Obviously, these prospective studies are harder and more expensive to do and take longer to produce results. But they usually yield a stronger conclusion. Indeed, John Breitner of the University of Washington announced in Philadelphia the results of three prospective studies, involving more than 8,000 people (and, in the peculiar language of scientific investigation, more than 30,000 person-years). Across the board, statins had no protective effect.

E-mail your comments or suggestions to Josh Fischman: pulse@usnews.com Sign up for our biweekly E-mail newsletter: Health Smart

6/16/04: Quick migraine treatment means less pain 6/3/04: Turning your immune system into a bad yo-yo 5/26/04: The danger of doctors' ties 5/26/04: Zinc helps treat pneumonia A complete archive of Pulse columns

What's going on? Barry Reisberg, an Alzheimer's expert from New York University, suggests an explanation: In the earlier, retrospective studies, the sample was biased. "You need to be 'smarter' in the first place to remember to take your pills regularly," he says. So while studies appeared to show a link between the people with the most statins and the lowest risk of Alzheimer's, the real association might have been with people who had the most on the ball to begin with. They took better care of themselves and had more of what neuroscientists call "cognitive reserve," and they ended up functioning better. The same phenomenon could explain why retrospective studies indicated that the antioxidant vitamin E had a protective effect, but a new prospective study by Mayo Clinic researchers, lasting three years and focusing on people at high risk for getting Alzheimer's, found no such link.

There's little reason to avoid vitamin E—it's good for you—or to avoid statins if you're at risk for heart disease. But to protect your mind from Alzheimer's? Not really, DeKosky concludes. "You can't say, 'If I do this, I will not get the disease.' "

Previous Page