• By Laura Whybra
  • Posted on Friday 19th February, 2016

Muddy waters: Are we making medical decisions with biased evidence?

The results of many clinical trials never see the light of day. As the World Health Organization calls for all trials to be released into the public domain, physician and writer Ben Goldacre argues for practical steps to speed up change. But why should we care if some trials don’t publish their results? In How to Get All Trials Reported: Audit, Better Data and Individual Accountability, Goldacre writes about the deficiencies in open and honest clinical trial reporting and suggests ways that this critical issue can be addressed. His article was prompted by a “landmark position statement” from the World Health Organisation (WHO) in the same issue of PLoS Medicine. The WHO statement demands full publication and disclosure of clinical trial results, and calls such disclosure a “scientific, ethical and moral responsibility.”

Why care?

But why should we care if some results from clinical trials don’t get published? If researchers test something and it doesn’t work, then surely we don’t need to know about it: we should save the journal space for reporting the next breakthrough drug for cancer. Right? But we should care. A lot. Here is a scenario. Imagine you have a disease, or a problem; let’s say, anxiety. You go to the doctor and she prescribes a course of medication or enrolls you in a program. The doctor thinks this medicine or this intervention will help because she has done her research: diligently plowing through medical journals in a bid to stay ahead of the curve, to keep up to date with the latest developments in the field. All the papers published about this treatment say that it works, and so she prescribes it. Except that it doesn’t work. And all those studies that conclude that it doesn’t work, or that it’s no better than some other treatment you might have been given, are sitting in a filing cabinet, read by no one, and informing no best practice decisions. Doctors – through no fault of their own – may now be prescribing treatments on the basis of faulty information. And this affects all of us. That’s why we should care about it.

Transparency is crucial

Making the results of all clinical trials available – not just the results that turn out the way the researchers hoped – is crucial to building a balanced picture of how medical treatments work. In his article, Goldacre suggests that transparency can be achieved and offers three solutions to the problem. First, all clinical trials should be registered and reported using a standard database. This is already a rule in principle, yet is not always followed. A standard format for registration and reporting makes it impossible to switch pre-specified primary outcomes, for instance, in order to “spin” the results in a more positive light. Second, auditing is key. If trials are registered then they can easily be tracked. An audit of trial registers can highlight where authors have failed to publish any results after a reasonable period of time. Goldacre suggests that audit data can be built into the research system to push reluctant researchers to make their data public. For instance, institutional review boards could “ask one simple question of all those applying to conduct a trial: ‘Have you been involved in any clinical trial, which completed more than 12 months ago, for which the results remain inaccessible?’” Third, the reporting of historic trials is also important. Today’s clinical decisions are mostly based on products that were trialed and brought to market decades ago, not those that were licensed last month. This wealth of valuable information, often unpublished and hidden from view, should not be allowed to go to waste. If these practices were in place, clinicians and patients would be able to have greater faith in the reports they read and the medical decisions they make. While the recent WHO statement carries a lot of weight and is an important step, it is meaningless without practical action to support it. Goldacre and colleagues have begun that action by instigating the AllTrials.net initiative to publicize the issue. Their tagline, “All Trials Registered, All Results Reported,” makes it abundantly clear what their objectives are. Transparency in clinical trials reporting is a critical, overlooked, and underappreciated issue. Without it, we risk wasting vital resources duplicating something that someone else has already shown to be ineffective. Worse still, we risk causing harm to people. Complete trial reporting helps to put the “evidence” into evidence-based practice.


References

Goldacre, B. (2015). How to Get All Trials Reported: Audit, Better Data, and Individual Accountability. PLoS Med 12(4). e1001821. doi:10.1371/journal.pmed.1001821 Moorthy, VS., Karam, G., Vannice, KS., & Kieny, M-P. (2015). Rationale for WHO's New Position Calling for Prompt Reporting and Public Disclosure of Interventional Clinical Trial Results. PLoS Med 12(4): e1001819. doi: 10.1371/journal.pmed.1001819

Back to Archives