NEWS

Europe’s academics fail to report results for 90% of clinical trials

But nearly 70% of industry-sponsored trials report their results within a year of ending.

Search for this author in:

Results of clinical trials often go unreported for longer than mandated by the European Union.Credit: Eddie Keogh/Reuters

European academics are failing abysmally when it comes to reporting the results of clinical trials, a study has revealed.

An analysis of data from the European Union’s Clinical Trial Register — published in The BMJ1 on 13 September — shows that around 50% of the listed trials have not complied with guidelines that say results must be reported within 12 months.

The researchers who conducted the analysis found that only 11% of trials run by academic centres — such as those led by universities, governments, hospitals or charities — had published outcomes after completion (see ‘Failing to comply’').

Source: BMJ 2018;362:k3218 http://dx.doi.org/10.1136/bmj.k3218

Furthermore, only 11 of the major sponsors of clinical trials — entities that are responsible for at least 50 trials on the register — had reported 100% of results, all of which were companies.

This compares to a total of 32 major sponsors that had not reported any results from their trials. All of these are academic institutions, rather than companies.

Ben Goldacre, head of the Evidence-Based Medicine DataLab at the University of Oxford, UK, and lead author of the study, says this is a serious concern for science. But he thinks that “chaos, rather than malice” is behind academics’ poor reporting of trial results.

Companies have a stronger culture of complying with such guidelines, he says, so publication of results might more often be integrated in day-to-day operations.

“For academics, the reason for not reporting results may be a matter of poor knowledge management or staff turnover, where a clinical academic has gone back to work as a doctor,” Goldacre says. “But equally, some people may have an ideological commitment to a particular treatment, which may lead them to, either consciously or unconsciously, slow down the dissemination of findings contradicting that treatment.”

Delayed directive

The European Union’s Clinical Trials Directive, which regulates clinical trials, came into effect in 2004, and guidelines that set publication deadlines for trial results were adopted in 2012. But the deadline for compliance with these guidelines came into force only in December 2016.

The European Medicines Agency (EMA) in London, which hosts the trials tregister, is responsible for overseeing compliance with the rules. Under the directive, the EMA is required to tell trial sponsors whose reports are overdue on the system and urge them to provide this information.

But the researchers found no evidence that this has happened, and note in their paper that no one has ever been sanctioned for flouting the guidelines.

“We need to reflect on how we can improve our communication with academic sponsors and smaller sponsor organisations,” says Fergus Sweeney, head of inspections at the EMA. “This study is a very good example of how transparency can support public scrutiny and enable the development of tools to support that.”

Some researchers, the study notes, might hold back information because they are aiming to publish in high-profile journals, which can take longer than the 12-month deadline set by the guidelines.

But Elizabeth Loder, head of research at The BMJ, says it’s not necessary to wait for a full publication to disseminate information.

“Results can also be reported on trial registration sites or in preprints,” she says. “The important thing is that the information from studies, which rely on the goodwill of volunteers, should be available in a timely manner to help doctors and patients make decisions.”

To raise awareness of the problem, the AllTrials campaign, led by Goldacre, has launched a tool called EUTrialsTracker, which ranks clinical-trial sponsors by how many results they have reported.

doi: 10.1038/d41586-018-06676-8
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up

References

  1. 1.

    Goldacre, B. et al. Br. Med. J. 362, k3218 (2018).

Download references