A patient is not a dish off the menu

Atul Gawande, MD, is a technophile and a believer in the checklist, and he yokes these ideologies to an attractive metaphor in his newest essay for the New Yorker. The article is worth reading in its entirety, but it can be easily paraphrased.

The Cheescake Factory, like other successful restaurant chains, has “brought chain production to complicated sit-down meals.” They’ve done it by far-reaching standardization of the best possible processes, but not all the way down, since some freedom is left for the front-line practitioners, for example, the line cooks, to get the job done according to their personal practices. Flexibility is built into the restaurant’s practices, which change regularly according to new data. And customers are satisfied.

Health care in the United States, on the other hand, does none of this currently. Practices are incredibly diverse, often for no good reason. Particularly touching, as in so many of Gawande’s articles, was a story that he elicited from a Cheesecake Factory employee, Dave Luz, a Cheesecake Factory regional manager in the Boston area. Luz’s mother, aged 78, had a fall and was subjected to the depressingly normal dysfunctions, malfunctions, miscommunications, and screw-ups of an American hospital. And, when it was time for her to go home, no one coordinated anything. It was up to Luz to do everything, even get her dressed.

An aide was sent. She was short with him and rough in changing his mother’s clothes. “She was manhandling her,” Luz said. “I felt like, ‘Stop. I’m not one to complain. I respect what you do enormously. But if there were a video camera in here, you’d be on the evening news.’ I sent her out. I had to do everything myself. I’m stuffing my mom’s boob in her bra. It was unbelievable.”

A terrible story. Gawande concludes eventually with a not unexpected conclusion. Standardization should change American medicine. Best practices do not make it down to the medical equivalent of the line cook. Doctors bristle at being told to do things better. “Already, there have been startling changes,” Gawande reports with excitement. “Big Medicine is on the way.”

He tips his hat to the obvious objections. Are we ready to make medicine into Walmart? What about accountability and transparency? “Some will see danger in this. Many will see hope. And that’s probably the way it should be.”

But one problem doesn’t make its way into the article in more than dribs and drabs. Does it work? Does massive standardization do the trick? Of course, we know that checklists work wonders in the critical care setting (from which Gawande presents most of his anecdotal evidence).

But in nearly every other area of medicine, people are still hotly debating whether standardization, or as Gawande puts it, quality control, makes people less sick or helps them live longer. Does giving antibiotics within a specified timeframe actually keep people from dying from pneumonia, or increase the unneeded use of those medications? On a larger scale, do the incentives provided to accountable care organizations actually improve care?

There’s an even deeper problem. What do we do about the majority of medical concerns for which there is no standardization? What happens when clinical judgment, often based on little more than anecdote, meets patient preference? Perfecting the mashed potato tower and slicing the avocado to a quarter of an inch is perhaps quite like fine-tuning the ventilator settings or turning down the oxygen. Critical care involves many quantifiable judgments. But what about all the non-quantifiable judgments?

And what about all the parts of medicine which are not strictly quantifiable? That aide who manhandled Luz’s mother: what standardization would have kept that person from acting like a jerk? Employees are only as good as the organization hiring them, but human beings act only as humanely as their capacity for compassion allows. The technophile’s faith will always let him believe in the perfect cheesecake in the next concession over. But the astute skeptic, the careful consumer of metaphor will realize that sometimes a patient is a patient, not a dish off the menu.

Zackary Berger is a faculty member of the Johns Hopkins University School of Medicine, where he is an internist and researcher in general internal medicine.  He blogs at his self-titled site, Zackary Sholem Berger.

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

  • Michael Appel

    Dr. Berger:

    Thank you for the excellent essay. With all due respect for the excellent work you do, it appears that you are confusing “best practices” (ie: scientifically-proven superior performance elements) with the idea of “standardization.”
    As both a physician and career airline-pilot, please allow me to explain why that confusion completely undermines this analysis, and why Gawande’s assertion is perhaps the most essential argument FOR modeling healthcare after “Cheesecake Factory.”
    By assuming that “best practices” is EQUIVALENT to “standardization,” you are implying that the practice of medicine is optimized merely by having faith that the SYSTEM will deliver superior outcomes as a sum total of the brilliant contributions of each doctor. That presumption, in fact, completely removes all elements from the safety equation, with the exception of individual performance.
    But safety is rarely achieved through improving individual performance. Best practices is an “individual performance” element; standardization is a “system” concept. You are ignoring “the system”; the “low-hanging fruit” for major breakthroughs in healthcare safety.
    When you question Dr. Gawande’s advocacy of “standardization / quality control” by asking whether the timing of antibiotics (a “best practice” performance element) has been proven to “actually keep people from dying from pneumonia,” you are comparing the classic “apples to oranges.”
    This is like questioning whether “rotating at 3 degrees-per-second into the command bars for a V2 climb after a Vee-one cut” (ie: a “best practice” performance element for airline pilots) has been “proven” to reduce plane crashes.
    Although we cannot provide a prospective, double-blind, peer-reviewed published study with a large “n” and small “p” to prove that these performance elements of airline pilots are directly responsible for the “9-sigma” level of reliability in commercial aviation, one thing IS certain: the culture that surrounds respect for system standardization is ultimately THE REASON aviation is so safe. The record speaks for itself.

    /Michael Appel, M.D.

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

      I would think that in aviation there is no need for a “prospective, double-blind, peer-reviewed published study” of anything, because the situation can be a) described by a set of differential equations and solved and b) tested in a wind tunnel or a flight simulator.

      The problems with comparing commercial aviation with medicine are many, and the most glaring differences are that in commercial aviation you have a handful of identical aircraft, a finite number of airports and an exactly predefined flight course between those airports. The nasty problem in medicine is that neither the origins nor the destinations are fixed or well defined, the routes are set and changed on the fly, and patients come in an infinite number of models and have no standardized parts or instruments to indicate their performance at any given time. There are no wind tunnels, no fatigue testing, no reliable flight simulators and no differential equations whatsoever.

      No one is debating the usefulness of certain best practices (e.g. Dr. Pronovost’s checklists, immunizations schedules and much more), but standardizing something that is not completely understood, with no evidence to support this standardization, and enforcing these standards through “Big Medicine” (i.e. corporate interests) is fraught with more dangers to individual patients, than allowing enough degrees of freedom to account for the current state of the art.

      • Michael Appel

        Standardization is not a cure-all prescription for high-reliability. Rather, it is merely an operating framework which enables superior performance from individuals that constitute a system. That is why the “wind tunnel / differential equation” analogy is irrelevant here: Boeing designs airplanes using differential equations and wind tunnels. Yet planes usually crash DESPITE sound design and testing. It’s the PEOPLE that crash them.

        Standardization is the FIRST STEP in quality control. Those of us who embrace the aviation analogy are not arguing that “the standard” is guaranteed to lead to good outcomes. Rather, “standardization itself” — processes which are well-designed, logical, efficient, and consistent — is the means by which PEOPLE are turned into high-reliability components of a system.

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

          The “wind tunnel/differential equation” analogy is what is informing your standards. As a commercial pilot,
          you are operating on an object built to standards. An object that comes
          with a standard manual of operations and an exact definition of how it
          will react to your tiniest inputs within very narrow tolerance margins.
          easy would it be to create standards if on any given flight a pilot
          would have to operate an aircraft built by one of several thousand
          manufacturers, with generally similar looking instruments, but different
          and unknown in advance CL/CD, engine specs and a host of other parameters? You could probably build guidelines and general best practices, but not deterministic “standards”.

          If we are interested in turning people into high-reliability components of systems (and I sort of think that we may actually prefer to do the opposite), then we must “standardize” the actions and reactions of all people in the system, including those people who are acted upon by the system, i.e. patients. If we cannot standardize the parts carried by the production line of medicine, there is absolutely no value in standardizing and mechanizing the actions of line workers, and there is no value in having line workers at all.

          I think education and teachers are a much better example of the value and effects of standardization in a people-on-people system, which commercial aviation is not.

          • Michael Appel

            Then what does the airplane “standard manual of operations” say I should do when I believe my copilot has just made an error in judgement?

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

            Forgive me Dr. Appel, I thought we were discussing standards for treatment processes such as timing of antibiotics, aspirin regimens or when to order what, not standards for basic social behavior such as being attentive and polite to colleagues, and maybe even washing hands now and then.
            If you are referring to Dr. Gawande’s list item of having team members introduce themselves by name before surgery, then by all means, that’s a great thing, but I would not classify this as a “standard”.

  • http://twitter.com/sinusMD Giri Venkatraman

    Apart from the issues you raised, which I agree with, Dr. Gawande’s own parent organization doesn’t seem to have adopted the Cheesecake factory mantra. Partners is a 5 hospital system, with massive duplication of services, no standardization, and their entire goal seems to be corner the market…Big Medicine at Ruth’s Chris prices.

  • http://www.facebook.com/people/Ruth-Cobabe-Brandt/1003202243 Ruth Cobabe Brandt

    Throughout my life I have had the uncomfortable misfortune of being what I refer to as a “chronic patient”. This has placed me in the position of having doctors young and old take one look at my chart and make an immediate and foolish determination that I am a hypochondriac. I have been an inpatient in more hospitals than I care to count, and I have had over 20 major surgeries, none of them elective. In essence, I believe I have enough experience in healthcare to be qualified to comment on this topic.

    Standardizing healthcare would be a tremendous mistake. Patients are not a piece of cheesecake. The best care I have ever received in hospitals was at smaller facilities, where they were able to give more personalized care. The worst care I have ever received was at larger hospitals, hands down. One hospital I stayed at sent me home while I was in complete liver failure. I died at home the next morning. While at the hospital where I had stayed for 6 days, I received no IV fluids. I had a hep-lock for narcotic administration, and was drugged beyond my capacity to reason, so I never drank any fluids. All I could do was sleep. The attending doctor actually accused me of being there for narcotic abuse. This was within hours of my having had optic nerve sheath fenestration. In this enormous, world-famous hospital, I had been transferred from the orthopaedic floor to the cardiac floor, and finally to the neurology unit, where I was supposed to have been to begin with. That was where my attending physician accused me of drug abuse, and I checked myself out of the hospital. It had been such a nightmare, I couldn’t stand it anymore. It was the worst care I had ever received. When I wanted to leave I asked the nurse to pull my hep-lock, but an hour later she told me she was still too busy watching the physical therapist work with the patient in the bed next to me. I pulled the hep-lock myself, and got dressed, while my husband searched several floors of the hospital for a wheelchair so we could leave.

    In another example of this so-called standardized care, my dad had a heart attack and was taken to a large hospital. He had angioplasty and a stent was placed. He remained in atrial fibrillation, and after a few days was sent home, still in a-fib. In the middle of the night, his first night home, he suffered a massive stroke. He died in that same hospital a few days later. I have no date that if he had been in a smaller, non-”standardized”-care hospital, they never would have sent him home in a state of atrial fibrillation. He was clearly not safe to go home. My father died unnecessarily because he was caught in the machine of standardized medicine, just as I had been several years earlier.

    The best doctor I ever had died in an airplane crash around 10 years ago. He was a young doctor, and a good man who was on his way to a conference on treating patients with humanity. He was probably just an average doctor, as doctors go, but he was amazingly humane and took his time treating the patient as a human being, rather than as a cog in the machine. I’ve always been struck by the irony of this young, highly compassionate doctor, dying on his way to a conference on the subject of treating patients with humanity. He was the last doctor in the world who needed it, but he was the first in line to sign up for it. The world is a lesser place for his absence. All doctors could have learned something from him.

    We need less standardized medicine, and more humanized medicine. But that’s just the opinion of this “chronic patient”.

    Ruth Brandt

Most Popular