Enlightening stuff re: anti-depressants, drug companies, FDA

Information Is the Best Medicine
By JOHN ABRAMSON

The measures to make the results of drug trials more accessible to doctors and the public, as well as the federal government's probable stiffer warnings on the use of antidepressants for children, are definitely steps in the right direction. But my experience as a family doctor on the front lines of medicine leads me to believe that these moves won't be enough to curb the influence of the drug companies on our health care.

The impetus for these changes comes largely from revelations in the Food and Drug Administration's review of the new class of antidepressants, known as selective serotonin reuptake inhibitors, in the treatment of depression in children and adolescents. Many studies, the review found, show that the drugs are no more effective than placebos, but significantly increase the risk of suicidal tendencies. So why are doctors writing millions of prescriptions each year to treat depressed children with these drugs?

Much of the problem stems from drug companies' unequal treatment of the clinical trials they sponsor. Findings that support drug sales tend to get published in medical journals, and become accepted as fact. Unfavorable findings often don't see the light of day. This leaves even the most dedicated doctors and best informed patients (and parents) unaware of important evidence about the drugs they're prescribing and using.

In response, the editors of 11 of the world's leading medical journals have agreed to publish only studies that are registered at the outset. Registration creates a trail, making it more difficult to suppress unfavorable results. Legislative efforts are also under way to make completed studies available to the public.

There are many examples, however, of drug sales skyrocketing despite the availability of research showing negative results or serious side effects. An article published in the Archives of General Psychiatry in April 2000 provides a preview of the limited effect that we can expect from registering and posting results of all studies. In this case, researchers obtained the results of all clinical trials of new antidepressants like Prozac, Paxil and Zoloft done from 1987 to 1997, published and unpublished, from the F.D.A. under the federal Freedom of Information Act.

The pooled results showed that an older class of antidepressants, known as tricyclics, was actually more effective, belying all the hype about the "revolutionary'' new antidepressants. The researchers reported that among the longer-term studies (which better reflect actual use), the new antidepressants were 8 percent more effective than placebos, while tricyclics like Elavil were 12 percent more effective. The most disturbing finding was that more than twice as many depressed adults on new antidepressants kill themselves than those taking placeboes. The difference was 8.4 versus 3.6 suicides per 1,000 patients a year, respectively.

How much impact did the availability of these results have on doctors' prescribing patterns? Evidently not much: the new antidepressants remained the best-selling class of drugs in the United States in 2000 and 2001.

The medical journal editors understand that just registering studies isn't enough. Their real goal is full transparency in clinical trials, meaning that research plans and complete data from studies should be available for independent review. But even this is not sufficient to counteract commercial influence, as the cases of the blockbuster drugs Celebrex and Vioxx demonstrate.

The results of large manufacturer-sponsored studies of these two arthritis drugs were published in The Journal of the American Medical Association and The New England Journal of Medicine, respectively, in 2000. The original data upon which both articles are based have been posted on the F.D.A. Web site since February 2001.

Both drug-company sponsored articles led doctors to believe that their patients would be better served by prescribing the new drugs. But the data posted on the F.D.A. Web site tell a very different story. Patients on Celebrex did not have significantly fewer serious gastrointestinal problems than those who took the older types of arthritis medicine.

In fact, during the second six months of treatment (data not even mentioned in the article), users of Celebrex developed serious gastrointestinal problems more often than users of the older drugs. The bottom line is that Celebrex is much more expensive, provides no better relief of arthritis symptoms, and overall causes 11 percent more serious complications than the other drugs.

The data about Vioxx offer an answer to the supposedly unresolved question - whether it increases the risk of heart attacks, strokes and blood clots. Even among people without elevated risk, those taking Vioxx suffered twice as many serious cardiovascular complications as those who took naproxen (sold as Naprosyn or Aleve). Most important, the data on the F.D.A. Web site show that patients taking Vioxx developed 21 percent more serious complications than those who took naproxen.

Since that data was posted more than three years ago, American doctors have prescribed more than $15 billion worth of Celebrex and Vioxx. Full transparency, it appears, doesn't solve the problem of commercial influence either.

Given the drug industry's domination of our medical knowledge, nothing short of an oversight board - modeled after the Federal Reserve Board, will make a real difference. The board's members must serve lengthy terms to avoid political influence and have no commercial ties. This body would make use of the excellent work already being done by the reviewers within the F.D.A. (though too often their analyses remain invisible to the public) as well as the commercial, nonprofit, and government-sponsored research that is already being done.

Like the National Institute of Clinical Excellence in Britain (which has a budget of less than $30 million a year), this board would independently evaluate scientific evidence relevant to selected diseases and treatments to determine optimal medical care including lifestyle changes. The board's recommendations would define the standards of medical care based on what is best for Americans' health, not what will generate the most profit.

If the latest round of reforms relating to drug research stops with registering and posting the results of clinical studies, it will stand only as window-dressing on a method of producing and disseminating medical knowledge that is better designed to serve drug makers' interests than to improve Americans' health.

John Abramson,a clinical instructor at Harvard Medical School, isthe author of the forthcoming"Overdosed America: The Broken Promise of American Medicine. "

--Published in The New York Times,September 18, 2004