A happy face on antidepressants?

[Guest guest]

http://www.oregonlive.com/news/oregonian/index.ssf?/base/business/12005439042749\

90.xml & coll=7

A happy face on antidepressants?

Health - Oregon research says published reports give an unbalanced measure of

drugs' effectiveness

Related Documents (PDF):

1

Thursday, January 17, 2008

JOE ROJAS-BURKE

The Oregonian Staff

http://www.oregonlive.com/cgi-bin/prxy/accessor/nph-repository-cache.cgi/base/pd\

f_captions/1200543905274490.pdf

or

http://tinyurl.com/2l39a6

The portrayal of antidepressant drugs in medical journals significantly

overstates their effectiveness, according to a study led by Oregon researchers.

Nearly a third of the clinical trials of antidepressants carried out by drug

companies produced questionable or negative results that never appeared publicly

in print, researchers reported in the New England Journal of Medicine.

" The doctor and the patient have only been aware of good news about these drugs

in terms of efficacy, " said lead author Dr. k , a former drug reviewer

for the federal Food and Drug Administration who now holds positions at the

Portland Veteran Affairs Medical Center and Oregon Health & Science University.

The findings do not imply that the drugs don't work, but rather that doctors and

patients lack a full, nuanced picture of their effectiveness.

The results also highlight a widespread bias problem in reporting on drug

treatments of all kinds, said Dr. Liebeskind, associate director of

neurology at the University of California at Los Angeles.

Denied access to all completed studies, Liebeskind said, doctors and patients

can't make the best possible decisions. Doctors may prescribe drugs that

patients don't need or recommend drugs that are less effective than

alternatives.

Researchers, he said, may unwittingly duplicate clinical trials that have

already been performed, wasting time and money -- and needlessly exposing

volunteers to study risks.

and colleagues dug up FDA records on 74 clinical trials with 12

antidepressant drugs, including Paxil, Prozac and Zoloft. In some cases, the

researchers relied on requests under the Freedom of Information Act to obtain

the federal agency's records. Drug companies seeking to gain marketing approval

must submit detailed reports on clinical trials to the FDA. The agency posts

some but not all of the study data on its Web site.

Antidepressant drugs relieved symptoms better than a placebo in about half of

the clinical trials: 38 out of 74, according to FDA reviewers. But published

reports in medical journals on the same 12 drugs presented a much rosier

picture: 48 of 51, or 94 percent, reported positive results.

FDA reviewers judged that antidepressant drugs failed to work better than a

placebo in 24 of the 74 trials. Only three of these negative studies appeared in

journals.

Positive spins

Most disturbing, Liebeskind said, was finding that published reports on five of

the 24 negative studies portrayed the results as positive. " The publication

authors put a positive spin on things, " he said.

For people who use antidepressants, the study doesn't imply that the drugs

aren't worth taking. said the analysis confirmed that each of the drugs

is superior to a placebo. But he said the drugs are not as effective as

published reports suggest and, with deeper analysis of existing trials,

researchers might be able to identify which brands are more effective than

others.

Drug companies aren't solely to blame for nondisclosure, said Dr. Abaid

with the University of North Carolina at Chapel Hill. Surveys have shown that

medical researchers and journal editors may consider it a waste of time

publishing insignificant or inconclusive studies, although journals in recent

years have tried to correct the bias.

But drug companies can control publishing decisions, for instance, by requiring

doctors to sign nondisclosure agreements, Liebeskind said.

Pfizer Inc., maker of Zoloft, said the company took steps three years ago to

make drug trial results more widely available. " We committed to disclose

clinical trial results within one year after study completion for all of our

marketed products, " spokesman Jack said.

A spokeswoman for Organon, another drug company cited in the study, said of

Organon's four studies listed as unpublished, two appeared in papers combining

results of multiple studies. Spokeswoman Mols said the four studies date

back to the 1980s. More recently, she said, Organon has begun publishing all

clinical trial results on its Web site.

" It's a different time now, " Mols said.

Online database

Ken , a spokesman for the Pharmaceutical Research and Manufacturers of

America, said the industry group established a public, online clinical trial

database in 2004. While that database is voluntary and not comprehensive,

pointed out that Congress last year passed legislation that will require

the National Institutes of Health to set up and run a mandatory clinical trial

database.

said the mandatory database will be " a tremendous step forward. " But, he

said, it leaves a black hole of data on drugs already on the market.

asserts that the FDA should commit more money to publicly posting the

data already in its hands. " The FDA is just sitting on this trove of

information, " he said.

FDA spokeswoman Sandy Walsh said, " That's easier said than done. "

For about three years, Walsh said, the agency has posted trial results, good and

bad, on the FDA Web site once a drug gains approval. But it can take weeks or

months because officials have to pore over the documents line by line to black

out commercial and confidential information.

Joe Rojas-Burke: 503-412-7073, joerojas@...