Geoffrey Boulton and colleagues' Comment (May 14, p 1633)1 lays out the case for sharing scientific data, and advises that the UK's Royal Society has established a working group to examine the implementation issues and to make recommendations.
However, completion of an open access project on an international scale could take years, perhaps a decade or more, even with funding in hand and a consensus blueprint to guide the effort. Meaningful steps can be taken now. Peer-review journals such as The Lancet can advance the goals of transparency and public confidence in the findings of scientists by adopting a data availability policy that requires authors of empirical papers, as a condition of publication, to submit electronic copies of their data, protocols, computer code, and relevant documentation to the journal's online archive of published papers.
There are successful precedents. In 2005, the American Economic Review imposed such a mandate on authors and found, as one might expect, that it improved the accuracy of research results.2 The journals Nature and Science have similar requirements to ensure the availability of data and code to its readers. The time has come for medical journals to open the shutters on the scientific process. The Lancet should take the lead.
I declare that I have no conflicts of interest.