Open Access News

News from the open access movement


Tuesday, May 23, 2006

Journal editors should help compel access to trial data

Richard Horton, Trial registers: protecting patients, advancing trust, The Lancet, May 20, 2006 (accessible only to subscribers). An editorial. I don't have access and quote from an excerpt posted by William Walsh:
This flourishing of clinical research for the benefit of patients is the overarching objective of WHO's effort to gain agreement on a practicable way of registering all clinical trials, from the moment of their inception and for a minimum set of scientifically and ethically essential elements. In today's Lancet, Ida Sim and her colleagues at WHO set out two principles that must underpin trial registration. First, that all interventional trials, including early-phase studies...should be registered. And second, that all elements of the 20-item minimum dataset must be disclosed at the time of registration...Journal editors can help. Just as the editors of general medical journals played a part in raising the profile of trial registration, so editors could again help to move the current debate beyond merely an aspiration and towards a necessity. Results disclosure will be the next major step on the road to full transparency of all relevant information about a particular trial. Editors could drive that disclosure process by insisting that trialists and sponsors deposit key results information into a publicly accessible database, akin to GenBank. But...some editors are operating policies that inhibit rapid disclosure of trial data. Editors have been tough on pharmaceutical companies resistant to registering their trials. They have urged companies, such as GlaxoSmithKline, to put public interest before commercial interest. Yet journals now find themselves in a similar position to industry, but with respect to results disclosure. They --we-- have a self-interested motivation to delay full disclosure of results until publication of a final paper --the Ingelfinger rule. In the past, editors have sought to control access to research results by insisting on a journal's priority in releasing data in advance of any other venue. Happily, that rule has broken down in the face of multiple outlets for data --in particular, presentation at scientific conferences. Given the ever-widening flow of information, it would be only a small step to recognise posting of trials data on an independent results database as an ethical imperative.