A new study in the New England Journal of Medicine (NEJM) found that, despite being required to do so by law, most clinical trials aren’t reported within a year of completion. Under the FDA Amendments Act 2007 (FDAAA), most clinical trials in the US on treatments approved by the FDA have to publicly report their results on the register ClinicalTrials.gov within one year of completion.

Researchers from the Duke Clinical Research Institute identified over 13,000 trials registered on ClinicalTrials.gov that were likely required by the FDAAA to report their results. Only 13.4% of those trials reported results within the first year. Industry sponsored trials were more likely to report results on time than trials sponsored by academic or government sources. Five years after a trial had ended, 41.5% of industry-funded trials, 38.9% of trials funded by the National Institutes of Health (NIH) and 27.7% of academic trials not funded by the NIH had reported results.

Dr Ben Goldacre, author of Bad Pharma and co-founder of the AllTrials campaign:

After the 2007 laws requiring trials to post results within 12 months, it was widely claimed that the problem of withheld trial results had been fixed. This NEJM paper is the second major study to show that those rules have been widely ignored. It highlights the need for enforcement, and accountability – naming and shaming – around withheld trial results. It also demonstrates once again that academia, as well as industry, need to improve performance, and stop withholding information from the people who need it most: doctors, researchers, and patients.

The next steps are clear. The FDA needs to start issuing the $10,000 a day fines that they are allowed to impose, but have never used: that much is simple. But they also have to help the public by providing clear information on bad performers. At the moment there is considerable obfuscation around which trials have been granted exemptions from the requirements to report results, and this is harming outside efforts at increasing accountability. The FDA need to display openly and in public all the applications they have received from anyone asking to withhold their trial results, and list all the exemptions granted, so that no company or researcher can hide in the shadows.

Dr Monique Anderson, lead author on the study:

Patients who participate in clinical research have the expectation that the risk of participation will be offset by the creation of generalizable knowledge and the advancement of science, and that is achieved through the availability of clinical trial results. Sponsors who lead clinical trials have an ethical and legal obligation to publicly report their findings, whether the results are positive or negative.

The US Government is currently consulting on two proposals that would clarify and expand requirements for which trials need to be registered and report results. It would also mean that every NIH funded clinical trial would have to be registered and report its results. They are welcoming comments from people around the world. Read more and respond before Monday, 23rd March 2015.