The fine line that drug companies must tread when reporting data about their products is one focus of a new investigation by the UK's Royal Society into how the results of scientific research are made public.

On August 11, the society formally announced the launch of the investigation, which was prompted by a string of controversies about how and when scientists communicate their research results to the public.

A working group under the chairmanship of Patrick Bateson, the society's biological secretary, will produce a guidance document in the fall. It will be sent to anyone receiving funding from the Royal Society and to the fellows. It will also be made available to the wider scientific community.

The group will look at the peer review process, but it will also be paying close attention to the way companies, charities, and other groups make their data public and to the quality control measures they have in place.

The situation companies can find themselves in is particularly interesting, Bateson told The Scientist. This is partly because they are legally obliged to alert the stock market to share-price–sensitive news as soon as possible, something that may be at odds with rigorous peer review. "Unfortunately, there is a real conflict here," he said.

Commenting on the investigation, the director of the Association of the British Pharmaceutical Industry, Trevor Jones, said the issue did not come up very frequently. "Every so often, difficulties can arise for companies who have share-price–sensitive information that needs to be announced, which is also awaiting peer review. Fortunately, this very rarely occurs," he said in a statement to The Scientist. "And when it does, there are a number of official sources that provide guidance about what a company should do."

Another member of the working group, Paul Harvey, head of the University of Oxford's zoology department, added there was another area of concern for drug companies.

"There's a clear problem that if there's a hint—though not statistically significant—that a drug may have a negative side effect, that has to be reported… but if there's a hint that it has a positive effect that is not statistically significant, then you do not include that."

Bateson added: "There's an aspect of it where they're actually reporting stuff which probably is not scientifically valid but they're doing it to cover their backs."

Mike Owen, a senior vice president at British pharmaceutical company GlaxoSmithKline, is a member of the working group. He has been informing his colleagues of the issues in this area, Harvey said.

Another area the group will be looking at closely is the quality control applied to information released by charities, or nongovernmental organizations (NGOs). To clarify the situation, the group has written to charities asking them whether their public documents are subject to peer review or to other forms of filtering.

"There's a danger with an NGO that they can release information to the wider world that has not been peer reviewed," Harvey said. "We're trying to look at groups like that where peer review may not exist or where scientific facts aren't put forward fairly objectively."