Letters Sharing data from clinical trials

Is sharing data from clinical trials always a good idea?

BMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f3379 (Published 28 May 2013) Cite this as: BMJ 2013;346:f3379
  1. Peter D White, professor of psychological medicine1
  1. 1Queen Mary University London, St Bartholomew’s Hospital, London EC1A 7BE, UK
  1. p.d.white{at}qmul.ac.uk

Not many clinicians or scientists would argue with the campaign by “AllTrials” to register and report the full methods and results of clinical trials.1 But is it sensible to go so far as to “encourage authors of all BMJ papers to share their datasets publicly,” so that all may see?2

We routinely reassure participants in clinical trials that their data will be held securely and confidentially. Research ethics committees rightly insist on locked filing cabinets and ensuring that only the researchers have access to digital data. Is this reassurance consistent with public release of patients’ confidential data without their consent? Although only “anonymised” data are proposed for public release, are data truly anonymous when details of age, sex, and perhaps locality are linked to past and current medical details?

And what will potential trial participants of the future think of the reassurance of confidentiality when they know that their “anonymised” data will be publicly available for anyone to access? Will this encourage more patients to take part in trials or will it have the opposite effect?

I prefer the Medical Research Council’s current policy on access to research data. The council considers release only to bona fide researchers, who work for bona fide research organisations, and who sign up to the same standards of respecting the confidentiality of the data as did the original researchers.3


Cite this as: BMJ 2013;346:f3379


  • Competing interests: PDW has received several Freedom of Information requests from members of the public for all the data from a recent trial of non-pharmacological treatments of chronic fatigue syndrome.