Tuesday, July 30, 2013

Medicine and Science

Some doctors are very science-oriented. Some are not.

This truth has dawned on me as I have pondered a long series of medical trials in which common medical procedures have been evaluated and found wanting. Here is the latest announcement in this vein:
We usually assume that new medical procedures and drugs are adopted because they are better. But a new analysis has found that many new techniques and medicines are either no more effective than the old ones, or worse. Moreover, many doctors persist in using practices that have been shown to be useless or harmful.

Scientists reviewed each issue of The New England Journal of Medicine from 2001 through 2010 and found 363 studies examining an established clinical practice. In 146 of them, the currently used drug or procedure was found to be either no better, or even worse, than the one previously used. The report appears in the August issue of Mayo Clinic Proceedings.

More than 40 percent of established practices studied were found to be ineffective or harmful, 38 percent beneficial, and the remaining 22 percent unknown. Among the practices found to be ineffective or harmful were the routine use of hormone therapy in postmenopausal women; high-dose chemotherapy and stem cell transplant, a complex and expensive treatment for breast cancer that was found to be no better than conventional chemotherapy; and intensive glucose lowering in Type 2 diabetes patients in intensive care, which not only failed to reduce cardiovascular events but actually increased mortality.
The story goes on, but I think you get the point. Many doctors routinely ignore medical studies that purport to say what works and what doesn't. Why?

I suppose one reason is that the findings of medical research seem to change all the time, and some doctors have probably built up a healthy skepticism about, for example, the latest announcements about healthy diets. So they take all such studies as provisional at best.

But I think there is something else going on. Some doctors don't pay much attention to scientific studies because they are focused on their patients and what seems to help them. These doctors want to do something for their patients; they want to help their patients be healthy and happy. They prescribe a treatment; their patient gets better; they prescribe it again. This drives medical scientists crazy. They think it is no better than medieval medicine; I was sick, I made a vow to St. Roche, and then I got better, so it was St. Roche who cured me. But the experience of prescribing a treatment and seeing the patient recover seems to be a very powerful one for physicians.

Many patient-focused doctors seem to believe that if big studies show that some treatment does, on average, nothing, that is because it helps some patients and hurts others. This was explicitly stated in a recent NY Times essay by Clifton Leaf, a medical writer who thinks people are too genetically diverse for a huge study to say how a drug will work in any one patient. Doctors raise these objections on the pseudo-scientific basis of genetic diversity, but I think they really flow from clinical practice. They prescribe the same treatment to two patients, and one gets better while one gets sicker; they decide that sometimes this works and sometimes it doesn't, and the only way to know is to try. One thing both clinicians and patients hate is to be told that nothing can be done, and both are likely to try some treatment for incurable conditions, hoping that it will help this particular patient even if the chance is low.

When it gives us treatments that work well, everybody likes medical science. But when it doesn't, doctors don't give up; they fall back on much older traditions of care.

1 comment:

Unknown said...

If a given drug helps, say, 10% of patients, hurts 20%, and a given patient accepts the risk, it seems to me the only problem with this is the increased expense. After all, it does help some patients (whether the explanation is genetic diversity or something else). And until we state explicitly as a matter of policy and principle that most medical options--that is, all but the very sure ones that actually pass the followup tests--are like other consumer goods and only open to those willing and able to pay for them, I don't think we'll be able to deny much medical care solely on the basis that it's expensive. Of course, once we state explicitly that most medical options are closed to any but the rich, Republican votes will go down, and pressure to raise taxes will go up.