by Atul Gawande
It is still not apparent to me what the clues were that I was registering when I first saw Eleanor’s leg. Likewise, it is not obvious what the signs were that we could get by without an amputation. Yet as arbitrary as our intuitions seem, there must have been some underlying sense to them. What there is no sense to is how anyone could have known that, how anyone can reliably tell when a doctor’s intuitions are heading down the right track or spinning wildly off.
For close to thirty years, Dartmouth physician Jack Wennberg has studied decision making in medicine, not up close, the way Gary Klein has, but from about as high up as you can get, looking at American doctors as a whole. And what he has found is a stubborn, overwhelming, and embarrassing degree of inconsistency in what we do. His research has shown, for example, that the likelihood of a doctor sending you for a gallbladder-removal operation varies 270 percent depending on what city you live in; for a hip replacement, 450 percent; for care in an intensive care unit during the last six months of your life, 880 percent. A patient in Santa Barbara, California, is five times more likely to be recommended back surgery for a back pain than one in Bronx, New York. This is, in the main, uncertainty at work, with the varying experience, habits, and intuitions of individual doctors leading to massively different care for people.
How can this be justified? The people who pay for the care certainly do not see how. (That is why insurers bug doctors so constantly to explain our decisions.) Nor might the people who receive it. Eleanor Bratton, without question, would have been treated completely differently depending on where she went, who she saw, or even just when she saw me (before or after that previous necrotizing fasciitis case I’d seen; at 2 A.M. or 2 P.M.; on a quiet or a busy shift). She’d have gotten merely antibiotics at one place, an amputation at another, a debridement at a third. This result seems unconscionable.
People have proposed two strategies for change. One is to shrink the amount of uncertainty in medicine—with research, not on new drugs or operations (which already attracts massive amounts of funding) but on the small but critical everyday decisions that patients and doctors make (which gets shockingly little funding). Everyone understands, though, that a great deal of uncertainty about what to do for people will always remain. (Human disease and lives are too complicated for reality to be otherwise.) So it has also been argued, not unreasonably, that doctors must agree in advance on what should be done in the uncertain situations that arise—spell out our actions ahead of time to take the guesswork out and get some advantage of group decision.
This last goes almost nowhere, though. For it runs counter to everything we doctors believe about ourselves as individuals, about our personal ability to reason out with patients what the best course of action for them is. In all the confusion of different approaches that different doctors take to a given problem, somebody must get it right. And each of us—used to making decisions under uncertainty every day—remains convinced that that somebody is me. For however many times our judgment may fail us, we each have our Eleanor Bratton, our great improbable save.
It was a year before I saw Eleanor again. Passing through Hartford, I called in on her at her family’s home, a roomy, spic-and-span, putty-colored colonial with a galumphy dog and beds of flowers outside. Eleanor had moved back home to recover following her twelve days in the hospital, intending to stay only temporarily but instead finding herself nestling in. Returning to a normal life, she said, was taking some getting used to.
Her leg had taken time to heal, not surprisingly. In her final operation, done during her last days in the hospital, we had needed to use a sixty-four-square-inch skin graft, taken from her thigh, to close the wound. “My little burn,” she called the result, rolling up the leg of her sweatpants to show me.
It wasn’t anything you’d call pretty, but the wound looked remarkably good to my eye. In final form, it was about as broad as my hand and ran from beneath her knee to her toes. Inevitably, the skin color was slightly off, and the wound edges were heaped up. The graft also made her foot and ankle seem wide and bulky. But the wound had no open areas, as there sometimes can be. And the grafted skin was soft and pliant, not at all tight or hard or contracted. Her thigh where the graft had been taken was a bright, cherry red, but still fading gradually.
Recovering the full use of her leg had been a struggle for her. At first, coming home, she found she could not stand, her muscles were so weak and sore. Her leg would collapse right under her. Then, when she’d built the strength back, she found she still could not walk. Nerve damage had given her a severe foot drop. She saw Dr. Studdert and he cautioned her that this was something she might always have. With several months of intense physical therapy, however, she trained herself to walk heel-toe again. By the time of my visit, she was actually jogging. She’d also started back working, taking a job as an assistant at one of the big insurance company headquarters in Hartford.
A year on, Eleanor remained haunted by what happened to her. She still had no idea where the bacteria came from. Perhaps the foot soak and pedicure she had gotten at a small hair-and-nail shop the day before that wedding. Perhaps the grass, outside the wedding reception hall, that she’d danced barefoot through with a conga line. Perhaps somewhere in her own house. Any time she got a cut or a fever, she was stricken with mortal fear. She would not go swimming. She would not immerse herself in a bath. She would not even let the water in the shower cover her feet. Her family was planning a vacation to Florida soon, but the idea of traveling so far from her doctors frightened her.
The odds—the seeming randomness—were what disturbed her most. “First, they say the odds of you getting this are nothing—one in two hundred fifty thousand,” she said. “But then I got it. Then they say the odds of my beating it are very low. And I beat those odds.” Now, when she asked us doctors if she could get the flesh-eating bacteria again, we told her, once more, the odds are improbably low, one in two hundred fifty thousand, just like before.
“I have trouble when I hear something like that. That means nothing to me,” she said. She was sitting on her living room sofa as we talked, her hands folded in her lap, the sun rippling through a bay window behind her. “I don’t trust that I won’t get it again. I don’t trust that I won’t get anything else that’s strange or we’ve never heard of, or that anyone we know isn’t going to get such a thing.”
The possibilities and probabilities are all we have to work with in medicine, though. What we are drawn to in this imperfect science, what we in fact covet in our way, is the alterable moment—the fragile but crystalline opportunity for one’s know-how, ability, or just gut instinct to change the course of another’s life for the better. In the actual situations that present themselves, however—a despondent woman arrives to see you about a newly diagnosed cancer, a victim bleeding from a terrible injury is brought pale and short of breath from the scene, a fellow physician asks for your opinion about a twenty-three-year-old with a red leg—we can never be sure whether we have such a moment or not. Even less clear is whether the actions we choose will prove either wise or helpful. That our efforts succeed at all is still sometimes a shock to me. But they do. Not always, but often enough.
My conversation with Eleanor wandered for a while. We talked about the friends she’d gotten to see now that she was back in Hartford and her boyfriend, who was something called a “fiber-optic electrician” (though what he actually wanted to do, she said, was “high voltage”), about a movie she had recently gone to, and about how much less squeamish she’s discovered herself to be after going through her whole ordeal.
“I feel a lot stronger in some ways,” she said. “I feel like there is some kind of purpose, like there has to be some sort of reason that I’m still here.
“I think I am also happier as a person”—able to see things in perspective a bit more. “Sometimes,” she went on, “I even feel safer. I came through all right, after all.”
That May she did go to Florida. It was windless and hot, and one day, off the eas
tern coast above Pompano, she put one bare foot in the water and then the other. Finally, against all her fears, Eleanor jumped in and went swimming in the ocean.
The water was beautiful, she says.
Notes on Sources
INTRODUCTION
5 Specialized medical journals are where doctors find much of their information on practical problems. Thus, the specific dangers of the large chest mass in children are detailed in articles such as Azizkhan, R. G., et al., “Life-threatening airway obstruction as a complication to the management of mediastinal masses in children,” Journal of Pediatric Surgery 20 (1985), pp. 816–22. For the most part, the lessons in articles like these are learned the hard way—from experience. Disaster occurs, and we call that a tragedy. But if someone writes it down, we call it science.
At least two articles explain strategies doctors have found using heart-lung pumps to safely manage patients with tumors like Lee’s: one, from a team at the University of Pennsylvania, is in the ASAIO Journal 44 (1998), pp. 219–21. Another, from a team in Delhi, India, is in the Journal of Cardiothoracic and Vascular Anesthesia 15 (2001), pp. 233–36. Both teams describe finding the strategies not through careful research but the way many breakthroughs are found—through happenstance and necessity.
EDUCATION OF A KNIFE
20 K. Anders Ericsson’s book on human performance is The Road to Excellence (Mahwah, N.J.: Lawrence Erlbaum Press, 1996).
27 The Great Ormond Street Hospital’s landmark report on their learning curve for doing the switch operation is Bull, C., et al., “Scientific, ethical, and logistical considerations in introducing a new operation: a retrospective cohort study from paediatric cardiac surgery,” British Medical Journal 320 (2000), pp. 1168–73. The British report is quoted in Hasan, A., Pozzi, M., and Hamilton, J. R. L., “New surgical procedures: Can we minimise the learning curve?” British Medical Journal 320 (2000), pp. 170–73.
28 The Harvard Business School research is reported in several chapters and papers, including Pisano, G., Bohmer, R., and Edmondson, A., “Organizational Differences in Rates of Learning Evidence from the Adoption of Minimally Invasive Cardiac Surgery,” Management Science 47 (2001); Bohmer, R., Edmondson, A., and Pisano, G., “Managing new technology in medicine,” in Herzlinger, R. E., ed., Consumer-Driven Health Care (San Francisco: Jossey-Bass, 2001).
THE COMPUTER AND THE HERNIA FACTORY
37 Edenbrandt’s study: Heden, B., Ohlin, H., Rittner, R., and Edenbrandt, L., “Acute myocardial infarction detected in the 12-lead ECG by artificial neural networks,” Circulation 96 (1997), pp. 1798–1802.
Baxt, W. G., “Use of an artificial neural network for data analysis in clinical decision-making: the diagnosis of acute coronary occlusion,” Neural Computation 2 (1990), pp. 480–89.
38 The Shouldice Hospital has published its results widely. One summary is in Bendavid, R., “The Shouldice technique: a canon in hernia repair,” Canadian Journal of Surgery 40 (1997), pp. 199–205, 207.
43 Meehl, P. E., Clinical Versus Statistical Prediction: A Theoretical Analysis and Review of the Evidence (Minneapolis: University of Minnesota Press, 1954).
Dawes, R. M., Faust, D., and Meehl, P. E., “Clinical versus actuarial judgment,” Science 243 (1989), pp. 1668–74.
44 A nice tutorial on, and summary of, medical neural networks is in Baxt, W. G., “Application of artificial neural networks to clinical medicine,” Lancet 346 (1995), pp. 1135–38.
45 Herzlinger, R., Market-Driven Health Care: Who Wins, Who Loses in the Transformation of America’s Largest Service Industry (Reading, Mass.: Addison-Wesley, 1997).
WHEN DOCTORS MAKE MISTAKES
57 Brennan, T. A., et al., “Incidence of adverse events and negligence in hospitalized patients: results of the Harvard Medical Practice Study I,” New England Journal of Medicine 324 (1991), pp. 370–76.
Leape, L. L., “Error in medicine,” Journal of the American Medical Association 272 (1994), pp. 1851–57.
Bates, D. W., et al., “Incidence of adverse drug events and potential adverse drug events,” Journal of the American Medical Association 274 (1995), pp. 29–34.
Localio, A. R., et al., “Relation between malpractice claims and adverse events due to negligence: results of the Harvard Medical Practice Study III,” New England Journal of Medicine 325 (1991), pp. 245–51.
63 Reason, J., Human Error (Cambridge: Cambridge University Press, 1990).
64 The history of anesthesia’s success against error is described in Pierce, E. C., “The 34th Rovenstine Lecture: 40 years behind the mask—safety revisited,” Anesthesiology 84 (1996), pp. 965–75.
65 Cooper, J. B., et al., “Preventable anesthesia mishaps: a study of human factors,” Anesthesiology 49 (1978), pp. 399–406.
69 The work of the Northern New England Cardiovascular Disease Study Group has been published in numerous studies, but a summary can be found in Malenka, D. J., and O’Connor, G. T., “The Northern New England Cardiovascular Disease Study Group: a regional collaborative effort for continuous quality improvement in cardiovascular disease,” Joint Commission Journal on Quality Improvement 24 (1998), pp. 594–600.
70 Brooks, D. C., ed., Current Review of Laparoscopy, 2nd ed. (Philadelphia: Current Medicine, 1995).
NINE THOUSAND SURGEONS
75 One can find information about the annual American College of Surgeons convention at www.facs.org.
WHEN GOOD DOCTORS GO BAD
88 The tale of Harold Shipman is described in detail in Eichenwald, K., “True English murder mystery: town’s trusted doctor did it,” New York Times, 13 May 2001, p. A1.
Prosecutors in John Ronald Brown’s trial alleged his involvement also included attempting to plug a Los Angeles woman’s leaking breast implants with Krazy Glue. He was convicted of second-degree murder in the death of Philip Bondy, whose leg was amputated, and sentenced to fifteen years. For more details see Ciotti, P., “Why did he cut off that man’s leg?” LA Weekly, 17 December 1999.
For the story of James Burt see Griggs, F., “Breaking Tradition: Doctor Steps in to Stop Maiming ‘Surgery of Love,’ ” Chicago Tribune, 25 August 1991, p. 8.
94 For data on substance abuse in doctors see Brook, D., et al., “Substance abuse within the health care community,” in Friedman, L. S., et al., eds., Source Book of Substance Abuse and Addiction (Philadelphia: Lippincott, 1996).
Two comprehensive sources of information on the occurrence of mental illness in the general population are Mental Health: A Report of the Surgeon General (Rockville, Md.: U.S. Department of Health and Human Services, 1999); Kessler, R. C., et al., “Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey,” Archives of General Psychiatry 51 (1994), pp. 8–19.
The estimate of the percentage of problem physicians comes from Marilynn Rosenthal’s review “Promise and reality: professional self-regulation and ‘problem’ colleagues,” in Lens, P., and Van der Wal, G., eds., Problem Doctors: A Conspiracy of Silence (Netherlands: IOS Press, 1997), p. 23.
Marilynn Rosenthal’s book is The Incompetent Doctor: Behind Closed Doors (Philadelphia: Open University Press, 1995).
98 Kent Neff presented the data from his work with problem physicians at the Annenberg Conference on Enhancing Patient Safety and Reducing Errors in Health Care, Rancho Mirage, California, 9 November 1998.
FULL MOON FRIDAY THE THIRTEENTH
111 Scanlon, T. J., et al., “Is Friday the 13th bad for your health?” British Medical Journal 307 (1993), pp. 1584–89.
William Feller’s study of the Nazi bombing of London was published in his book An Introduction to Probability Theory (New York: Wiley, 1968).
On Texas sharpshooting, see Rothman, K. J., American Journal of Epidemiology 132 (1990), pp. S6–S13.
112 Buckley, N. A., Whyte, I. M., and Dawson, A. H., “There are days . . . and moons: self-poisoning is not lunacy,” Medical Journal of Australia 159 (1993), pp. 786–89.
&nbs
p; 113 Guillon, P., Guillon, D., Pierre, F., and Soutoul, J. H., “Les rythmes saisonnier, hebdomadaire et lunaire des naissances,” Revue Francaise de Gynecologie et d’Obstetrique 11 (1988), pp. 703–8.
The two best summaries on the moon and human behavior are Martin, S. J., Kelly, I. W., and Saklofske, D. H., “Suicide and lunar cycles: a critical review over twenty-eight years,” Psychological Reports 71 (1992), pp. 787–95; and Byrnes, G., and Kelly, I. W., “Crisis calls and lunar cycles: a twenty-year review,” Psychological Reports 71 (1992), pp. 779–85.
THE PAIN PERPLEX
118 There is a wide literature on the puzzle of chronic back pain. Among the texts and studies I found useful were Hadler, N., Occupational Musculoskeletal Disorders (Philadelphia: Lippincott Williams and Wilkins, current edition 1999); and Haldeman, S., “Failure of the pathology model to predict back pain,” Spine 15 (1990), p. 719.
The rather curious findings from back MRIs for people in the general population is from the Cleveland Clinic: Jensen, M. C., et al., “Magnetic Resonance Imaging of the lumbar spine in people without back pain,” New England Journal of Medicine 331 (1994), pp. 69–73.
119 Hilzenrath, D., “Disability claims rise for doctors,” Washington Post, 16 February 1998.
120 Descartes’s description of pain is found in the Meditations (1641).
Henry K. Beecher’s studies of battlefield pain were published in two places: “Pain in Men Wounded in Battle,” Bulletin of the U.S. Army Medical Department 5 (April 1946), pp. 445; and “Relationship of Significance of Wound to Pain Experienced,” Journal of the American Medical Association 161 (1956), pp. 1609–13.
Ronald Melzack and Patrick Wall’s classic paper describing Gate-Control Theory was “Pain Mechanisms: A New Theory,” Science 150 (1965), pp. 971–79.
