The finding flies in the face of a large body of evidence showing industry studies tend to promote new drugs and downplay potential side effects.

That potential bias has fueled concerns that medical care could be guided by warped science, not least because more and more research is being done or paid for by companies with money riding on the results.

The new report, published in the journal Arthritis & Rheumatism, shows 37 of 49 drug trials funded exclusively by manufacturers had positive results. When non-profits such as the government or a foundation picked up the tab, 11 of 16 trials found the studied drug to be effective.

That difference -- seven percent -- is too small to be statistically reliable, meaning that a trial's outcome didn't depend on who sponsored it.

Still, Dr. Nasim Khan, who led the new work, was quick to add a caveat.

"A single study limited to (clinical trials in a single disease) is insufficient to completely dispel worries about potential for bias when financial conflicts of interest exist," Khan, of the University of Arkansas for Medical Sciences in Little Rock, told Reuters Health by email.

Other experts were intrigued by the new data.

"It's an interesting finding, there is no question," said David J. Rothman, who heads the Center for the Study of Society and Medicine at Columbia University in New York and has studied conflicts of interest in medicine.

"It would be nice if industry studies were not biased as the literature suggests," he told Reuters Health. "Nobody wants to demonize the industry."

BIAS UNIVERSAL

But Rothman said Khan's team is emphasizing the wrong point.

"The finding is not that industry doesn't have a proclivity to publish positive results. The finding is that not-for-profits also have a proclivity to publish positive results," he said. "The problem may be the journals rather than industry or not-for-profits."

Past research has shown that experts who keep the gates at medical journals tend to prefer "splashy" findings -- in other words, studies that show one treatment works better than another. (See Reuters Health story of November 22, 2010.)

There is even a journal dedicated to publishing negative findings -- the Journal of Negative Results in Biomedicine -- in an attempt to give doctors a more balanced view of medical treatments.

The new study did show that industry research is published less frequently than trials funded by non-profits. But the reports that drugmakers did publish appeared to be more thorough than the others, Khan said.

His findings are based on a review of 103 past trials of rheumatoid arthritis drugs such as methotrexate or corticosteroids. Just over half of those studies were funded by drugmakers, whereas 18 percent were paid for by non-profits, six percent had mixed funding and the rest didn't specify the funding source.

Both Rothman and Dr. Gisela Schott, who has also studied drug trial financing, had doubts about the methods used by Khan and his colleagues.

In 2010, Schott published a comprehensive summary of earlier research concluding that industry-funded drug trials give a distorted picture of the evidence that is likely to favor the company's product.

For example, one classic study from 1998 shows scientists from the tobacco industry were seven times more likely to deem secondhand smoke safe than non-industry researchers.

TRIALS NOT ANSWERING RELEVANT QUESTIONS

Schott, of the German Medical Association's Drug Commission in Berlin, also looked at the conclusions of drug trial reports, which may spin the results in a positive direction -- something Khan's team did not.

She also said the new work didn't address whether trials asked questions that are relevant to doctors and patients.

"For a prescribing doctor this is, for instance: is the new treatment better than the old one?" she told Reuters by email. "Studies which test only against placebo are not of much use in this situation."

That concern is bolstered by a new study, published in the Archives of Internal Medicine earlier this month. (See Reuters Health story of February 17, 2012.)

In that study, researchers found that new rheumatoid arthritis medications were usually only compared to an inactive placebo pill instead of currently used medications.

That leaves doctors and patients unable to judge whether they would be better off by switching to the new drug, and it raises ethical concerns about withholding effective medicine from patients in the trials.

To Khan, however, it's not black and white.

"Industry funded clinical trials are more focused on studying newer drugs for RA (rheumatoid arthritis), while non-profit funded trials are more likely to study established drugs and different strategies to treat RA," he said. "Each of these purposes have important role in advancing care of RA patients."

SOURCE: bit.ly/zqqqa5 Arthritis and Rheumatism, online January 24, 2012.

• A Permanent Economic Emergency