Many people in science would rather not talk about the problem of research misconduct, much less act on it. After all, who directly involved would benefit from a serious crackdown? Certainly not the institutions at which the misconduct takes place — they are nominally responsible, but can face legal repercussions, embarrassing headlines and a public-relations disaster if they expose cheating academics. It is much easier to shuffle miscreants out of the side door with vague references and a promise of silence, effectively pushing the problem somewhere else, and onto someone else.

So it is perhaps a sort of progress that the British Medical Journal and the international Committee on Publication Ethics were able to organize a meeting on the subject in London last week, gathering representatives from universities, funders, journals and lobby groups to discuss how the problem could be tackled in the United Kingdom (see Nature http://doi.org/hmx; 2012). The meeting broke little new ground, but its organizers do, at least, deserve credit for trying.

Stronger action and punishments are needed to discourage misbehaviour.

A big part of the problem is the lack of perceived risk associated with misconduct. Some fraudulent researchers might be sociopaths who don't care about the rules, but many others simply believe that they can anticipate the outcome of a research project, and see no downside to fabricating the required results to save time, or tweaking results to achieve a stronger signal. Either way, stronger action and punishments are needed to discourage such misbehaviour. (Meanwhile, for colleagues considering blowing the whistle, the risks are glaringly huge — witness the plight of scientists, such as cardiologist Peter Wilmshurst, who have raised questions and have faced the full force of Britain's ludicrous libel laws as a result.)

Could publications such as this one do more to deter cheats? Unfortunately, we are often in no position to flag up even proven cases of misconduct, and thereby highlight the risks that miscreants run with their careers. Yes, it is a journal's primary job to clean up the literature, but when papers are retracted owing to misconduct, the libel laws (again) often prevent our editors from saying so. We know that this leaves the affected communities frustrated and in the dark. It leaves us frustrated, too.

So, with journals unable to push towards greater integrity and universities often unwilling to do so, should funding agencies be leading the charge? It is, after all, their money that is wasted if misconduct does occur.

Funding agencies in the United States do sometimes investigate misconduct. Research funded by the National Institutes of Health and some other government agencies falls under the remit of the Office for Research Integrity (ORI), which has the power to bar researchers from receiving future funding. However, as Nicholas Steneck, director of the research-ethics programme at the Michigan Institute for Clinical and Health Research in Ann Arbor, told the London meeting, this process probably misses most major misconduct. And the ORI can't initiate investigations: institutions must conduct their own inquiries first.

In the United Kingdom, there seems to be little appetite for launching an overarching ORI-type regulator. Certainly, the existing independent advisory group, the UK Research Integrity Office in Falmer, is clear that it has no desire to take on such a role. British funding councils — in collaboration with the country's universities — have chosen instead to produce a 'concordat' detailing good practice, to which institutions will be expected to sign up. This is laudable, but unlikely to strike fear into fraudsters and fabricators.

So, how can Britain highlight cases of misconduct and discourage it in future? Ultimately, the incentives probably need to come from on high, and the government could get the ball rolling by commissioning an anonymous survey on misconduct that UK researchers have witnessed and perpetrated. An official audit would offer a strong platform for others to build on — perhaps with a parliamentary inquiry and subsequent report on the damage done to UK science by misconduct, and an assessment of the options for tackling it and the investment needed. Funders and universities could then work together to establish common definitions of what counts as misconduct, and how it will be punished. And if a reform of the libel laws goes ahead, journals and other scientists would be able to do more to highlight and expose miscreants.

Sounds ambitious? If the solutions were easy, there wouldn't be a problem to discuss. But there is, so we must face it.