Do You Believe Doctors Are Systems, My Friends?

In the current issue of The New Yorker, surgeon Atul Gawande provocatively suggests that medicine needs to become more like The Cheesecake Factory – more standardized, better quality control, with a touch of room for slight customization and innovation.

The basic premise, of course, isn’t new, and seems closely aligned with what I’ve heard articulated from a range of policy experts (such as Arnold Milstein) and management experts (such as Clayton Christensen, specifically in his book The Innovator’s Prescription).

The core of the argument is this: the traditional idea that your doctor is an expert who knows what’s best for you is likely wrong, and is both dangerous and costly.  Instead, for most conditions, there are a clear set of guidelines, perhaps even algorithms, that should guide care, and by not following these pathways, patients are subjected to what amounts to arbitrary, whimsical care that in many cases is unnecessary and sometimes even harmful – and often with the best of intentions.

According to this view, the goal of medicine should be to standardize where possible, to the point where something like 90% of all care can be managed by algorithms – ideally, according to many, not requiring a physician’s involvement at all (most care would be administered by lower-cost providers).  A small number of physicians still would be required for the difficult cases – and to develop new algorithms.

A variant of this view, discussed by technologists such as Vinod Khosla, and commentators such as John Goodman, imagines that one day even the low-cost providers can be cut out of the loop, and patients (consumers) can do most of the work with their computer and perhaps a few gadgets.

Doctors, as you might expect, tend to reject this vision of standardization, as it ruthlessly undercuts the view that physicians are particularly wise, special, insightful – and worthy of autonomy – and instead seems to assert that medicine should be run like an assembly line, with limited opportunity for customization.

The patient perspective may be more complicated: on the one hand, it’s absolutely true that the current system generally fails even the most basic standards of customer service.  There’s virtually nothing in the current system that appears to be designed around patients; Gawande shares an example about the care a patient received in the emergency room that could have been told, with few modifications, by virtually everyone I know, and holds true whether you’re talking about a small community hospital or one of the nation’s leading teaching centers.

In other words, learning a bit about customer service from The Cheesecake Factory – or a number of other industry-leaders – would do medicine a world of good.

On the other hand, I worry that a lot of medicine really isn’t quite as reducible, as standardizable, as many of the advocates and management gurus would like to believe, and by doing the classic economics trick of “assuming a can opener” – assuming medicine is standardizable because, gosh, wouldn’t it be nice if it were – these experts may not be helping as much as they’d like to imagine.

Patients deserve and increasingly demand far better “customer service” than they currently receive; at the same time, at least the patients who are fortunate enough to have physicians tend to give them surprisingly high marks.  Critics contend this is how the system is harmful to patients – patients can be suckered by good bedside manner, and not realize how poor the care actually is.

Yet, I’d argue that given the incredibly limited amount of solid evidence for most things in medicine, the individual relationship between physician and patient can be of remarkable therapeutic value (although perhaps less so in surgery than in internal medicine).

Thus, while Gawande claims that “Patients just won’t look for the best specialist anymore, they’ll look for the best system,” I wonder whether this is generally true and realistic.  While Romney was roundly criticized for suggesting that “corporations are people,” I wonder if Gawande is overreaching in making the reverse claim: essentially, that “doctors are systems,” and that patients should, and will, reach for the best system, not the best person.

I certainly appreciate where Gawande’s coming from here, but I also worry that we might lose something important by accepting this premise – something vital and distinctive about the patient-doctor relationship that is unlikely to be captured with the same depth and nuance if it’s instead between a patient and a system.

As I’ve previously discussed, it’s a challenge to balance consistency and innovation, and it’s difficult to know how best to remove the “bad” variability in care delivery while still supporting the customization of care so central to medicine, and to healing.  We might want our $15 dinners to look the same, but I doubt most of us want our medical care delivered in as rote a fashion.

The question is whether there’s a way to improve care and preserve individualization, and avoid imposing what is effectively centralized control and a litany of standardized processes.

Doctors may be running out of time to figure this out.  The writing is on wall (and, it seems, everywhere else as well).  If doctors don’t want to wind up as commoditized participants in a Taylorized vision of medicine they will need to recognize their limitations and seriously up their game.

I’ve too much respect for the practice of medicine and too much concern for the care of patients to believe that the Cheesecake Factory really defines how we want American medicine to be served.

David Shaywitz is co-founder of the Center for Assessment Technology and Continuous Health (CATCH) in Boston.  He is a strategist at a biopharmaceutical company in South San Francisco. You can follow him at his personal website. This post originally appeared on Forbes.

24 replies »

  1. I do not need a personal relationship with my airline pilot or air traffic controller when I fly, because I have a relationship built on trust and a consistently high quality product.

    I propose that, to a significant degree, the current emphasis on the need for a *personal* relationship to a health care provider represents a compensatory mechanism. Since we (both patients and clinicians) know how unreliable the medical product is, instead we demand faith in the fact that our clinician will try hard and be nice to us.

    If AI and medicine ever reach the point where medical diagnosis and treatment are as reliable as airline safety or as functional as online banking or making travel arrangements, I predict that the demand for a personal relationship with a caregiver will be replaced by demands for bandwidth and a user-friendly interface.

    Peter Elias

  2. 3 challenges with your thoughts:

    1. You ask if the Cheesecake Factory system is so good why are there other restaurants. You are mistakenly comparing one chain with an entire industry. That would be like saying if Johns Hopkins is so good, why are their other healthcare facilities anywhere near them? No one is making the argument that we should all move to one system that is housed under one parent company with no competing systems.

    2. You ask if anyone at the Cheesecake Factory knows your name or cares about you. You’re making a broad generalization that big/systematized = impersonal. Actually, the Cheesecake Factory near me in fact is very friendly and personal. I’ve had several servers over time that do know me and the experience is quite personal. Again, having a more systematic approach to healthcare doesn’t automatically equal being treated like a robot. It is still the same caring people you’re interacting with, just with a more proven process for making sure you not only feel cared for but actually get better medicine.

    3. Finally, you talk about systems breaking down when problems become complex or nuanced. The faulty assumption here is that an evidence-based system of best practices by default eliminates the ability for doctors to use their wisdom and experience. I don’t think anyone is arguing that we should have a checklist that cannot be deviated from for any reason. The argument is that great doctors + proven systems = better, more efficient care.

  3. Alright, this a repost from inside Linked In, Uncle Sam needs your input and I am really impressed wiht the “passion”!

    To all of you who commented on this thread, “thank you very much”! Very little do I know about the medical industry but now, thanks to your input, I am a little better off! The first twenty years or so running our humble company were spent being “schooled” and here as I develop an advocacy for the “prevention” of Hospital Associated Infections (HAI) in our Veteran’s Hospital’s, “school” begins, I should be getting pretty good at it by now!
    The last year was spent learning the mechanics of the medical system and how to be a supportive advocate. HAI is an odd thing, my efforts will do little until I understand the culture of the community even though my focus is on leadership, above the “silos”!
    “Prevention” is something new as of today for me, it seems to be circulating a the U.S. Department of Health and Human Services, cool, let me know if any of you healthcare professionals have a problem with that! “Eradication”, to me, means to remove something, when you are talking about Ventilator Acquired Pneumonia(VAP) there is one that many are having success “preventing”!
    There is quite a verity of comments on here, now that you are all warmed up, send in your thoughts:

    Re: Public comment of Phase 3 in the 5-year National Action Plan to Prevent Healthcare-Associated Infections: Roadmap to Elimination (HAI Action Plan).

    “If we can improve the quality of care, that will translate into lower cost,” Anthem President Pam Kehaly said. “These are real dollars”


    Something of a ”mergence of cultures”, due to my lack of resources and knowledge to “support those who help others” like you and a few Jarheads in our V.A. Hospitals, so from boot camp, “knowledge is king” and from our humble machine shop, “ I still do not know what I do not know yet”!
    Looking over Ms. Pam Kehaly’s bio on the web last year I found her advice to her staff, “have fun”, it helps a lot in my position, I can use all the help I can get!
    So once again, thank all of you for what you do for the people of our community and your patience as I develop my advocacy!

    Michael H. Slavinski

    Ps: if this did not totally kill this thread I have much to learn about your culture!

  4. Lovely, well-considered comment. Thanks! Changes such as the ones you see in the last part of your comment will be a godsend to cognitive practitioners. It seems to me, as it does to “Denise Vincent” below, that a great deal of what ‘cognitivists’ do now could be managed as well or better by para-professionals, either alone or in groups. The next step in your analysis would seem to be how to leverage those changes to make the life of the cognitivist not only more accurate [using your new tools] but more satisfying and productive as well. See Christensen’s “The Innovator’s Prescription” for a longer discussion of this issue. Tremendous potential for cost savings, but how you gonna’ make it happen?

  5. Thanks for your comment. “Can’t” and “never” are too definitive. Of course there are docs, office managers, and staff who CAN do CQI. But, realistically, what percentage of small, private practices will this ever represent? Not enough to make any significant difference. If, however, the practices get help with it, in its many forms, from an IPA or ACO, then it might happen. Having just spoken to the medical director of our local IPA, I can tell you that it’s the proverbial herding of cats and doesn’t go very well.

  6. 90% of care could be managed by algorithms because of the huge load of BS that caregivers are forced to wade through due to waste and ineffeciencies, not because anyone is questioning physician expertise. Dr. Shaywitz completely misses that the best reason to standardize care is to more fully leverage physician brain power. I suspect that Dr. Shaywitz believes that physicians are the line workers in the Cheesecake Factory Scenario, not realizing that in healthcare, the support staffs are the line workers, while physicians are the managers.

  7. RobertL39, I respectfully disagree with your assertation that small medical practices can’t pursue Continuous Improvement. Please check out Dr. Sami Bahri (there’s a blog about his incredible work at theleancorner.com), practicing dentist and author of Follow The Learner. He and his team have absolutely incorporated CQI concepts (primarily Lean) and his patients adore what they’ve done.

  8. It’s between the waiter and patron at the restaurant table of medicine
    Hospitals can be dangerous and inefficient; therefore it is easy to connect with Atul Gawande’s recent New Yorker essay “BigMed” suggesting that the streamlined, production processes found at the Cheesecake Factory can and likely will be applied to healthcare. Yet hospital care should not be confused with the full spectrum of healthcare. One must make the distinction between the cognitive process of medical diagnosis occurring in exam rooms, with the procedural basis of surgical care and hospital recovery. While Dr. Gawande has provided a wonderful revealing portrait of cost-effective, fast, food preparation and delivery at the CheeseCake Factory , he has focused on the process of creating the meal, not the process of deciding what meal to make. Successful surgery, for the wrong diagnosis, is a problem. If we are to solve some of healthcare’s largest failings we should focus on what happens as physicians try to address their patient’s problems, diagnose and make decisions, at the table of medicine called the exam room.
    Consider the continuum of the patient encounter, from first symptoms, through diagnosis and therapy at a restaurant called Med. At Med I spend all of my shifts with my patrons at my tables. This is an unusual restaurant since the patrons are never sure of what they want to eat and appear every 20 minutes with ever changing lists of unique groups of ingredients to share with me. There are varying ingredients and thousands of meals that can be created. The patrons know the ingredients, but not the meal that they would like to eat. From memory I respond to the customers list of ingredients and ask many questions, take the pulse and other vital signs of the customer, order blood samples, radiographic studies and then decide for the patron which meal their ingredients add up to. All from memory. At Med, restaurant patrons also ask for foods and “food tests” they have seen on television all purported to be risk free. Further complicating the process is my customer is not out for a fun and relaxing evening, they are in small booths in skimpy, open at the back gowns, often anxious and uncertain if they will be harmed or poisoned by my foods, or simply receive a meal they do not want. Some are in pain and some are depressed, while other customers are totally unrealistic about the meal that is to be delivered. You see at Restaurant Med, where patrons only can speak to their wait staff about ingredients, and demand the modern but unhelpful ovens they heard about from friends and the media, it is really difficult to create meals that patrons thoroughly enjoy.
    An appendectomy should be consistently performed and priced, but how do we consistently perform and price considering the ambiguity inherent in diagnosis itself? Unlike a restaurant, where customers choose a meal by ordering a meal, at restaurant Med some higher force gives an unfortunate person an undifferentiated and undiagnosed problem that needs and deserves an answer. As it turns out, none of the patrons really want to be eating at restaurant Med, as they always receive a meal they did not ask for.
    Patients do not choose their diagnoses from menus; doctors must discover and diagnose them.
    If your waiter tries to memorize all the orders at all the tables, you might get the wrong meal, and if your server is in a hurry, thai dipping sauce might be spilled on your new silk blouse. Likewise if physicians are in a rush, they might not take a thorough history, perform a complete physical exam, or have an accurate and thorough list of diagnostic possibilities, ultimately resulting in the wrong diagnosis. If your physician believes he or she can memorize all the questions, tied to all the possible diagnoses you also might receive the wrong diagnosis. With that wrong diagnosis you might end up in a hospital more efficient than the Cheesecake Factory with doctors efficiently ordering unnecessary tests, and performing wrong surgeries for the wrong diagnosis all with the ease and speed of the best assembly line on the planet.
    Diagnostic and patient management error caused by cognitive mistakes in the exam room are all too often overlooked and unmentioned in the discussion of repairing our broken healthcare system. There are over a billion outpatient visits in the US each year, and numerous studies have shown 15-20% of these visits have an inaccurate diagnosis. Autopsy data proves this, malpractice insurers know this, and policy makers avoid it. Add diagnostic error in the emergency room and walk-in clinics to error in the out-patient offices of medicine and you have more than 200 million errors. If we are to resolve some of healthcare’s deepest woes we need to address diagnostic errors and the decision-making occurring at the restaurant table of medicine, the exam room. A bright light needs to be shined on the simple fact that there is too much to know, to ask and to apply during a 15 minute encounter unless the patient has the simplest of medical questions or problems. Medical informaticists, researchers and innovative companies are focusing on this essential limitation of medical decision-making by designing information systems to be used by physicians at the point of care, during the patient encounter. Problem oriented systems can also be designed for use by patients in advance of the visit, and the future holds home-based information coordinated with professional clinical decision support. These new information tools are beginning to take the guessing out of which ingredients (symptoms) relate to the meals that the patient ultimately receives (diagnosis and treatment). If medical care is truly to be driven back to primary care we need to arm the waiters of medicine with purposefully designed tools and training to resolve ambiguity, aid diagnosis and inform therapy in the exam room.

    Art Papier MD

  9. Which leads back to my main point: you/we will get ACOs whether we help them or not. The choice is to help them, being sure they “know something about clinical medicine”, or don’t help and cross your fingers. Are doctors going to lead this, or continue their rear-guard actions of the past?

  10. RobertL: You might assume that but I’m far from convinced the people doing it will assume that.
    Therein lies the whole debate.

  11. Read your first regulation:

    “…become accountable for the quality, cost, and overall care of the Medicare fee-for-service beneficiaries assigned to it”
    If you don’t know any clinical medicine, I’d suggest you might not want to assume accountability for the quality and overall care of the beneficiaries assigned to your ACO.

  12. Robert L says:
    Sure sounds like an ACO to me. Is that BIg Medicine?

    Here are the DHHS regs:
    Incorporating DHHS final regulation adjustments on October 20, 2011, Section 3022 outlines the following requirements for ACOs:
    The ACO shall be willing to become accountable for the quality, cost, and overall care of the Medicare fee-for-service beneficiaries assigned to it
    The ACO shall enter into an agreement with the Secretary to participate in the program for not less than a 3-year period
    The ACO shall have a formal legal structure that would allow the organization to receive and distribute payments for shared savings to participating providers of services and suppliers
    The ACO shall include primary care ACO professionals that are sufficient for the number of Medicare fee-for-service beneficiaries assigned to the ACO under subsection
    At a minimum, the ACO shall have at least 5,000 such beneficiaries assigned to it in order to be eligible to participate in the ACO program
    The ACO shall provide the Secretary with such information regarding ACO professionals participating in the ACO as the Secretary determines necessary to support the assignment of Medicare fee-for-service beneficiaries to an ACO, the implementation of quality and other reporting requirements under paragraph (3), and the determination of payments for shared savings under subsection (d)(2)
    The ACO shall have in place a leadership and management structure that includes clinical and administrative systems
    The ACO shall define processes to promote evidence-based medicine and patient engagement, report on quality and cost measures, and coordinate care, such as through the use of telehealth, remote patient monitoring, and other such enabling technologies
    The ACO shall demonstrate to the Secretary that it meets patient-centeredness criteria specified by the Secretary, such as the use of patient and caregiver assessments or the use of individualized care plans
    The ACO participant cannot participate in other Medicare shared savings programs [20]
    The ACO entity is responsible for distributing savings to participating entities
    The ACO must have a process for evaluating the health needs of the population it serves

    Maybe I’m missing something but where is the requirement in here anyone needs to know anything about actual clinical medicine?

  13. And to Margalit Gur-Arie (thank you for your comment!):
    Would one of you please translate what you mean by “Large Systems Medicine” or “Big Medicine”? There are many different definitions, some of which we should probably avoid and some of which perhaps embraced [e.g., would you call Mayo Clinic “Big Medicine”?] And where is it written that we must “…accept Big Medicine.”? Do you imply that this means everything else is excluded?
    from Cory above “…people with medical experience who are willing to take a disinterested look at these things from a patient care standpoint, taking in to account all facets of medical care including cost.” Sure sounds like an ACO to me. Is that BIg Medicine?
    I think you both need to read Clay Christensen’s book. And realize that by just complaining and not offering an alternative [see Christensen again] you’re likely to get Big Medicine, like it or not.

  14. I welcome your “impertinence” because it is not impertinence at all. You are right. This is a false dichotomy. Good medical care looks to me like it has very little to do with size, so why is the conclusion that we must accept Big Medicine? There is no evidence to support that conclusion, so perhaps one should not be so quick to offer it as a solution.

    And I agree with Cory below.

  15. Pardon my impertinence, but it feels to me like you are creating a false dichotomy between Big Medicine [not defined except as ‘Partners’] and “one-of-a-kinds”, by which I presume you mean independent private practices. Small offices simply can’t/won’t do extensive process improvement. That’s not their job, unless they’d rather do that than practice medicine. Each form has something to offer, just as Per Se has something to offer that Cheesecake Factory doesn’t. And vice versa. “One size don’t fit all.” The dichotomy is false.
    Similarly impertinently, I’d like to suggest that you re-read Dr. Gawande. How did small private practice, or Big Medicine, rein in costs in McAllen? Is his ‘generalized conclusion’ that there a huge amount of wasted money out there that can be brought back into the system to provide needed care correct? You bet. This extrapolates to Big Medicine how? Agriculture has morphed for a variety of reasons, virtually none having to do with Agricultural Extension agents. His ‘generalized conclusion’ that best practices may vary by area and be extrapolatable elsewhere is wrong, how? Spreading Best Practices is not inherently Big Medicine and to conflate them is inaccurate and misleading. You may want your lap Nissen [“Pit Crew” references–read ‘tire changed’] done by the cowboy who does a little bit of everything including neurosurgery [read ‘fuel added’], but I don’t, thanks. As he has agreed, pit crews need a boss [read, usually, ‘Doctor’] to provide The Big Picture and help things run well. And this is wrong, how?
    Don’t be so quick to defend your turf and strike at Big Medicine, though a lovely target it is. Read, in particular, Christensen and Arnold Relman. Will your ‘solution shop’ be part of Big Medicine? Guess that depends on how good a shop you run!
    Lastly, I’d argue that the real reason to fear Big Medicine is because of your reaction; physicians have decided that hunkering down, taking shots at Big Medicine, and resisting change is their only possible salvation. So the insurance companies, large companies, and maybe even the Feds will force Big Medicine on us because doctors have not not only not led the way forward but persist in trying to make sure that we never move any direction at all. We’ll have only ourselves to blame.

  16. “This post really gets my goat. “, said someone.

    Patients are not cheese cake or meat or goat cheese or lettuce wraps. And, they, well you know then but, do not want to be eaten. So then, the infamous writer Atul has it wrong. He and most of the bloggers and commenters here have not read on the Cerner outage that eliminated the medical records of thousands of the sickest patients in one fell swoosh.

    If there is a need for process control and improvement, imust be devoted to
    the HIT vendors who have zero process control cause they answer to know one.

    They use the patients as meat or guinea pigs for their unregulated experiments.

  17. “Dr. Shaywitz and Cory have a romanticized notion of medicine, but the truth is that with the right incentives, a welcoming, patient-centered delivery orientation and clinical decision support tools, the patient experience can vastly improve, the quality of care become predictably far better and the cost drop precipitously.”

    On your first point, that’s true. Some romanticized medicine is good some is bad. Let’s not forget medicine has been going on far longer than CQI and there were some pretty smart people doing 50, 100, 500 years ago who established certain principles we are willing to abandon without much thought.

    On your second point, I’d be willing to concede it, but like anything else in medicine I’d want to see it actually does what they say it does. And long experience has taught me that quick fixes in medicine are often not what they seem, which brings us to Ms. Gur-Arie’s prescient point in her last paragraph.

    The argument for large systems medicine, like the argument for EMRs is rigged. Anyone who does not agree is a Luddite and since the culture is gravitating that way anyway, you cannot win the argument. But it does not mean the argument is right. There is certainly some truth in it but we need people with medical experience who are willing to take a disinterested look at these things from a patient care standpoint, taking in to account all facets of medical care including cost. It sure looks like we are building more anymore barriers between us and patients rather than breaking them down. But maybe that’s romanticized.

  18. Bobby, I am not arguing with Toussaint’s writing. I am not even arguing with Dr. Gawande’s suggestions for CQI. I am just arguing with his unfounded assertion that something must be Big in order to be good.

    All the conveniences that Big things brought us, came at a steep price, sometimes financial other times worse than that. I’m not sure we can afford the collateral damage associated with Big Med.

  19. Gotta say, you never disappoint.

    Though, I would take issue with this:

    “If so are there any other chains that are better than individual establishments? Why are all one-of-kind restaurants lumped together? Surely, one can identify certain factors that make one-of-kinds infinitely better than others and than chains? Without clear answers to these questions, why must we accept Big Medicine?”

    You’d have to actually read Toussaint to understand that that is not what’s HE’s calling for. Quite the converse, beyond some general principles that are by now time-tested in clinical settings..

  20. I think there is a subtle point getting lost in this exchange.
    Nobody ( I presume) is debating the advantages of performance improvement, evidence-based decision support, data and information technology, team work, coaching and training and all the good stuff that could be applied, to a certain extent, to medicine today and to a much larger extent to medicine tomorrow…. but what does this have to do with Big Medicine?

    The problem, IMHO, with this cheese thing is the same problem with many other Gawande articles: he picks one facet of a multifaceted situation, analyzes it and then proceeds to derive generalized conclusions that have very little basis.

    Is the supposed efficiency of CF derived from its being a chain? If so are there any other chains that are better than individual establishments? Why are all one-of-kind restaurants lumped together? Surely, one can identify certain factors that make one-of-kinds infinitely better than others and than chains? Without clear answers to these questions, why must we accept Big Medicine? Because it works so well at Partners?

    He did the same thing with McAllen/El Paso, the same thing with pit crews/cowboys, the same thing with agriculture, and yes, with aviation….
    More here: http://onhealthtech.blogspot.com/2012/08/dr-gawandes-new-shiny-thing.html

  21. “There is plenty of good – vast enhancements in cost, quality and safety – than can come from active process improvement.”

    Dr. Toussaint and others with established cred peg this at at least 30% in savings (process improvement side). Moreover, this is no longer QI “theory.” See the long Thedacare experience in particular.

    With respect to “loss of physician autonomy,” Lawrence and Lincoln Weed have demolished that lament (“Medicine in Denial”).

    But, yeah, let’s just add the “restaurant” analogy to the hated “aviation” analogy (and, by extension, anything touting “high-performance teams”).

  22. As Dr. Gawande describes so engagingly, most modern doctor’s offices and health systems are brimming over with dysfunction, and this is very often reflected not only in the patient experience but the quality of care. Dr. Shaywitz and Cory have a romanticized notion of medicine, but the truth is that with the right incentives, a welcoming, patient-centered delivery orientation and clinical decision support tools, the patient experience can vastly improve, the quality of care become predictably far better and the cost drop precipitously. We are doing these things in my firm’s health centers, which is why we recruit in more than 60% of employee populations in the first year. The clinic becomes stickier after that.

    The criticism here seems to be that standardization is a dehumanizing process. In fact, doing things to patients by caprice and unsupported by evidence is the assault on them.

    There is plenty of good – vast enhancements in cost, quality and safety – than can come from active process improvement. Creating a better patient care environment makes managing patient complexity less cumbersome, not more, with a better likelihood of positive results.

    Doctors who don’t get this are way behind. They should wake up, and maybe visit organizations, as Dr. Gawande has, where health care is better, with the results to prove it.

  23. No one would argue that our medical system is perfect, and all too often it is less than good. Streamlining and improving some processes would be a good thing, but the idea that medical care should be just like Cheesecake Factory is pretty ridiculous when you think about it.
    Consider- if Cheesecake Factory was so good, why doesn’t everyone go there? Why do we have other restaurants? Why is it likely that someday, Cheesecake Factory, like most restaurants and franchises, will go under? It’s not a commentary on Cheesecake Factory but on the restaurant analogy.
    MacDonalds would have been a similar and more understandable analogy. Yes, medical care could be standardized like MacDonald’s and in some respects this would be good. It delivers reliable standard meals. But is it the place to go for your anniversary? Does everyone like it? Do places do certain things better, whatever better, for a restaurant, is?
    There are a number of factors to consider. First, the difference between customer service and good medicine. good doctors have known for centuries that there are many bad doctors out there who get by with good bedside manner. IT’s not entirely a bad thing. And in fact one of the biggest problems in medicine today is the growing depersonalization of medicine. Those personable hacks have a role, albeit a limited one, in the equation. Think about it- would standardizing like Cheesecake Factory make that depersonalization better? Does anyone at Cheesecake Factory know your name when you walk in or care about you in anything other than a commercial sense?
    Second the difference between routine medicine and outliers. Routine medicine can generally be taken care of by algorithms. And it will problem outperform most physicians, if we make the assumption in advance a certain problem is routine (shouldn’t we be testing the algorithms with controlled studies before implementing them? Just asking). the algorithm can do that up to a point and that is where the Cheesecake Factory model holds its greatest promise. I have read much of that standardization literature for the ICU. IT works for routine stuff and fails miserably for complicated problems (just like the Spellcheck system that misspelled tow words in that paragraph -I misspelled a couple also so neither one of us by ourselves was perfect ).
    So the final issue is the complex or tough problem. In this case the patient who looks for the best system rather than the best doctor is a fool. Give me the surgeon who knows the nuances of the subtle case, or the ID doc who has seen the unusual infection, or the intensivist who understands how sick the patient is by looking beyond the numbers. Especially if I am the one who is sick.
    In sum, making medicine more like Cheesecake Factory has its benefits but to think its a fully formed solution or the way to go, at the expense of traditional medical wisdom and experience, is a sign of people who don’t really understand how medicine works, even if they are Dr. Gawande.

  24. This post really gets my goat. It completely misunderstands, perhaps deliberately, the role of the physician in a well-designed system of medical care. It seems to me that if Dr. Shaywitz has actually practiced in a hospital, the dysfunction of what we now call a system would be blindingly obvious. I have experienced it as both doctor and patient.
    The idea of standardizing care is far from what he describes. Yes, it seeks to eliminate the variations in physicians’ care which are entirely unexplainable, such as geographic variations for which there is no other explanation.
    But the truly terrible variation is in the PROCESSES of care, leading to the rampant medical errors which kill people. Find me a doctor who thinks we have reliable processes and it will be a doctor who has never practiced.

    I find this post entirely unhelpful and even dangerous to the effort to educate physicians and patients as to what a ‘system’ in medical care really is. Stop pontificating and start reading people like Brent James, Gary Kaplan, Paul Levy and Bob Wachter.