What Doctors Don't Know About the Drugs They Prescribe (TED Talk)

Saturday, April 6, 2013

What Doctors Don't Know About the Drugs They Prescribe (TED Talk)


This TEDTalk is accompanied by an original blog post from the featured speaker. Watch the talk above, read the blog post and tell us your thoughts below.


Doctors need the results of clinical trials to make informed choices, with their patients, about which treatment to use. But the best currently available evidence estimates that half of all clinical trials, for the treatments we use today, have never been published. This problem is the same for industry-sponsored trials and independent academic studies, across all fields of medicine from surgery to oncology, and it represents an enormous hidden hole for everything we do. Doctors can't make informed decisions, when half the evidence is missing.

Most people react to this situation with incredulity, because it's so obviously absurd. How can medics, academics, and legislators have permitted such a huge problem to persist? The answer is simple. This territory has been policed -- and aggressively -- by the pharmaceutical industry. They have worked hard to shut down public discussion on the topic, for several decades, with great success.

They say, for example, that the problem is modest, and that critics have cherrypicked the evidence: but this is a lie. The best evidence comes from the most current review of all the literature, published in 2010. It estimates that half of all completed trials are left published, and that trials with negative results are about twice as likely to be buried.

Then they pretend that the problem is in the past, and that everything has been fixed. But in reality, none of these supposed fixes were subjected to any kind of routine public audit, and all have now been well-documented as failures. What's more, they all shared one simple loophole: they only demanded information about new trials, and this is hopeless. Anything that only gets us the results of studies completing after 2008 does nothing to fix medicine today, because more than 80% of all treatments prescribed this year came to the market more than ten years ago. We need the results of clinical trials from 2007, 2003, 1999, and 1993, to make informed decisions about the medicines we use today. This isn't about catching companies out for past misdemeanors, it's a simple practical matter of making medicine optimally safe and effective.

The arguments go on, with ever more red herrings: industry spokespeople pretend that information about trials -- such as Clinical Study Reports -- can't be released without breaching the confidentiality of individual patient participants. But in reality, the EU Ombudsman has already forced the European Medicines Agency (EMA) to release hundreds of these exact same documents. He stated clearly that the administrative burden of removing any individual patient information is minimal. Next, they claim the cost of sharing trial information is prohibitive: but both the EMA and GSK have committed to releasing all the Clinical Study Reports that they have, and the EMA has already shared millions of pages of documents, quite happily. Sometimes industry people even claim -- in hugely patronizing tones -- that it's better for only regulators to see trial results, behind closed doors, because the public would panic if exposed to dissenting views.

There's more. Sometimes they pretend that the academic journals are the bad guys, for rejecting papers with negative results, when the evidence shows this was barely ever an issue, and, in any case, there are now endless open access journals, specifically designed to accept negative results. Then there are the hole-pickers: people who pay lip service to the problem, with a brief claim that they are "on your side," then expend all their worldly effort trying to pour cold water on the problem, pretending that things aren't so bad after all.

Continue reading here.

Related Posts


This site's content is licensed under a Creative Commons Attribution 3.0 License. | Terms of Service

Do You Like This Post?