The Open Curriculum

The Body · Level 2

Better + Complications

Atul Gawande · 2002–2007

Thesis

Medicine is an imperfect science practised by fallible people under uncertainty, and the difference between ordinary and excellent care comes less from brilliance or new technology than from diligence, honest measurement of results and the willingness to keep improving the systems that doctors work within.

Strength of evidence: mixed. The books are essays built on cases; the data they cite, such as wartime survival rates and infection reductions, come mainly from observational before-and-after studies rather than trials.

The main ideas

The learning curve

Atul Gawande, a surgeon at Brigham and Women's Hospital in Boston and a staff writer for The New Yorker, wrote Complications (2002), a National Book Award finalist, during his surgical residency. Its first essay confronts an uncomfortable fact: surgeons can acquire skill only by practising on patients.

His case is placing central lines, catheters threaded into large veins near the heart. As a resident he failed repeatedly and had to be rescued by senior surgeons; by his later years of training he did the procedure almost without thinking. Every surgeon's competence was built the same way, on patients who received care from someone still learning.

The mechanism is that manual and judgment skills cannot be acquired from books or simulation alone. They require repetition with real consequences, so someone must bear the cost of a learner's inexperience.

Implication: the system of training, not individual consent, determines who bears that cost. Patients who can choose their doctor tend to choose the experienced one, which pushes the learning curve onto patients who cannot.

Machines and factories

Gawande argues that for narrow, repetitive tasks, routine beats individual brilliance. He describes a 1996 study at Lund University Hospital in Sweden, in which the head of coronary care, Hans Ohlin, an experienced cardiologist, and a computer program each read 2,240 electrocardiograms. The computer identified 738 heart attacks, the cardiologist 620, about 20 percent fewer. He connects this to the psychologist Paul Meehl's findings that statistical rules often outperform expert judgment.

His second case is Shouldice Hospital near Toronto, which repairs only hernias. Its surgeons each perform hundreds of hernia operations a year, and its recurrence rate is far below the average elsewhere.

The mechanism is that repetition and standardization reduce variation, and variation is where errors occur. A specialist doing one operation many times, or a machine applying the same rule every time, avoids the inconsistency of a generalist.

Implication: for routine tasks, prefer the system or specialist with the highest volume and the most standardized method, even over a more prestigious generalist.

Errors are systemic

In "When Doctors Make Mistakes," Gawande argues that error in medicine is not mainly a problem of bad doctors. All doctors make serious mistakes; the question is what happens next. He describes the morbidity and mortality conference, where surgeons review cases that went wrong, and its limitation: it focuses on the individual's error rather than on the system that made it likely.

His model of a better approach is anesthesiology, which from the 1970s studied its errors systematically and redesigned equipment and procedures, and became dramatically safer as a result.

He also examines the opposite problem: doctors who decline. "Hank Goodman," a pseudonym, was an excellent orthopedic surgeon whose judgment deteriorated, partly through depression, until he was treating serious infections with pills instead of surgery. His colleagues were slow to act. The medical profession, Gawande argues, has weak mechanisms for confronting colleagues who are failing.

Implication: when errors recur, redesign the process rather than blaming the person. And when a colleague's performance is declining, act early, because no one else will.

Mystery and uncertainty

The second and third parts of Complications concern what medicine does not know. A Boston architect suffers severe back pain that no test can explain. Gawande uses the case, and Australia's epidemic of repetitive strain injury in the 1980s, to argue that pain is shaped by the brain and by social expectation as well as by injury. He also describes a television anchor whose uncontrollable blushing was ended by a nerve-cutting operation, and a man whose life was transformed by gastric bypass surgery.

In "The Case of the Red Leg," a young woman's swollen leg looks like an ordinary skin infection, but Gawande, having recently seen a case of the rare flesh-eating infection necrotizing fasciitis, suspects it and orders a biopsy. The decision rests on intuition shaped by a recent experience, which might as easily have misled him.

He laments the decline of the autopsy, which once revealed how often diagnoses were wrong. And in "Whose Body Is It, Anyway?" he argues that patients often want competence and kindness from their doctors more than the burden of making every decision themselves.

Implication: treat medical decisions as judgments under uncertainty. Ask what the doctor does not know, as well as what they recommend.

Diligence

Better (2007) asks how doctors can perform better, and begins with diligence. In "On Washing Hands," Gawande reports that staff at his hospital washed their hands one-third to one-half as often as they should, and that incentives and reminders barely helped.

What worked, at the veterans' hospital in Pittsburgh, was an approach called positive deviance: asking staff at every level, from cleaners to surgeons, how to solve the problem, and adopting their ideas, such as putting hand-gel dispensers exactly where they said they were needed. Infections with the drug-resistant bacterium MRSA fell sharply.

The same theme runs through his account of American military surgery in Iraq and Afghanistan. About 10 percent of wounded soldiers died, compared with about 24 percent in Vietnam. Gawande examines how military surgeons achieved this, and presents it as an improvement in how care was organized rather than a technological breakthrough.

Implication: before seeking a new technique, ask whether the simple, known practices are being done every time. Ask the people doing the work how to make them reliable.

Measuring performance

Gawande argues that measurement drives improvement. In "The Score," he describes how Virginia Apgar's simple score for newborns, introduced in 1952, gave obstetrics a way to measure its results, and drove the field to standardize, with fetal monitors, epidurals and a move from forceps toward cesarean sections. The score made childbirth safer, and also made it more industrial.

In "The Bell Curve," he examines centers treating cystic fibrosis, whose results were published by the national foundation. The results formed a bell curve. A respected center in Cincinnati turned out to be merely average. At the top was Warren Warwick's center in Minneapolis, whose patients survived to a median age of 47, against a national average of 33, because Warwick pursued every small improvement in lung function with relentless intensity.

Implication: find out where your performance sits on the curve. Without honest measurement, every practitioner assumes they are above average.

Ingenuity and the positive deviant

Better ends with ingenuity. In a government hospital in Nanded, India, a handful of general surgeons serving more than a million people had taught themselves an extraordinary range of operations out of necessity.

Gawande closes with five suggestions for becoming what he calls a positive deviant: ask an unscripted question, to see patients as people; do not complain; count something, to learn from data; write something, to reflect and contribute; and change, by being willing to adopt new ideas.

Implication: improvement in any profession depends less on talent than on the habits of curiosity, measurement, reflection and willingness to change.

Limits

Related