Most retractions in scientific journals are due to misconduct, not errors, according to a new analysis. (CBC)

Fraud, plagiarism and similar kinds of misconduct are behind the majority of retractions in scientific journals according to a new analysis that ties the dramatic increase in retractions over the last few years to the growing difficulty in getting projects funded.

Researchers looked at 2,047 biomedical papers where there had been retractions and found that only 21 per cent of them were due to errors. Two-thirds – 67 per cent – were the result of misconduct, including fraud or suspected fraud (43 per cent), duplicate publication (14 per cent) or plagiarism (10 per cent).

"Biomedical research has become a winner-take-all game — one with perverse incentives that entice scientists to cut corners and, in some instances, falsify data or commit other acts of misconduct," says senior author Arturo Casadevall, a professor of medicine at Albert Einstein College of Medicine at Yeshiva University in New York.

Earlier studies had suggested that most retractions were due to inadvertent errors. But the researchers behind this new analysis say that is because the retraction notices written in journals, which are usually written by the papers' authors, frequently gloss over the real reason the retraction is being issued.

"Authors commonly write, 'We regret we have to retract our paper because the work is not reproducible,' which is not exactly a lie," Casadevall writes in a release.

"The work indeed was not reproducible — because it was fraudulent," he says. "Researchers try to protect their labs and their reputations, and these retractions are written in such a way that you often don't know what really happened."

Retraction rates soar

The greater a publication's influence, the more likely it has been to retract earlier findings. Researchers attributed that to what they termed a "prevailing culture" in science that overwhelmingly rewards scientists for publishing in those journals and, in the process, encourages a few to cut corners.

The study found that the percentage of retractions for fraud or suspected fraud has risen tenfold since 1975. The authors link this rise in fraud – especially in the last several years – to the reality that it is becoming increasingly difficult to get research money.

"Scientists are human, and some of them will succumb to this pressure, especially when there's so much competition for funding," Casadevall says.

"Perhaps our most telling finding is what happened after 2005, which is when the number of retractions began to skyrocket. That's exactly when [National Institutes of Health] funding began to get very tight." The NIH is a branch of the U.S. Department of Health.

The one bright spot in the study was that almost half of the retractions – 43 per cent – came from just 38 labs worldwide. Fraud is relatively rare, with only a handful of instances for every 100,000 published studies.

"So while we're not looking at a systemic disease, so to speak, in the scientific community, our findings do indicate a significant problem that needs to be addressed," according to Casadevall.

The study was published online today in the Proceedings of the National Academy of Sciences.

With files from The Associated Press