“EU agency lifts lid on drug data secrets,” the newswires blared on the otherwise calm summer Sunday afternoon of July 15, 2012.
The usual full-time critics of the pharmaceutical industry, in alliance with the pirates and privateers who abhor the idea of anything digital not being available at a mouseclick, have long demanded it. Now the European Medicines Agency (EMA) did not only cave in to these demands earlier than expected, but is even spearheading them. Apparently the decision to make public all clinical trial data submitted to the EMA has been made; the conference announced for November is only intended to fine-tune the release modalities.
This move traces back to the victory of the Nordic Cochrane Centre in a complaint against the EMA to the European Ombudsman (Case 2560/2007/BEH) who had stated on November 24, 2010 that EMA’s refusal to grant access to specific clinical study reports and protocols constituted an “instance of maladministration.” In response, EMA has decided to go far beyond the scope of this verdict. As a matter of routine, they will disclose not only study reports and protocols, but patient-level clinical raw data.
This amounts to a straight adoption of the Cochrane Collaboration’s mantra, “Free Access to all Data from all Clinical Trials” by the European drug regulatory authority. Its difficult to see this as anything but an all-out victory of the self-appointed watchdog organization which exists to conduct meta-analyses that almost invariably dismiss published drug efficacy data as insufficient or irrelevant. Equally invariably, Cochrane analyses are extensively cited by public healthcare providers in their decisions to refuse reimbursement for new drugs.
One might ask, what should be wrong with opening up anonymized clinical trial databases to public investigation? – Nothing at all really, if we were living in an ideal world where opening these data vaults would just mean that more medically and statistically educated minds, imbued with the desire to make objective conclusions even more scientifically valid, are brought to bear on these data.
But we do not have this kind of society.
In his comment on the EMA’s policy change Peter C. Gotzsche, director of the Nordic Cochrane Centre, made no attempt to conceal what his organization intends to do with these data:“The pharmaceutical industry routinely exaggerates the benefits of its drugs and plays down the harms or hides them … We may discover that many drugs are less effective than we thought they were and also more harmful.”
The matter goes far beyond Cochrane’s anti-industry bias. It is all very well hearing EMA’s Senior Medical Officer, Hans-Georg Eichler, assure that “public health scares must be avoided” and that “what the debate now is about is finding the right conditions to minimise the potential for false findings.” These are empty words. Once the data are accessible to everybody (and they will be, because inevitably they will be leaked, hacked, or forced out by court verdicts) simply everybody can interpret them at will — regardless of qualification or bias. The media will be extremely happy to publish these “insights” for their scary and scandalous “news value.”
Care to imagine for a moment what sort of year-round field day drug damage lawyers and anti-vaccination campaigners will soon have? — Oh, to be fair: Mr. Gotzsche did mention this risk in his highly accessed position paper (Trials2011; 12:249; PDF) — in a single short sentence, immediately dismissing it in the next one: “Societies that have only one official version of the truth are not societies we would like to live in.”
Please correct me if I am missing something here; but I was educated to believe that “versions of truth” exist only in religion, philosophy, and politics. Old-fashioned as I am, I prefer to hold that in the exact sciences (and that’s what evidence-based medicine is all about), truth is defined by objective and quantifiable facts, not through diversity of personal opinion.
Then remember — we are talking patient-level raw data here. Stripped of all personal information, sure. And nobody will ever, under any circumstance, be able to link anything back to individual study participants, right? – And pigs can fly.
As the European Medicines Agency prepares sponsors of drug approval studies to accept that they will no longer have control over the submission data that they have dearly paid to generate, some more questions are in order. Such as, What sort of clinical data will be submitted under these conditions? Will there be more data filtering and massaging on the corporate level? What sort of intangible and unprovable pressure will be exerted on corporate scientists so that certain data are not recorded, or analyzed, perhaps not even generated? – Little doubt for me what the response of the Gotzsches of this world would be: “This only proves the need to police pharmaceutical companies.”
Sometimes I get the impression that Europe never tires of contemplating new modalities of pushing itself into scientific and economic oblivion.