select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to drug trial protocol redactions by industry sponsors

Researchers describe the redactions in protocols for industry-sponsored randomised drug trials and evaluate whether there was a legitimate rationale for the redactions, published in the Journal of the Royal Society of Medicine.


Dr Gillies O’Bryan-Tear, Chair of the Policy and Communications Group, Faculty of Pharmaceutical Medicine, said:

“From the outset, in the title, it’s clear what the authors have set out to prove in this review of industry sponsored protocols: that the industry conceals its study designs and in so doing, is harming patients. Nothing could be further from the truth. It is surprising that there still exists in academia the view that the pharma industry is trying to hide important safety and efficacy data from public scrutiny. Nowhere in the paper is the reader reminded that all protocols and analyses have to be submitted to the regulatory authorities, by law, where they are subjected to the most rigorous review. It is simply not in the interests of companies to conceal important data, as the consequences are so profound; in any case, pharma companies and the medical and clinical staff within them share the same objectives as healthcare workers everywhere: to improve patient suffering and to bring safe and effective new treatments into medical use.

“Turning to particulars: the analysis is a subjective review of redactions to pre-defined sections of a number of protocols obtained under Freedom of Information from several companies and from some academic groups (not specified, though the companies were). The review focusses on the sponsored trials since there were few redactions noted in the non-sponsored trials. However, there was no analysis of those redactions in the non-sponsored trials. There was clearly some reluctance of some of the companies to share either the protocols, or details within them. I think this may be because, by the authors’ admission, this was the first time such a study had been done. Companies have for years agreed to publish all of their studies, negative or positive; and they already put summary information of all their protocols on government web sites. However they are clearly not used to requests for protocols, and this was demonstrated by inconsistencies in what the companies provided and what they redacted. It’s not hard to understand why a company may not share detailed information about assumptions behind sample sizes – such calculations assume certain differences of treatment effect which companies may not wish to share – but it’s harder to understand why some companies redacted details of how adverse events were to be collected, since this is absolutely standard across all industry sponsored trials and indeed the methodology is standardised and mandated by the regulatory authorities. I can only assume companies were simply unsure how much to reveal and erred on the side of caution. Similarly, it’s hard to understand why companies would not release all the inclusion and exclusion criteria since these are often placed on the government web sites such as

“I can understand why companies may not wish to reveal the details of data ownership and publication policies, since these are commercially sensitive areas and private to the investigators as well as the sponsor.

Unfortunately the paper indulges in some well-worn shibboleths about industry trials: “There are no good reasons for redacting the definition and analysis of the primary outcomes, but there are unsound ones like hiding scientific misconduct”. Are the authors accusing some of these companies of scientific misconduct? If so, they should say so and report the companies, citing evidence.

“It is sad to see papers like this published, seeking to reduce the credibility of industry sponsored trials, which are subject to more regulatory scrutiny than the documents of most other industries and certainly more so than academic studies. Academia and industry should be partners in helping to alleviate the suffering of patients, and this paper does not further that noble cause.”


Prof. Tony Fox, Professor of Pharmaceutical Medicine, King’s College London, said:

“Redactions have legitimacy when:

  1. a) It is necessary prior to patent issuance (patents are initial full publications): usually molecular structures
  2. b) Where it would cause identifiable patient information to be published (this can include extremes of age, people with very rare diseases identified by their diagnosis, etc.)
  3. c)  Where it violates some other restriction, e.g. could be perceived as promotion of an unapproved drug or an old drug for an unapproved indication
  4. d) Certain (rare) types of research with national security-related issues.

“As an expert member of the London SE Research Ethics Committee (NHS Health research Authority) I have never been presented with a protocol containing redactions of the type described in this paper.  I would agree that no ethical assessment could be made under those conditions.  One wonders how these protocols could achieve approval by a competent ethics committee in Denmark when there is so much missing information that thwarts assessments of equipoise, beneficence and autonomy.

“I should also point out that the UK has a reputation for transparency of clinical trial protocols. Glaxo (now GSK) actually began publishing long before (etc) See:  Sykes R Being a modern pharmaceutical company BMJ 1998; 317: 1172.  As for law, the EMA does have a guidance and a policy on what can be legitimately redacted.”


* ‘Redactions in protocols for drug trials: what industry sponsors concealed’ by Marquardsen, et al. will be published in Journal of the Royal Society of Medicine on Friday 26th January.


Declared interests

Dr Gillies O’Bryan-Tear: “I work part-time in the pharmaceutical industry.”

None received.

in this section

filter RoundUps by year

search by tag