On 29 June 2011 the European Medicines Agency (EMA) collected a group of journal editors to introduce to us to, and get our views on, their fairly recently launched European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP). ENCePP, and its e-register of studies, has been developed in collaboration with leading European researchers in the fields of pharmacoepidemiology and pharmacovigilance to try to improve the quality and capacity of post-drug-authorisation research in the EU.

Why have a new European register of research studies?

Of course this was the first question we all asked. Although there aren’t formal registers for observational studies as there are for clinical trials, many investigators do now register their observational studies prospectively on www.clinicaltrials.gov, and indeed the BMJ now actively supports the prospective registration of study protocols and posting of results on registries for all types of studies.

So if the capacity to register non-randomised studies already exists why should there be a new registry of this kind?

The ENCePP steering groups replied that it has become clear that post-authorisation medicines research needs to be more formally organised in Europe. Doing so would lead to a shift away from a reactive policy in post-authorisation regulation (waiting for a signal of an adverse effect to emerge through reports) to a more proactive approach.

As a regulator the EMA has been asking for more post-authorisation medicines studies to be done and it hopes to facilitate higher quality research through the ENCePP network. Henry Fitt of the ENCePP steering group said that when they began setting ENCePP up they looked at doing it within the European clinical trials registry but in the end decided to set up a separate initiative (probably a wise decision considering the bad press that the EU-CTR has had).  There’s a precedent for this as the FDA has a separate initiative (Sentinel) and Canada has its Drug Safety and Effectiveness Network (D-SEN).

The EMA hopes that ENCePP will encourage protocol-driven post-authorisation medicines studies rather than the sort of hypothesis-generating study that is more usual in this research area. And prospective registration of medicines research may be helpful in reducing duplication of work.

What’s the benefit for researchers in registering a study with ENCePP?

It’s worth saying at this point that registration of a European pharmacoepidemiology study with ENCePP is entirely voluntary and is likely to remain so in the foreseeable future. The BMJ certainly does not require it. The ICJME requirement for registration of studies applies only to randomised trials at present.

There are also two ‘tiers’ of registration of a study with ENCePP. First there’s the so-called ‘ENCePP study’ where investigators are obliged to register the study prospectively, to upload a formal a priori study protocol before beginning the study, to clearly set out the roles of the study funder and the researchers in advance, to adhere to certain methodological standards, to document any changes to the plan for the study, and to publish full results without delay on the ENCePP website. An ‘ENCePP study’ is awarded a mark of quality (the ‘ENCePP study seal’  when the protocol and methodological checklists are uploaded, and must meet various ongoing requirements to maintain this ‘badge of honour.’ However, researchers may register a study that does not meet these requirements with ENCePP, either prospectively or retrospectively, and may share their research question, methods and findings on the ENCePP website. They just don’t get the mark of quality and don’t get to call the study an ‘ENCePP’ study. This voluntary registration is encouraged to increase research transparency in the field and to encourage networking between researchers.

Researchers at the recent meeting said that there are several advantages to registering with ENCePP and being an official ENCePP study centre (apart from their natural desire to do good and useful research). Many researchers in this field have contracts with the pharmaceutical industry and may feel more empowered to insist on adhering to a code of conduct in research if they can say that as an ENCePP centre they are required to register their study and adhere to requirements for methodological conduct and reporting. In addition, responsibilities of funders and investigators have not always been clearly communicated in the past, and since ENCePP requires these to be clearly documented researchers are further empowered to get clarity on roles and responsibilities from the outset.

Researchers hope that if they tell journals to whom they submit their research that the study is registered with ENCePP, editors will recognise what that means in terms of the quality of the study.

Posting of results of studies on the ENCePP website

The researchers and members of the ENCePP steering group present at the meeting seemed to be afraid that ENCePP’s requirement for investigators to post their findings ‘without delay’ on the register would prove a problem when it came to publication of their study in a peer reviewed journal (as study results would already be in the public domain). This puzzled the journal editors present, who professed to hold a liberal view about posting of study results on trial websites. The BMJ’s guidance for authors of research articles clearly states that ‘the BMJ does not consider posting of protocols and results in clinical trial registries to be prior publication’ (see also this editorial).

Virginia Barbour of PLoS Medicine said that it would be ridiculous to try to prevent researchers from posting their findings on a registry before publishing in a peer reviewed journal as this would ‘introduce another level of publication bias.’

Toby  Lasserson of  The Cochrane Library said that publication on register sites ‘facilitates accountability and encourages transparency’ (not least because the journal staff and peer reviewers have the opportunity – if they wish –  to look at the full technical report, protocol etc. when evaluating the submitted study).  Joerg Hasford of Pharmacoepidemiology and Drug Safety pointed out that a journal article is an entirely different animal from a full technical report as it has had the benefit of peer review and is packaged in a way that makes it accessible to a wider readership.

Open for business

ENCePP launched its searchable E-Register of Studies link in November 2010 and to date 12 studies have been registered, 4 of which have the ENCePP quality seal.

There is a plan to make study data available in the future and the ENCePP steering committee is now consulting on the finer details of how to enforce their broad commandment ‘thou shalt share the data’.

How to ensure patients’ anonymity, adhere to governance rules and deal with ownership of data are issues that must be resolved. But they are confident that they can figure out how access to data will be ensured whilst respecting anonymity and commercial interests.

Kirsten Patrick is clinical reviews editor, BMJ.

Competing interests: None declared.