How should doctors decide what to do: evidence, intuition, or both?

Most people think medicine is a science. Mostly, that’s true. But only mostly. In fact, medicine makes frequent use of science, but it’s not entirely science — it’s more of a mishmash of science, experience, intuition, guesswork, and blind luck. Many wish this were not so, but it is.

The last decade has seen an attempt to bring more scientific rigor into medical practice. The movement is called Evidence Based Medicine. The notion seems simple, one few could argue with: take a critical look at all the research that’s been done about a particular medical treatment and see if, on balance, the treatment works. The process has several important principles, among which are to establish in advance how much credance we should place on various research studies, especially when they conflict with one another. To do this we assign a hierarchy of reliability of the evidence. The weakest evidence is expert opinion alone — after all, experts can be wrong. The reliability of evidence works its way up from there, through case reports to uncontrolled trials to the strongest evidence — the randomized, placebo-controlled trial.

These trials compare the results between two groups of patients: those who got the treatment and those who didn’t. One key to this is the placebo part: neither group knows until the trial is over who got the treatment and who got the “placebo,” the sugar pill. A second key is patients are randomly assigned to the treatment or the placebo groups. A final crucial element is that the doctors caring for the patients don’t know themselves who’s getting the treatment and who’s getting the placebo until after the trial is done. Then the investigators look at the data and see if the treatment works, if it’s better than the placebo.

Sounds simple. It isn’t, though, especially as it applies to the Holy Grail of evidence based medicine. For one thing, for some things there isn’t a good placebo — a major operation, for example. For another, physicians have only studied a tiny fraction of all medical conditions, typically those which affect a lot of people, are controversial for one reason or another, or which look financially promising to drug companies.

So how do I and my colleagues decide what to do? We use hard evidence if there is any. We do what makes sense in light of what we do know about the condition or others like it. We tend to do what we have been taught, and we respect the opinions of our medical forebearers. Sometimes we have no idea what to do, in which case it is usually better to do nothing. In short, we still rely to some extent on experience, intuition, guesswork, and blind luck. For myself, I actually like things to stay at least a little bit that way.

If you want to learn more about evidence based medicine, the guiding organization is the Cochrane Collaboration, a huge group of valiant volunteers who scour the medical literature to collect information about specific ailments and write reviews about what the data show. The Cochrane site is here.