TORONTO — Medical authors with ties to diabetes drug makers were more likely to publish favourable articles on a controversial diabetes medication than authors who had no such ties, a new study reveals.

Over 90 per cent of scientists who wrote positively about the drug Avandia (rosiglitazone) in studies, commentaries and letters published in medical journals had financial relationships with drug companies, said the study, written by researchers from the department of internal medicine at the Mayo Clinic in Rochester, Minn.

Authors with financial ties to industry were more than three times more likely to write favourably about the drug than those who didn't. And those with industry ties who wrote commentaries were six times more likely to back continued use of the drug, which came under fire after a large review published in 2007 linked the GlaxoSmithKline drug to a significant increased risk of heart attack.

The scientists who did this new study, published Friday in the journal BMJ, said they couldn't prove financial ties provoked the positive reviews.

But people who study the influence the pharmaceutical industry wields over scientific publishing were more inclined to draw the link.

"My conclusion is that their conclusion is wimpy," said Dr. Jerome Kassirer, former editor of the New England Journal of Medicine and a long-time critic of the ties between drug companies and the medical profession.

"What else could that association be related to?" asked Kassirer, author of "On The Take: How Medicine's Complicity With Big Business Can Endanger Your Health."

"At face value, when you see results like this, you have an extremely strong suspicion that the industry affiliation had an influence on their opinion."

After the link between the drug and heart attacks surfaced, controversy raged in the medical literature about the extent to which the risk was real. Dr. Victor Montori and some colleagues at the Mayo decided to try to see whether industry ties were influencing the debate.

They found 202 studies, reviews, commentaries and letters to journals discussing Avandia and heart attack risk, then looked for conflict of interest declarations from the authors.

Only 53 per cent of the publications had conflict of interest declarations — surprising in itself as a number of journals now routinely insist on collecting and publishing that information.

For the record, the Mayo authors declared that they had no financial ties to any commercial entities that might have an interest in the subject of their study and nor did any of their spouses, partners or children.

Montori, a diabetes specialist, and his team found that of the authors for whom conflict of interest statements were available, 90 had a financial conflict. That represents 45 per cent of the authors of the studies for which conflict statements were available.

In some cases, some authors declared no conflict of interest, but the team found they actually had undeclared financial ties by looking up other scientific papers those researchers had written.

"Authors who were unfavourable on the issue of rosiglitazone (Avandia) safety were largely free of identifiable financial conflicts of interest," Montori and his colleagues wrote.

In addition to expressing concern about how the scientific record may have been skewed by the financial interests of some researchers, the Mayo team said the findings suggest journals need to be more demanding of their authors.

"We are hoping for the science to be the science. But in this case, science is conducted by scientists. And scientists are human. And humans have ways of introducing their own preferences into things," Montori said in an interview.

"If you're trying to be an evidenced based clinician and you want your practice to reflect the science, the biggest enemies are the forces that distort ... the scientific record."

"And there is good evidence that for-profit interests have both contributed extensively to the record — and for that we must all be grateful — and have also contributed significantly to its distortion, for which they should be severely punished."

Dr. Joel Lexchin, a pharmaceutical industry critic who teaches at Toronto's York University, said researchers, journals, drug companies and governments all play a part in how the pharmaceutical industry has been allowed to influence science.

"All of them bear some part of the responsibility," Lexchin said. "And who's left in the middle not knowing what to do are the doctors who are prescribing these drugs and the people who are taking them."