Instances when a reviewer steals an idea from a grant or paper under review are hard to trace—and even harder to quantify.

In a study published in March, scientists reported that “theft of ideas from conference papers and grant proposals” and “manipulation of the review system” are common problems (J. Empirical Res. Human Res. Ethics 1, 43–50; 2006). Nearly half of the 51 scientists interviewed said they were aware of cases where colleagues used another's ideas without obtaining permission or giving credit.

I'm quite sure the majority of reviewers are quite honest. Harvey Marcovitch, Committee on Publication Ethics

But because those instances are largely anecdotal, there was no system in place to tackle them. Beginning last year, however, the US Office of Research Integrity (ORI) took on the authority to intervene in cases where the research is funded by a federal agency. The new definition of scientific misconduct for the first time includes problems that emerge during peer review.

For instance, in one of five cases of possible 'reviewer misconduct' collected by the UK-based Committee on Publication Ethics (COPE), a scientist who had reviewed a paper later submitted a manuscript that included models taken from the original paper. The committee also cites cases where reviewers did not keep manuscripts confidential or did not disclose conflicts of interest.

The ORI has also handled a handful of cases of misconduct during review of grant applications—most of them involving plagiarizing ideas from the applications—at the US National Institutes of Health.

“I'm quite sure the majority of reviewers are completely honest,” says Harvey Marcovitch, chair of COPE and an editor of the British Medical Journal. “But I think every editor would tell you about instances of professional jealousy or even personal dislike between authors and reviewers that can lead to injustice.”