15 February 2002
MORE PHARMA-AUTHOR LINKS
After last week's news that perhaps as many as "50% of the articles
on drugs in the major journals across all areas of medicine" are actually
ghostwritten by industry, raising "the question of whether the eminent
scientists whose names were on the papers ever saw the raw data from the
trials"... Scandal of scientists who take money for papers ghostwritten
by drug companies
News: Authors of guidelines have strong links with drugs industry
Alison Tonks, Bristol
BMJ 2002;324:383 ( 16 February )
Most guidelines on clinical practice are written by experts with undisclosed links to the pharmaceutical industry, researchers from Toronto, Canada, say in an article in the journal of the American Medical Association (JAMA2002; 287:612-7[Medline]).
In a survey of nearly 200 authors of 44 clinical guidelines, 87% of respondents admitted to financial links with one or more pharmaceutical companies. Over half of the authors had been paid to conduct research, over a third had been an employee or consultant, and two thirds had received fees for speaking.
On average each respondent had links with 10 companies, including companies whose products they recommended in guidelines. Only one of the 44 guidelines carried a declaration of the authors' competing interests.
"I'm not at all surprised by these findings," says Dr Bob Goodman, internist at Columbia University in New York and founder of No Free Lunch, the campaign for independent prescribing. "Other studies have already shown extensive links between physicians, researchers, and even policy makers and the pharmaceutical industry. It's particularly worrying, though, in the case of practice guidelines. These documents are widely distributed and intended to change physicians' practice.
"Any influence of a drug company on an individual author is multiplied thousands of times. Worse, there's a subjective element to the recommendations in clinical guidelines that makes them particularly vulnerable to bias."
Most (93%) of the study's respondents said their relationships with pharmaceutical companies did not affect their recommendations on treatment. But evidence cited by the researchers makes it clear that accepting money from drug companies alters prescribing, drives requests for additions to hospital formularies, and contributes to publication bias.
The researchers were unable to check whether authors' financial interests influenced the treatments recommended in guidelines, because there were too few independent guidelines in the sample to make a meaningful comparison.
The study looked at guidelines on the management of 10 common diseases, including asthma, coronary artery disease, heart failure, depression, and peptic ulcer. All the guidelines were endorsed by professional societies in North America or Europe and were published between 1990 and 1999.
The researchers contacted 192 authors, but only 52% responded, despite a second mailing. They blame the low response rate on authors' reluctance to admit to links with drug companies and speculate that those who did not reply had even more to declare than those who did. If so, the links between authors of guidelines and the drugs industry are even more widespread than the study indicates, they conclude.
The researchers want a formal process built in to guideline development
that forces authors to declare their financial interests. They also want
written declarations of competing interests on every guideline.
"the New England Journal of Medicine published a negative review of a widely hailed book linking chemical pollution to cancer but failed to disclose that the review was written by the medical director of a major polluter. In another case, the journal Neurobiology of Aging published a panel report endorsing a proprietary blood test for Alzheimer's disease but did not disclose that one of the panelists held a patent on the test and another was a co-founder of the company that planned to market the test. Nonetheless, most scientific journals still do not disclose authors' conflicts of interest..."
Scientists Call on Journals to Disclose Authors' Conflicts of Interest
ngin bulletin archive