BOB HIRSHON (host):
How medical studies go wrong. I’m Bob Hirshon and this is Science Update.
Hormone replacement therapy reduces heart attacks. A low-fat diet reduces cancer. You may remember these findings, made by statistical studies of large groups of people. And you may also remember that they were later refuted. So what went wrong? Epidemiologist John Ioannidis of the University of Ioannina in Greece says claims like these get refuted as much as 85 percent of the time. It’s in part because researchers tend to report only their highly significant results and not their less sensational ones.
JOHN IOANNIDIS (University of Ioannina, Greece):
So what is published is really a very selective cut that may have happened just due to chance—that we found that something seemed to be effective and useful for health while in fact it was just a chance finding.
He says the media and the public should beware of claims until they’ve been confirmed by rigorous clinical trial. I’m Bob Hirshon, for AAAS, the science society.