An Indecent Proposal That Just Might Solve the Primary Care Crisis: Meet the 35 Hour Work Week

A few weeks ago, The Health Care Blog published a truly outstanding commentary by Jeff Goldsmith, on why practice redesign isn’t going to solve the primary care shortage. In the post, Goldsmith explains why a proposed model of high-volume primary care practice — having docs see even more patients per day, and grouping them in pods — is unlikely to be accepted by either tomorrow’s doctors or tomorrow’s boomer patients. He points out that we are replacing a generation of workaholic boomer PCPs with “Gen Y physicians with a revealed preference for 35-hour work weeks.” (Guilty as charged.) Goldsmith ends by predicting a “horrendous shortfall” of front-line clinicians in the next decade.

Now, not everyone believes that a shortfall of PCPs is a serious problem.

However, if you believe, as I do, that the most pressing health services problems to solve pertain to Medicare, then a shortfall of PCPs is a very serious problem indeed.

So serious that maybe it’s time to consider the unthinkable: encouraging clinicians to become Medicare PCPs by aligning the job with a 35 hour work week.

I can already hear all clinicians and readers older than myself harrumphing, but bear with me and let’s see if I can make a persuasive case for this.

The crisis we face

First, consider the situation:

The most pressing and urgent health services research problem society must solve is how to restructure healthcare such that we can provide compassionate, effective healthcare to an expanding Medicare population, at a cost the nation can sustain.

This is a problem with very high human stakes at hand. As we know, most older adults end up undergoing considerable health-related suffering at some point, with family caregivers often being affected as well. Much of this is due to the tolls of advancing chronic diseases, such as diabetes, heart disease, COPD, arthritis, dementia. (See this handy CMS chartbook for the latest stats on chronic disease burden in the Medicare population.) And a fair part of the suffering is inflicted by the healthcare system itself, which remains ironically ill-suited to provide patient-centered care to those medically complex older adults – and their caregivers — who use the system the most.

Needless to say, the financial stakes are high as well, with projected Medicare expenditures usually cited as the number one budget buster threatening the nation’s financial stability over the next 50 years.

A necessary part of the solution

Next, consider an essential component to compassionately and effectively meeting the healthcare needs of the Medicare population:

Medicare beneficiaries – and their family caregivers – must be partnered with good PCPs who can focus on person-centered care, and can collaborate with them as they navigate the many health challenges of late life.

Especially once they are suffering from multiple chronic illnesses and/or disability, seniors – and their families — need a stable relationship with a clinician who can fulfill the role of trusted consultant and advisor as they go through their extended medical journey. Healthcare for older adults almost always becomes complex and stressful for seniors and their families. Even educated and activated patients who are willing and able to direct their own care will need a generalist who can maintain a longitudinal health dialogue with them, and who can help them sort through complicated medical situations as they arise.

Now, much as been made of teams in primary care, and the importance of moving past our historic model of PCP as the person who knows it all, and does it all. This change is long-overdue, and I’m thrilled to see it coming. When properly implemented, I’m quite sure that team-based care will help older adults obtain the comprehensive primary care services they need and want.

But even with excellent team-based care, I believe most older adults will want and need a PCP to function as their high-level medical strategy consultant and collaborator.

Common challenges for PCPs of older adults

For instance, consider the kinds of issues I routinely addressed as a general internist for older adults:

  • Following up on 6+ chronic conditions and 12+ medications, in an integrated whole-person fashion. Good luck outsourcing this to disease management.
  • Following-up on the work of multiple specialists, many of whom hadn’t explained their thinking to the patient and family. Yes these specialists should get better at explaining their thinking. No, they will probably not resolve the conflicts between their recommendations and some other specialist’s recommendations.
  • Resolving the conflicts inherent in attempting to follow clinical practice guidelines in patients with multiple conditions. For a fun read on how elderly patients routinely generate a gazillion conflicting clinical practice guidelines, read this JAMA article.
  • Adjusting care plans as a function of goals and what seems feasible for the patient. It is pointless to recommend chronic disease management per best practices if it doesn’t seem feasible to the patient and family. Also, many disease management approaches must be modified in the face of conditions such as dementia, cancer, advanced COPD, etc. I’ve spent my fair share of time taking diabetics with mild dementia off sliding scale insulin regimens. (Hello endocrinologists, please stop recommending labor-intensive blood sugar management.)
  • Explaining why certain commonly requested interventions – antibiotics, diagnostic tests, specialty consults – might not be helpful. People have questions. Answering questions takes time and attentiveness. It’s obviously much easier to rely on the historic approach of doctors and just tell people what to do, but that’s not good care.
  • Helping patients and families prioritize and identify a few key health issues to work on at any given moment. Many older patients have 15 items on their problem list. Prioritizing is key. (Not losing track of all the issues is hard though.)
  • Helping patients and families evaluate the likely benefits and burdens of possible medical approaches. Should that lung nodule be biopsied? Should knee replacement surgery be considered now, or still deferred? So many of the decisions we face have no clear right answer.
  • Helping patients and families cope with the uncertainties of the future. For instance, it’s impossible to predict how quickly someone with dementia will decline and become unable to live at home, but these issues are of grave concern to families and they need a clinician to talk to.
  • Addressing end of life planning. I’ve found this is often trickier in the outpatient setting than on an inpatient palliative care service.
  • Weighing in on family conflicts. I’ve had to watch patients and spouses squabble in the visit over what the patient is and isn’t able to do. Similarly, adult children worried about a parent will call and ask for me to intervene. (Stop her from driving! Make him take his pills!)

I must say that I love doing the work above. It’s deeply satisfying to help patients make sense of all that is medically happening to them, and to support them as they cope with their health challenges. But it’s also, as you can imagine, difficult work that is cognitively and emotionally demanding. The pressure of 15-20 minute visits makes things harder than they should be, but even if we went to 30-45 minute visits, the work will remain fundamentally intense and somewhat taxing for the provider.

Can anyone seriously argue that we won’t need PCPs to do the work above for Medicare beneficiaries over the next 20 years? (Plus we’ll need them do manage dementia, falls, and all the other geriatric problems too.)

Ok. Then if we agree that the work above is essential to the wellbeing of millions of older adults, and is a crucial component to providing overall cost-effective healthcare to the Medicare population, we must get serious about how we can recruit and keep clinicians as Medicare PCPs.

The benefits of a 35 hour work week

If the work of  Medicare PCP could be organized so that it fit into a 35 hour work week, we’d see the following benefits:

  • More clinicians would be willing to do, or stay, in the job. Let’s face it, we have ample evidence that work-life balance is important to the younger generation of physicians, especially those with young children. As much as this dismays the older generation of physicians, this trend seems to be here to stay, so perhaps we should learn to work with it. Debt relief – the usual hope for attracting people to primary care – is never going to be enough on its own.

Given that we are asking PCPs to actively engage with patients and families, embrace shared-decision making, adapt to technological changes, and make a whole host of behavior changes, making sure that clinicians in this role aren’t burnt out by long working hours just makes sense.

Summing it up

The impending shortage of PCPs constitutes a national emergency. In order to provide the growing Medicare population with compassionate, effective healthcare at a sustainable cost, seniors will need stable relationships with PCPs who can function as their strategic medical consultants, collaborate in helping to meet healthcare goals, and provide emotional support.

Doing this type of PCP work can be extremely rewarding, but it’s also cognitively and emotionally demanding.

Structuring the job of Medicare PCPs into a 35 hour work week would probably attract more clinicians to the job. It would also help PCPs maintain the cognitive and emotional resources needed to do the job consistently well, and could reduce burnout in this group of key clinicians.

Leslie Kernisan, MD, MPH, has been practicing geriatrics since 2006, and is board-certified in Internal Medicine and in Geriatric Medicine. She blogs at GeriTech.

34 replies »

  1. Agree that many employers are providing very poor working conditions to salaried physicians. This is bad because:

    1. Makes it very hard for us to provide good care to patients, esp the complicated patients who most need MD-level care
    2. It’s hazardous to our own health & wellbeing; bad for us and in the long-run bad for patients too.

    What we need to do: push back against these terrible working conditions! It is expensive for the employer to hire someone and have them later quit, or take extended leave due to burnout. (I know many PCPs who have taken fairly sudden leaves due to depression/stress; leaves the clinic in a pickle!)

    There are lots of ways to push back, although many of them are hard for docs in the short-term. It’s not financially feasible for most people to quit, and often all the employment options stink. Also, when we pull back from our work, in the short-term patients lose access (although the quality of what they were accessing via an unhappy doc is highly debatable).

    Still, I know that big employers such as Kaiser and GroupHealth have made changes in response to the concerns/complaints of their PCP corps.

    I do think it’s important to highlight that we cannot take good care of the patients, and do what everyone is asking us to do, without the right kind of work conditions.

  2. I would like to take into consideration the shift of private practice to employed practice in primary care. More than two thirds of primary care physicians are employed. Basically because of ridiculous red tape, reimbursement and bureaucracy have all but destroyed thousands of small practices in the U.S.
    Many of us are working 80 hours per week or more in order to honor our contracts.
    This is where I question the so called “primary care shortage” of which the evidence is mixed at best.
    You have a lot of big players, such as big medical groups and hospitals drooling over a surplus of primary care physicians competing with each other so that they can basically turn you into an employed slave. This is already happening. The message is simple “be a slave or unemployed”.
    This may sound eerie, but you have to face reality. The worst part is that the people that should be in our corner actually encourage this.
    The AAFP has this amazingly exaggerated propaganda to increase the number of physicians in the U.S. a strategy that has been both debunked and discreddited several dozen times. But I wonder if the increase in revenue and political power that comes from an increase in sheer numbers family docs to members of AAFP chapters influence their opinions at all?

  3. Actually, Peter, we need to move it from HOSPITALS and from technology related payments (eg. Part B technical fees) to primary care physicians.
    There are lots of specialists (the 24/7 kind) who are ALSO not replacing fast enough- general surgeons, etc. Even specialists are underpaid for making the correct diagnostic call, and overpaid for using technology.

    You’re right about moving the money.

    Also, part of the reason why managed care might play a part, if we can get the models right, is that they move money from unnecessary use of the health system (e.g. avoidable or preventable use) to primary care physicians by paying them better/differently. . . .

  4. Exactly. Payment reform has to come first if we want to save primary care. Your plan would give PCPs the option of working long hours and making a salary more in range with what their peers make, or cutting back hours, spending more time with patients, and still making an acceptable salary.

    Sadly, I don’t see it happening.

  5. True indeed. I believe that adjusting the RVU reimbursement rate for primary care upwards by 75% would go far towards the key goals here: more talented docs would see primary care as a viable career choice, and at a reasonable schedule of 40 to 50 hours, 35 for those perhaps sharing the income burden with a spouse. Hopefully the pcmh movement as well as the HIT standardization efforts underway will reach maturity such that consultant-partner-in-care-pcp’s can finally rely on a team and decision support tools, not just tolerate the hodge podge we have now. But a very smart first step would be ramping up the per-RVU payment rates for primary care – something which portends great possibilities and little downsides.

  6. Pay should be based on a per patient basis. It would be similar to capitation and be based on a much simpler version of the HCC Risk Adjustment model. The pay will not be PMPM mumbojumbo that confuses PCPs but a yearly amount. (Frequent flyers can be trued-up/ reconciled to pay the PCP if the patient is an outlier).There should be 5 to 10 “disease groups” and a physician would be paid a yearly fee to take care of each patient. A relatively healthy patient would pay $750 per year and a patient with 3 or more disease groups, including mental health, could range from $1,750 to $3,000 per year. Health plans would allocate a nurse to meet the patient in the PCPs office and the nurse will do most of the legwork. This is the way to go but Public Health Administrators cannot comprehend common sense. This needs to be proposed by physicians. Feel free to contact me regarding this revolutionary solution to the Medicaid/Medicare Duals Demonstration Project … promedpractice@yahoo.com

  7. Excellent suggestions. I would add dumping the PCMH concept as it is currently defined: it’s 95% administrative busy work that benefits no one.

  8. I really appreciate the kind words at the beginning of Leslie’s post. This thread is the heart of the matter. When I talk to a lot of my physician friends who are working seventy hours a week, what I learn is that they are spending about half their time with patients (e.g. precisely 35 hrs a week).
    The rest of the time revolves around documentation, authorizations and the dance of mistrust with insurers and Medicare.

    We need to dramatically simplify how physicians are paid by moving primary care to a subscription model. We need to subject physician reporting requirements to a cost/benefit analysis and eliminate so-called “core measures” that don’t measurably reduce patient risk. To do this will require a “base closing commission” for physician reporting requirements.
    We also need to end PQRI and meaningful use, and cancel the planned implementation of ICD-10.

    Unless we can dramatically increase the percentage of time physicians actually spend with their patients, we are going to have an horrendous shortage of physicians just when we boomers need them the most.

  9. I left primary care not because of low pay, but because of excess stress and unmanageable expectations. (Like that I see 20 patients per day, and not focus exclusively on frail patients because that would mess up my productivity; that I was actually good at helping frail patients didn’t count for much.)

    But many younger docs go into something higher paying because then you can still make a decent living while working less than fulltime.

    There are lots of ways to potentially work out the money part, but society would have to start by believing that this is an important job for docs to do, and that the value of a doc is not measured by how many patients they can churn through in a week.

  10. I just don’t see how a 35 hour work week with longer appointments, which would have to be coupled with even lower pay than what PCPs now get, is going to attract more students. My point is that it’s all about the money, with hours only being a minor secondary issue.

  11. “A shorter work week and longer appointments also means a dramatically lower income.”

    Not if they’re paid salary with pension etc.

  12. I’m proposing the 35 hour week carrot specifically for those docs who will be PCP to the medically complex Medicare patients.

    We should probably solve the primary care workforce question differently for Medicare than we do for the rest of the population.

  13. I’m still confused about what the poster is proposing. A shorter work week and longer appointments also means a dramatically lower income. No one is advocating payment changes in the near future dramatic enough to alter that equation. The overwhelming reason students are shunning primary care, particularly gerontology, is low pay. Seems like this proposal would just make things worse.

  14. The money’s in the system, we just need to move it from specialists to PCPs (as a matter of policy), that would recognize the PCP’s value. The system would probably save money overall by reducing medical opinions of specialists because that’s how they make their living.

    But going to a 35 hour week means less patients run through the grist mill, it seems we’d need many more PCPs for that. Not sure in Obamacare where that fits with more people supposedly in the system.

  15. this only makes sense if medical school is not only free but medicare pays a stipend. The days of paying to be in the ‘education’ salt-mine of debt only to be an employed at fixed rates is cognitive dissonance.

  16. Pay should be based on a per patient basis. It would be similar to capitation and be based on a much simpler version of the HCC Risk Adjustment model. There should be 5 to 10 “disease groups” and a physician would be paid a yearly fee to take care of each patient. A relatively jealthy patient would pay $750 per year and a p[erson with 3 or more disease groups, including mental health, could range from $1,750 to $3,000 per year. Health plans would allocate a nurse to meet the patient in the PCPs office and the nurse will do most of the legwork. Feel free to contact me regarding this revolutionary solution to the duals demo model. promedpractice@yahoo.com

  17. This is fine for primary care for general adult population, but isn’t suitable for older complex Medicare patients.

    It’s physician level work to figure out an approach to the many intersecting and overlapping chronic illnesses these patients have, and to advise patients on prognosis and trajectory.

    This isn’t to say that Medicare PCPs shouldn’t work with NPs/PAs; it’s great to be able to refer the patient to an NP/PA for something specific, like help transitioning a dementia patient from an overly labor-intensive insulin regimen to something simplified.

    But in general, the Medicare PCPs are all going to have to practice geriatrics and pre-geriatrics. It’s rewarding esp if you are given time to do the job as it should be done, but it’s challenging.

  18. Actually we can both be right. The key is to not conflate the role of PCP-as-partner-in-complicated-health-care-journey-over-time with provider-who-sees-you-promptly-when-you-need-urgent-care.

    Also important to not conflate the PCP work I describe in my post with a primary care team, or a medical home.

    What Jane Brody describes is people going to the ED because they aren’t able to access urgent care.

    Especially because health documentation and health information exchange is absymal, medically complex patients are generally better off when they get urgent care services from a provider who knows them, or at the very least, can access their primary care chart.

    You can embed a PCP who does what I do into a primary care home that provides extended hours for urgent care.

    Better yet, we should be making sure that all patients are able to keep all their information in a personal health record so that they can bring it to any doc in the middle of the night and get decent care.

    Instead we have a system that is still based on the idea that your PCP carries much of the important info about you in their head (it’s true, the notes that many PCPs write would be useless even if urgent care or the ED could access them), so if you can’t access your PCP you’re often at a big disadvantage.

    Meanwhile generalists who would like to have relationships and help older people with the most important aspects of their health are told that 1)you should be willing to see 20 people per day, and hey we might make you go to 24-28, and 2) you should be available at all hours to your patients, and 3) real doctors sacrifice themselves and should be willing to work 60 hours/wk. For the lowest pay in the doctoring profession.

    It’s an approach that’s failing, so I think it’s time to rethink the role of Medicare PCP.

  19. It’s more likely, especially with the recommendations here, that the majority of primary care practice is going to become like psychiatry practice is today. That is, much of this face to face time, counseling, whatever is needed, will be between the patient and NP or PA, and the doctor on-site will be there to sign off on the notes and prescriptions. Just like psychiatrists tend to do these days, working with psychologists.

  20. This little string of comments gives me a radical idea. How about linking physician pay with more variables than years of academic excellence, more years of apprenticeship, and certifications by institutions, professional groups and government licensing boards.
    More, you say? What more do you want?

    Pretend doctors and surgeons are like other highly compensated professionals — engineers, research scientists, accountants, actuaries, sales managers…what makes them different that they are already the highest paid professionals in the country?
    Here is a list of the highest-paid careers in America.


    Notice the words career and <America.
    I have the impression that doctors see themselves more in a “profession” than a “career.” The difference is not subtle. Ask any highly-compensated wage-earner in the country and they will tell you that getting to that level of compensation took years of work and loyalty to some larger entity (company, brand, group, institution, whatever) coupled with a track record that speaks for itself. Toward the top end of that list one group of highly-paid careers jumps off the page.

    Orthodontists — whose specialty is straightening teeth — have an average salary of $90,120 per year. That figure might seem low but it takes into account orthodontists who don’t have a private practice and work instead in general medicine and surgical hospitals.

    When they are not in “private practice” and work in general medicine or hospitals they earn less. Why would that be?

    I suggest that when they affiliate with an institution an important part of overall compensation can’t be easily measured in taxable annual earned income. The trade-off for them is for longer years of job security, retirement packages with which private practice cannot compete, legal and institutional backing if they run into trouble (professional or personal) and the prospect of working less in the future, should that be their choice, without cashing in all their chips. All that, not to mention whatever secondary lifestyle and personal relationship benefits might be forthcoming from letting somebody else worry about the details of running the operation. Those variables are huge. So much for careers versus professions

    I don’t think I need to explain the America part of the quote. Here is a link that makes the point much better than anything I might say.


    I’ve already said more than most doctors want to hear. It’s time for me to duck and cover.
    See ya later.

  21. Agree that the hassle factor is huge; makes the job very stressful and makes it very hard to give complex patients the mental and emotional focus they deserve.

    It would seem logical that clinicians should be paid more $ for those patients who have a lot more medical need, no?

  22. Very productive PCPs have usually failed to provide older patients and their families with the kind of guidance and comprehensive care that I describe in the post.

    Fee-for-15-min-in-person-visit-that-doesn’t-meet-your-most-important-health-needs is probably going out the door. What to replace it with is a good question, esp if you want to encourage PCPs to provide comprehensive person-centered care to medically complex older adults.

  23. Otherwise excellent post, btw. Fix the hassle part, and you’re on to something (oh, and of course there is that small matter of the fact that seeing fewer patients per day would mean more $ per patient is needed to keep primary care competitive )

  24. Money is most certainly the issue. No, not pay. It isn’t simply about paying primary care more. It is about trusting primary care. Ever as much if not more than how everyone likes to say how much we need primary care, they like also to say how poorly docs (usually meaning PC Docs) do at managing this or that, and how they need to adhere to this or that guideline and do all these administrative things to make sure they are providing quality care, because we just don’t trust them otherwise. This is the absolute stark truth. All the restricted formularies, prior authorization requirements, certifications for this and that, etc. all is because we don’t trust the doc with money. A medication that is on the restricted list because of “safety issues” suddenly is the drug of choice as soon as it goes generic. You can follow some guideline to a T but you still have to have someone in your office struggle for 45min with a prior auth for the indicated test. And on and on. Why would I want to do that even for 35 hours a week? Add in HIPAA (with its pending update), Stark, PPACA, Meaningful use, etc etc – it is simply overwhelming. And that is why primary care is dying.

  25. You can currently have a 20 hour work week if you want, you just make less money, so you have to address the money. Beyond that, you need to address how you get paid. FFS has its issues, but we have also found that buying up PCP practices and putting folks on salary kills productivity. You need a comprehensive plan.


  26. hi John,
    Yes, I’ve certainly had to spend time convincing families that it’s ok to let an elderly diabetic eat a doughnut. People need help seeing the big picture.

    Thanks for bringing up Shuster’s post; I loved her idea and am glad to see it brought up.

  27. I agree with Dr Foley. The post suggests a business plan which already exists. All you have to do is choose it as your business plan.

    I met a doctor golfing one day who worked in a “family friendly” group practice where they all only work 4 days a week.

    I am not sure how this has anything to do with medicare and the solution. When you come out of med school you can choose whatever business plan you wish. You may make less money working 35 hours a week, and you may make more money working 60 hours a week.

    Life is full of choices.


  28. Money is THE issue, which results in too many appointments in too long days. I don’t see where it’s addressed at all in this posting.

  29. And the pay for a 35 hour week? Based on the wage and price controls of the last 25 years, I would guess you would be much better off being a part time plumber. And plumbing likely has more “patient centricity” than most any 3rd party payor culture does right now. I started a cash based practice 13 years ago and have never looked back. We are, by definition, “patient centered”. If we are not, then they vote with their wallets. I only do 4 hours of care face to face per day and use the rest of my time to do email (all pro bono), phone consults, case research, care coordination, handball, cycling, family time, and dogs. I hope to be healthy enough to work til I am 75 as I cannot wait to get to the office everyday.

    And I don’t do “practice guidelines”, preauthorizations, or any other TPPer nonsense. Being a professional means answering to the patient first — all other non-deity authorities are secondary.

    Thanx for the article and let me know if you want to move to St Paul. We are hiring.

  30. You make a very good case, Dr. Kernisan. Love that “indecent proposal” in the title. Yours is clearly the voice of someone who has been on the front lines and can see a slow motion train wreck in the making. Anybody who has worked with seniors, including a lot of non-medical types, will read this with heads nodding in agreement — especially all those references to “and their families.” My experience has been that family members can be a lot more problematical than the people with medical problems. My observation is that as people age they come to terms with the downside far better than well-meaning friends and family pushing dietary and/or physical regimens that seem helpful but chip away at the quality of life, not to mention a volume of medicine nearly as big as another meal.

    You and Janice Shuster should get together and explore her excellent suggestion for a National Caregiver Corps so that PCPs of the future can coach helpers to do what doctors find time-consuming by delegating the spade work that is sure to follow.


    (Good luck luring medical students from the high incomes of specialties to the less luxurious incomes of PCPs. From what I’ve read money remains an issue.)