along the way…

Posted on Friday 28 November 2014

British Medical Journal
by Tom Jeffeson and Peter Doshi
27 Nov 2014
… On 24 November 2014, the European Medicines Agency [EMA] released a new “Guide on access to unpublished documents.” The guide follows in the steps of several other policy documents, charting the revolution from a closed shop to what is probably the most liberal experiment in regulatory data sharing on the planet.

The six page guide is written clearly, as you would expect from a document for “anyone” interested. The guide tells you how to apply for documents held by the EMA as part of the process of central pharmaceutical regulation. We learn that English is almost certainly the language of the released documents, but you can apply in any language of the European Union. Other sections tell you what will happen to your request, and what options you have in the event that your request is turned down or not answered in time.

Readers of the guide are warned that the release of large and complex documents may take place in batches over a long time. This certainly has been our experience and presents a serious problem for independent researchers who work on deadlines. We applied for clinical study reports from trials of a global public health intervention, and we’ve yet to see more than 10% of the text six months after the ball got rolling. The EMA tells us that the documents are in preparation, but so far there’s little meat.

Gone, it seems, are the good old days of data request. On 10 January 2011, we requested around 20 clinical study reports on Tamiflu. By the end of May, the EMA had sent us 25,000 pages of unredacted text. One presumes that the EMA’s workload has exploded since those early days. But is this the case? We await increased transparency on the to-ing and fro-ing with the marketing authorisation holders, to understand more precisely what happens as one waits. Where are the delays occurring? How can the system become more efficient? As we await answers to these questions, we think it’s time to address an equally serious problem: the lack of a menu.

While we applaud the EMA’s efforts to provide a guide that makes requesting documents easier, we are concerned that the guide does not tell us what’s on the menu at the EMA restaurant. For hungry people this is a bit of a problem, but even more so for the restaurateur…

So let’s have a list of holdings by marketing authorisation application, with the dates and types of documents held included. And while you are at it, please explain what’s in each document in plain language so that “everyone” can order the right dish.
Tom Jefferson and Peter Doshi shepherded the groundbreaking quest for the raw data on Tamiflu that ultimately lead to the Cochrane Collaboration meta-analysis the concluion that the multibillion dollar government stockpiling for an epidemic was ill advised. It has become the rallying point in the movement[s] clamoring for Data Transparency in medical clinical trials. In this blog post, they’re focusing on something the rest of us have no way of knowing much about – the process of actually getting the data in hand to reanalyze and in particular, what data is available.

It’s easy to see  conceptually why the consensus has settled on having the raw data as a powerful way to put a much needed damper on the massive over-prescription and over-use of medication in medical practice. But it rests on the availability of independent scientists to do the analyses – probably gratis. And that, in turn, rests on how easy this data is to come by and work with. These clinical study submissions each have thousands of pages – challenging in their own right. But in this blog, Tom and Peter are talking about something that comes before anything, knowing what studies are available – apparently no easy task. Only then can one begin the application process [which this EMA guide apparently clarifies].

And it’s not just an issue with the EMA. The various pharmaceutical companies are creating their own unique versions of the conditions and modes of access that might be available for future independent analyses. So, as exciting as the recent advances have been in the fight for public access to the raw data from clinical trials, as is always the case – the devil is in the details. And in the best of circumstances, the process will be hard with long wait times. Even if the NIH/HHS’s recent pledge to make sure that the requirements for posting the results of Clinical Trials on supplies the advertised result, the trial sponsors still have a year in which to post the data, and even then, it isn’t exactly "raw."

A couple of years, ago, Ben Goldacre famously ended his TED Talk on this topic by saying,
"I think that sunlight is the best disinfectant. All of these things are happening in plain sight, and they’re all protected by a force field of tediousness. And I think that with all of the problems in science, one of the best things we can do is to lift up the lid, finger around with the mechanics, and peer in."
No doubt about that,  but Jefferson and Doshi have gone through the actual process of getting through that "force field of tediousness" and are taking Goldacre’s comments a step further by pointing to just one of the many difficult elements along the way.
    Bernard Carroll
    November 28, 2014 | 1:59 PM

    It is just not in Pharma’s DNA to value transparency, so we should not be surprised that they have to be dragged kicking and screaming to any agreement or that they will try all means to sabotage any such agreements. Viewed from 30,000 feet, the tragedy is that a noble enterprise of drug development was hijacked by generic managers and marketeers who had never internalized the values of medical ethics. As many have observed, we should not expect things to change until executives do jail time.

    In the case of the new HHS-NIH announcements on transparency and reporting, we are waiting to see how credible the enforcement process turns out to be. And even if it is credible, Pharma will still find ways to game the system. For instance, the authority of HHS and NIH does not extend to all countries, so there is nothing to stop Pharma from running Phase II or Phase III trials or off-label trials in, say, Serbia or India, then choosing to bury the results if they don’t like them. And if they do like the results there is nothing to stop them publishing those results and publicizing those publications.

    November 29, 2014 | 10:10 AM

    Written for the “drive-by” reader, this “In These Times” articles might indicate that the word is indeed getting out. Let’s hope that as a few good doctors/writers preach to the choir that the choir continues to grow.


Sorry, the comment form is closed at this time.