Many GPs have a dim view of the intellectual gifts of surgeons. But Atul Gawande, Harvard-trained cancer surgeon, is different. He has a degree in politics, philosophy and economics from Balliol College Oxford. He broke off his medical studies to work as health advisor to Bill Clinton. During his residency he started writing for the intellectual weekly, the New Yorker. His two books of essays are available in the UK: Complications: A Surgeon's Notes on an Imperfect Science and Better: A Surgeon's Notes on Performance.

Of the two, I prefer Complications. Written during his years as a resident, it has a freshness which Better doesn’t quite match. In Complications, a situation  – learning to insert a central line, a case of necrotising fasciitis – triggers reflections on how we learn, how we make decisions, how we could do better. In Better, the established surgeon starts with a theme or a question ­– hospital-acquired infections, doctors and the death penalty – and subjects it to the same scrutiny. Complications may be more personal, but it may be Better with its broader scope, that can stimulate a widespread change of practice.

Managers and politicians often view medicine as an exact science and think that rigid guidelines will solve the problem of poor outcomes. “There should be no learning curve as far as patient safety is concerned” said a UK report. Gawande points out that at the front line, an imperfect science is practised by imperfect people with imperfect knowledge. We will never be right all the time, and we need to be honest with ourselves and our patients. But we can do better.

He looks hard at how. The exhilaration and mystery of getting it right, and the mundane but rewarding and often overlooked ways of improving.

There is no doubt that repeated practice and super-specialisation produce the best surgical results. Would you choose to have your hernia repaired by someone who does it day in, day out, or by a generalist, however gifted, who only repairs hernias now and again?

In some fields, the best way of achieving machine-like consistency is to use a machine. Computers beat humans at reading ECGs, for example. They aren’t distracted by bleeps or over-influenced by the case they saw yesterday.

Doctors are reluctant to admit that they can be upstaged by a specialist health worker or a computer, but they should not fear that the art of medicine is thereby dead. We won’t lack for emergencies where we have to rely on that sixth sense, professional intuition, for guidance, but if it fails us – fails the patient – we should have the humility to wonder how we might have done better. And humanity and kindness are as important as they ever were. Let us be glad that machines reduce the number of times we have to tell the patient, or their relatives, that we got it wrong. Let us be thankful that we are freed to concentrate on the traditional bedside role of the doctor – interpreting illness, and helping the patient make the best decisions.

Improving outcomes can be ridiculously simple in concept: standardised anaesthetic machines have reduced the risk of twiddling the knob the wrong way. But someone needs to ask the question, to collect data, to challenge the received opinion. To ask why patients at one cystic fibrosis unit in the USA did so much better than those at the other specialist centres. Could the bell-shaped curve be pushed to the right? The answer, to the amazement of those at the top of the curve was yes, it could.

Improvements can often be ridiculously cheap. But that doesn’t mean they are easily adopted.  Atul Gawande featured in the January 22nd edition of the BMJ as team leader for the WHO surgical safety checklist study. The world in which we live and work is increasingly complicated. Remembering all the things you need to do when starting a prescription for the Pill is bad enough, but on ITU, where the patient’s life is dependent on a hugely complex system of people and procedures, it is impossible. The WHO study found that if surgical teams used simple checklists, mistakes such as leaving sponges inside the patient dropped dramatically and lives were saved, in Toronto just as much as in Tanzania.

But why is there reluctance to adopt checklists? The NHS won’t be implementing them until February 2010. Why not right away?  Does the NHS need a year to save up for pencils and paper?

The truth is, behaviour is hard to change. Look at how badly handwashing is still done: we aren’t that much better than the colleagues who drove Ignaz Semmelveis from his job when he proposed that if doctors washed their hands between patients, fewer women would contract puerperal fever. Doctors’ amour propre doesn’t help. The drama of medicine is attractive – describing how you implemented a checklist isn’t going to draw an awed audience in the pub – and it can be hard to admit fallibility. There is nothing in it for the pharmaceutical industry, so there is no ‘free’ publicity with all the inducements to change behaviour that the drug companies do so well.

What it comes down to is teamwork. We in general practice are generally better at this than many hospital teams. We are less hierarchical, more used to learning from each other. Teams in hospital tend to be less stable, but Gawande demonstrates that those who introduce themselves to each other do better than those who do not. And he suggests that however frustrating work may be, we should resist the temptation to spend coffee time moaning. It doesn’t help. Instead, ask yourself a question and start collecting data . You may come up with an audit that changes practice.

For details of Atul Gawande’s books and links to his articles in the New Yorker, NEJM and other journals go to

First published in NASGP Newsletter 'The Sessional GP'  February/March 2009

No Comments Yet.

Leave your comments