wellness visits that for the majority of people show no difference in outcomes so I wonder is there really a shortage or just a broken monopoly dominated by providers?

  • Art_As_Social_Inquiry

    I learned that some top flight medical schools — Johns Hopkins, Yale, Harvard, Columbia, Cornell — do not have departments of family medicine — to train primary care docs. Because of Obamacare, the shift in medicine is to prevent disease. Family docs are taking center stage. I expect a shift in the medical schools. We cannot yet know how primary docs practices will morph but they are now assuming a very crucial role in the healthcare picture. They’re star is rising.

  • cookiebaker

    “Contrast all of this to the realities of office practice: Fifteen-minute visits with patients on multiple medications, oodles of paperwork that cause office docs to run a gauntlet just to get through their day, and more documentation and regulatory burdens than ever before (e.g. new IT and compliance mandates). Students see the high pressure that primary care docs are under and are increasingly making the logical choice.”
    I am a Family Nurse practitioner and I see that all of the practitioners are struggling with increased administrative burdens, less patient time and increased push for more patients visits per day. Most PCP’s do not like this at all and I suggest that primary care is losing providers in large part because the satisfaction of the work is withering. It is overload. In addition to less pay compared to specialists, the burnout levels are very high for the reasons that you cite, even for the most committed of us.
    I think that primary care needs to get figured out and improved ASAP because the need is there and we need to provide it. My suggestion is that the MD’s in primary care and the PA’s/NP’s all sit down and try to figure out how we can sustain meaningful work together. Money, training, administrative work are all a part of this picture, but so is the relationship between the providers. The NP’s do have excellent outcomes and patient satisfaction as PCP’s, so whether it is “cookie-cutter” medicine or not, the work is getting done, and done well.
    I would suggest two things:
    1. Look at successful primary care models from around the world and learn from them.
    2. Have primary care teams who all take care of patients on a shared basis and exchange information about patients both casually and formally. My supervisory MD is not really around and I almost never call for support because, often the supervisor is not available. I do call, or casually ask, MD colleagues for support, but they are not my supervisory MD. The supervisory role for me is essentially in name only. This is not to say that I don’t need MD’s to work with-I love having the support, but I also handle most of my patients independently, and that seems to be the norm for primary care these days. Everyone is really busy.
    Primary care is essential and we do have many smart and brilliant minds that we could put together to create a way to provide what is needed while maintaining our most precious base-patient/provider trust. We are a rich country. We have the resources. We need to make this a priority and build it.

  • andymc12342003

    “Nonsense. At least 60 % if not more of the care in the dermatologist
    office where I go is done by midlevels. Should the dermatologist’s
    reimbursement be significantly less is because of this?”

    “Yes, I do think you will have to discount patient’s not treated by
    the MD in these settings. You’ve got your head in the sand if you can’t
    see this coming”

    Different subject. Many would agree that a lower reimbursement is do in a particular individual visit performed by a mid-level instead of an MD, regardless of the specialty or clinic- but you noted that primary care was not getting screwed simply because- 60 % of the work could be done by mid-levels. Likewise, much maybe even 60 % of general dermatology could be done by competent mid-levels- it is in the busy practice where I go. So, certainly, their average salary (average derm salary is 283K for 2011) is exorbitant, and their reimbursement should reflect this, right? Sure, mid levels will play an increasing role everywhere but there is no ounce of proof that the will go into primary care in the numbers needed to fill the future need. By the way, one of the 6 or 7 recruiters that email me sent me a message about 2,700 FP practices that are actively looking for an MD- there was a tone of desperation in the email (and that was just 1 recruiter). The need is now and unless the corrupt nontransparent reimbursement system that screws primary care is fixed, I am not sure how to describe what will happen come 2014.

  • Guest

    Nothing factually incorrect with the story – but also nothing new – and no indication of a solution? Ok … so there’s 3.5 mins I’ll never get back. What a waste.

Most Popular