|
7/28/04
Alzheimer's on trial
By Josh Fischman
There's a dizzying phenomenon in medical research when it comes to the results of studies. Call it "trial whiplash." One trial of a drug or disease prevention strategy says X works; another trial says X doesn't work at all. Following the conflicting results can make your head swing back and forth fast enough that your neck hurts. And there were a lot of aching necks last week after researchers at a major Alzheimer's conference reported that heart drugs called statins and antioxidant vitamins didn't seem to protect people from the disease, even though earlier studies had said there was a protective effect.
|
|
 |
|
6/16/04:
|
|
|
6/3/04:
|
|
|
5/26/04:
|
|
|
5/26/04:
|
|
|
|
|
 |
|
It's not that scientists are waffling here (although they've been known to do that at other times.) What's going on is that there are two very different kinds of studies, and here they have produced very different results. Both are necessary for figuring things out, but only one gets you close to a final answer. Scientists, not just the public, often mix up the conclusions that can be drawn from each type. Says Steven DeKosky, director of the Alzheimer's Disease Research Center at the University of Pittsburgh: "We always confuse studies that look backwards with prevention studies that look forwards."
Studies that look backwards are part of the "hypothesis factory," the process scientists use to generate ideas about what might work. (For sophisticated and clear examples of how this works, check out the Alzheimer Research Forum. Not only is this a place where researchers meet and debate ideas—including things like the effectiveness of statins—but patients and caregivers also can get authoritative information about drugs and therapies and state-of-the-art thinking about the disease.) Take an elderly population, some of whom have Alzheimer's and some of whom don't. Looking back at their medical records, try to see if there were any differences in their lifestyles and medications. Several researchers have done this, and one of the differences seemed to be that more people with intact mental faculties had taken statins. Perhaps—and this is the hypothesis—statins offer some protective effect.
The trouble is that the past is a vague and shadowy place. Statins may not have been the only thing separating Alzheimer's patients from healthy patients. Maybe the healthy ones took other drugs. Maybe they exercised more. Maybe they did more crossword puzzles. The point is that, looking backwards, you can't isolate statins as the only meaningful difference. But at least these studies have identified statins as something worth investigating.
To get further, you need what researchers feel is the gold standard of trials: a prospective, double-blind, placebo-control study. Or in DeKosky's words, a study that looks forward. Take a group of people, as identical as you can realistically match them up to be. Then give some of them statins, the others a dummy pill called a placebo. The hope here is that the statins are the only meaningful difference between the two groups, so you have eliminated the confounding factors that might have confused the earlier, retrospective trial. Follow these people for many years, and see if the statins make any difference in the outcome.
|
