Trish Groves: Data sharing: where are we?

The movement towards open science is gathering pace, driven by scientific and ethical imperatives—not simply by the technological possibilities. In medicine such openness has real potential to benefit patients and society. Indeed, the arguments for open registration and publication of studies’ protocols, methods, and main results are underpinned by ethical arguments that are focused on patients and the public. The Ottawa Statement on trial registration is a prime example.

But truly open science means more than access to articles and to summaries of study designs and results, even when such open access comes with a licence that allows text mining and other forms of unrestricted reuse. Really meaningful reuse and further analysis of clinical studies depend on access to the anonymised patient-level data, along with the statistical code, details of any missing data, and other supporting information. This doesn’t necessarily have to be public access. There are good arguments in favour of either access with registration (for instance of personal details, proposals for further data analysis and reuse, and commitments to publish and share the new data), and for access on request (via an independent panel that considers proposals).

From a standing start, the race to share from clinical studies has really taken off. But, for those who’ve lost track, here’s the story so far—or at least the headlines, from a BMJ viewpoint:

Up to and including 2007









Coming up in 2013


  • European Parliament: first reading and vote on proposed revision and amendments to Clinical Trials Directive/Regulation.


  • EMA due to put its proposal on clinical trial data sharing out for consultation.


  • European Parliament: second reading on proposed revision and amendments to Clinical Trials Directive/Regulation.


  • EMA due to finalise data sharing policy.