It's time for AllTrials registered and reported
- By: Tracey Brown
- On: April 30, 2013, 12:53
The Cochrane Collaboration has become a lead partner in the AllTrials initiative (AllTrials.net), a campaign to ensure that all clinical trials are registered and the results reported, for all treatments in current use. Cochrane joins Sense About Science, Bad Science, the BMJ, the James Lind Initiative, and the Centre for Evidence Based Medicine in leading this international campaign, which has already signed up over 200 research bodies, regulators, and patient groups, and nearly 50,000 individuals. This, though, is just a beginning.
Everyone involved in medical research knows that we only see parts of the picture of how treatments work and in whom. There are 'known unknowns': studies not yet conducted and comparisons not yet made, for example. We can press for the funds and motivation to do them. There are 'unknown unknowns': limits to our current understanding that we can only guess at, such as the extent to which epigenetics might sweep away working assumptions about how we respond to drug treatments. But medical research is suffering from something far less explicable: 'unknown knowns'. Trials are run and data are gathered, but what many of them found is kept a secret. Researchers, doctors, and patients cannot benefit from knowledge that they are unable to obtain.
Around half of all the clinical trials that have been conducted have not yet been published, and trials with positive results are twice as likely to be published as others. The problem of failure to report the outcomes of trials exists for industry and non-industry trials, internationally, and at all stages of drug development. A cross-sectional analysis of trials registered and completed on the FDA-run registry ClinicalTrials.gov between 1999 and 2007 found that 56% of industry-sponsored trials and 40% of non-industry, non-government-sponsored trials had been published.
Compared to all the other challenges of medicine, the missing information about clinical trials is a routine barrier to knowledge, and its solution is a simple matter of communication about what is known. It is strange that we have lived so long with a problem that can be fixed relatively easily. It is unfair to patients and trial participants, it is frustrating to researchers and medical practitioners, and it is completely unnecessary.
One of the reasons that the problem of missing trials has not been addressed is that discussions have taken place behind closed doors, amid promises that "it is being fixed". Behind those closed doors, arguments to slow and complicate the path to improvement can thrive. These arguments have been redeployed against the AllTrials initiative. The problem is fixed. It shouldn't be discussed in public because it undermines trust in medicine. Publishing research results causes scare stories. The regulator has what it needs, and we should be content with that. Patients are alarmed that their information might be shared. It wastes resources. The problem is being exaggerated. And so on.
These are not, though, the arguments of people with greater insight into the problems of clinical trial reporting. They are the arguments of people who don't want change. The problem is not fixed. We have seen a slow improvement in registration compliance and reporting rates, but the medicines that are currently prescribed for patients already have marketing authorisation. This is why AllTrials is calling for retrospective publication of trials relating to treatments in current use. Talking only about the conduct of future trials is just kicking the ball further up the street, which is what has happened since Iain Chalmers first sounded the alert on the problem of non-reporting 20 years ago. Registering protocols and reporting outcomes do not present insurmountable patient confidentiality issues. GlaxoSmithKline, the first pharmaceutical company to sign up to AllTrials, has agreed to publish the results of all trials going back to its formation as a company, which shows it can be done. Under the glare of the public spotlight now on this issue, the arguments against communicating trial results disappear into the shadows.
AllTrials is an opportunity for us to come together to ensure that this public spotlight continues to shine on the dark corners of missing trials until there is change. There are no new arguments here. Patients have expected and continue to expect their treatments to be based on the best available evidence. Clinical trial participants expect to be contributing to knowledge about their disease and future treatments. Policy-makers expect to be able to make decisions about effective and efficient health care. We all expect researchers to know what is known and to be free to analyse it. All of this underlines the valuable and necessary role of The Cochrane Collaboration in this essential campaign. Please sign up and urge other individuals and organisations to do the same at AllTrials.net.
Managing Director, Sense About Science, London, UK. firstname.lastname@example.org
How to cite: Brown T. It's time for AllTrials registered and reported [editorial]. Cochrane Database of Systematic Reviews 2013 Apr 30;4:ED000057. dx.doi.org/10.1002/14651858.ED000057
1. Song F, Parekh S, Hooper L, Loke YK, Ryder J, Sutton AJ, et al. Dissemination and publication of research findings: an updated review of related biases. Health Technology Assessment 2010;14(8):1-220. www.hta.ac.uk/1627
2. Ross JS, Mulvey GR, Hines EM, Nissen SE, Krumholz HM. Trial publication after registration in ClinicalTrials.Gov: a cross-sectional analysis. PLoS Medicine 2009:6:e1000144. dx.doi.org/10.1371/journal.pmed.1000144
Competing interests: The author has completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available upon request) and declares (1) no receipt of payment or support in kind for any aspect of the article; (2) no financial relationships with any entities that have an interest related to the submitted work; and (3) no other relationships or activities that could be perceived as having influenced, or giving the appearance of potentially influencing, what was written in the submitted work.
Image credit: AllTrials
Contact the Editor in Chief, Dr David Tovey (email@example.com): Feedback on this editorial and proposals for future editorials are welcome.