KEY POINTS:
In the 1950s, one in 30 babies in the US died at birth. Why? How did they fix that? In every war they have fought in, American soldiers were more likely to survive their injuries than soldiers of the previous war - up to the Gulf War of 1991, when their survival rate was no better than it had been in Vietnam. Now, though, it is better again. What happened there?
Throughout America, specialist medical centres follow established best-practice in treating cystic fibrosis, but life expectancy varies enormously. Why?
Atul Gawande's answer to all these questions is simple - it's the result of doctors deciding they have to improve the way they work.
This is a controversial observation, in so many ways. Many professionals - in health and elsewhere - think they already do their jobs as well as they can. Deciding to be better takes a particular mindset.
In addition, medical progress is usually assumed to flow from breakthroughs in science and technology. Gawande suggests many of the big improvements in our time don't come out of the lab at all, but from changes to the way things are done.
He also says doing better requires the ability to recognise better when you find it, and for this you need to count things. In fact, that's his working definition of science.
In other words, though political slogan-writers may not want to hear it, paperwork is important. It allows doctors to track the relations between symptoms and diagnosis, treatment and outcome. It separates good medicine from a reliance on memory, anecdote and prejudice.
Doing better is not a new way to describe the pursuit of excellence. Where that earlier concept assumed we could become perfect cogs in a machine, Gawande knows we live in an imperfect world and doctors won't always get it right. They should always try, he says, and we should always expect them to. But we should also understand they will not always succeed.
Fundamentally, Gawande believes in "positive deviancy" - in not accepting the norms. In medical terms, that translates into not giving up on patients, but the lessons are by no means limited to medicine.
Gawande has precocious credentials: surgeon at a top Boston hospital, associate professor at Harvard, former Rhodes scholar and adviser to Bill Clinton, staff writer at the New Yorker and father of three. This is his second book. He is 32.
But what is most remarkable about him is not in the CV; it's the way he writes, which is conversational and clear, provocative but also reassuring. In the essays of this book he conveys a delight in both telling a story and discovering the big picture among the details.
His stories are extraordinary, not least because they show that doing better is not always about money. That high infant-mortality rate in the 1950s was due to the prevailing assumption, now shocking, that babies born malformed, too small or not breathing well were unlikely to survive. So they were listed as stillborn, placed out of sight, and left to die.
Then an anaesthetist called Sandra Apgar invented a way to measure the health of a baby: two points if it cried well, two if it moved all four limbs, and so on. And all babies were treated as if they were going to live.
Very quickly, they discovered that a baby with a low Apgar score at one minute after birth would often respond to oxygen and warmth and achieve an excellent score after five minutes. Mortality rates plummeted, neonatal intensive care sprang to life.
Soldiers now survive horrific injuries in Iraq because urgent surgery begins in the field, within the first crucial hour, and is completed elsewhere. Cystic fibrosis patients treated at the best hospitals live longer than others because their doctors go way beyond best practice.
In India, Gawande discovered surgeons doing so much with so little, he decided "the abilities of an average Indian surgeon outstripped those of any Western surgeon I know". For a Harvard medical professor, even one happy to be controversial, that's quite a statement.
- Profile books, $39.99
* Simon Wilson is an Auckland reviewer
