EMA: Damned if it Does, Damned if it Doesn't

The European Medicines Agency thought it might at last be back on the road to salvation when Guido Rasi swept into town in 2011 and started ordering greater transparency in the agency's operations. He was determined to seek a new type of trust with the agency's various publics—and particularly with those who had been most vocal in attacking it for secrecy. As this blogger remarked in April, the agency has ever since been bending over backwards to get transparency right.  Its crowning achievement in transparency—proactive release of clinical trials data—is only weeks from finalisation.

But instead of the road to salvation, the agency finds itself yet again on that road so well-paved with good intentions—the road to hell. It is again under fire—and, paradoxically, for subverting transparency with the very mechanism—pro-active release—that it is planning to introduce to boost transparency.

The criticism has sprung simultaneously from at least two separate points. The first to hit the headlines was the European Ombudsman, with a public statement on May 16 that the agency's new scheme looked as if it was going to undermine transparency. The second—on the same day—came from scientists linked to the Cochrane Institute and the British Medical Journal, and accused the agency of a "U-turn" from earlier promises, and of seeking to impose gatekeeping controls that would slow rather than speed access.

The European ombudsman investigates complaints about maladministration in the institutions of the European Union. The current incumbent, Emily O'Reilly, took over the post last year, and has been energetic in her pursuit of complaints—as well as avid in her endorsement of transparency. She enthusiastically welcomed the recent European Parliament vote backing new clinical trials rules with ambitious provisions on data access, and congratulated the parliament "for having successfully steered this legislation through to a very positive outcome."

In recent weeks, the EMA invited her to a meeting to discuss the way it was finalizing its plans for pro-active access to clinical trial data, but she declined. Instead, the agency sent her its latest draft of how it envisaged its new scheme operating—on May 7. On May 13, she wrote to Rasi outlining concerns that the plans would limit rather than widen access, contrary to its earlier aims.

On May 16 she released her statement, saying that the agency "is planning to limit access to clinical trial data by imposing strict confidentiality requirements and by allowing data only to be seen on screen using an interface provided by EMA, as well as imposing wide restrictions on the use of such data." She said the approach could "undermine the fundamental right of public access to documents established by EU law."

She focused on what she saw as over-restrictive protection of data defined as confidential because of its commercial significance, and conflict with existing EU legislation on access to documents. If the agency "considers that its proposed new policy provides the same level of transparency," O'Reilly said in her letter, "it is obvious that such an assumption would not be well-founded."

Also on May 16, Tom Jefferson, a reviewer at the Cochrane acute respiratory infections group in Rome, Italy, and Peter Doshi, an assistant professor of pharmaceutical health services research at the University of Maryland in Baltimore and associate editor at BMJ, issued a statement claiming that the agency plans represented “a stunning and surprising reversal.”

Academics and learned journals had been invited by the agency for a consultation on the drafts for the scheme's operating details—as had patient and consumer groups, health professionals, and industry over the last few days.

The agency, caught on the back foot by the simultaneous assaults, pointed out that the documents it had shared were drafts, not yet finalized. "Everything in the policy, terms of use and redaction principles, is draft at this stage," an agency spokesman told this blogger on May 19. "The final package will be presented to the Board on June 12."

The agency insists it is not trying to limit citizens' rights to information. Its new policy was intended as a supplement to, and not a replacement for, the rights of access that all EU citizens already enjoy under existing legislation. "Our objective is to facilitate proactive access to clinical trial data," the agency said.  The new EMA policy would go further than the existing rules, since it will offer data proactively, rather than merely in response to requests. And the new system will operate alongside the existing procedure, which anyone will remain free to use.

The irony in much of this episode is that the agency, by its own exercise in transparency, has triggered the flurry of attacks. It provided the draft documents to the ombudsman and it consulted with academia and journals in a bid to avoid criticisms of acting in secrecy. These are now being used as incriminating evidence in the criticisms levelled at it. Damned if it does, damned if it doesn't.