Washington Post: Sunday Book World

Cooking the Books

John Abramson, a clinical instructor at Harvard Medical School and the author of Overdosed America: The Broken Promise of American Medicine (HarperCollins, $24.95), has done what most physicians don't have the time or inclination to do: He's not only read the peer-reviewed articles upon which doctors base their prescribing habits but also has examined the actual study data for some of today's most popular drugs. As a primary care doctor in Massachusetts, he constantly found himself on the receiving end of drug-company pitches. And, he writes, he believed in the integrity of the medical information system -- until he sat down and examined the clinical trial results in detail.

The strength of Overdosed America is in these close readings of the research. Abramson walks the reader through the contradictions he's discovered between the exorbitant claims made for the products and the actual study data -- or between the data and the subsequent medical guidelines promulgated by government-sponsored panels, whose members, he reports, often have financial ties to the companies that make the drugs they're recommending.
Abramson argues, for example, that the benefits of the cholesterol-lowering drugs called statins are nowhere near as great as generally believed, and certainly do not approach the advantages conferred by lifestyle changes that would cost much less. Reviewing studies of the blockbuster anti-arthritis drugs Celebrex and Vioxx conducted years before the recent announcement of Vioxx's dangers, he reports that the two medications, although marketed as safer than an earlier and cheaper generation of products, can actually cause more severe side effects. And he recounts the now-famous story of how drug companies and doctors persuaded millions of menopausal women to go on hormone replacement therapy, which later proved to significantly increase the chances of suffering from breast cancer and heart attacks.
In one particularly enlightening section, Abramson analyzes various methods drug companies use to massage their research data and obtain the results they'd like. "Rigging medical studies, misrepresenting research results published in even the most influential medical journals, and withholding the findings of whole studies that don't come out in a sponsor's favor," he writes, "have all become the accepted norm in commercially sponsored medical research."
Beyond Abramson's focus on the details of medical studies, he covers much the same ground that Angell does, although not quite as comprehensively. Both authors offer a list of suggestions -- some practical, others probably wishful thinking -- for reducing the drug companies' influence. But these ideas remain overshadowed by the much larger and more compelling point: If this is the industry we hope will rescue us from disease, pain and unhappiness, we're in real trouble.
David Tuller is a contributing writer at Salon.com.