I am writing a book (due out in December) entitled "Tabloid Medicine: How the Internet Is Being Used to Hijack Science for Fear and Profit."  The book warns how pseudo-science intended to increase public fear about medical technology  is spread first through tne Web in virus like fashion and then to lazy and biased reporters who run with the "facts", to reinforce the narrative that untold risks are hidden by a conspiracy between industry and the FDA.  The people pushing this story line are furthering their own interests and fattening their bank accounts, at the expense of science-based approaches to improving health.  It leads to unscientific skewing of social and political institutions.

It should be in a loose leaf folder.

The most recent example is the article in today's WSJ: 

By JEANNE WHALEN And ALICIA MUNDY

Medicare patients in the U.S. who took GlaxoSmithKline PLC's diabetes drug Avandia may have suffered as many as 48,000 heart attacks, strokes and other problems between 1999 and 2009 that could have been averted had they taken a different drug, a Food and Drug Administration scientist contends in a new study.

The unpublished study by outspoken Avandia critic David Graham, an FDA drug-safety official, echoes his earlier assertions and another study led by the Cleveland Clinic's Steven Nissen that linked Avandia to additional heart attacks and strokes.

The FDA is scheduled to review the drug's safety in a meeting next month. Dr. Graham and others have argued that Avandia should be withdrawn from the market. An FDA official said Thursday that Dr. Graham's study would be included in safety data to be reviewed for that mid-July meeting.

Why should it be included?  Or rather, why, since it is just another example of assuming the worst based upon questionable data, is it being included at all.

Because the media and pols pressure the FDA to include it, whether it is crap or not. 

Oh, but wait, this isn't just ANY study.  This one was

"first published by the Pharmalot blog, which also posted a May 28 email from Dr. Graham to the FDA's senior leaders. In it, Dr. Graham said he wanted to submit the study to the Journal of the American Medical Association for publication. The Pharmalot blog is a respected blog covering the pharmaceutical industry written by Ed Silverman, an editor at large for the industry publication Med Ad News.

In the email, Dr. Graham complained that senior FDA officials were holding up his efforts to publish his work. However, an FDA official, who confirmed the email, said Thursday that the agency isn't suppressing the study and that it was subsequently submitted to the journal.

A JAMA spokeswoman said the journal doesn't ever confirm or deny receipt or acceptance of any manuscript until publication. Dr. Graham couldn't immediately be reached for comment."

We all know that respected peer-reviewed medical journal Pharmalot, which of course does not accept any advertising. 

(Drugwonks is respected too but if we were part of such a stunt the Tabloid Medicine machine would rise up in anger and call us tools or something,)

In any event, the preceding is just fancy footwork around having to undergo peer-review (which for an article like this should not be a hard sell at JAMA) and to make all sorts of claims without having to answer serious questions.  One of them being, if Nissen showed 43 percent increase risk cherry picking data even as he consults for Takeda, a company making a competitor product, how do you explain the 25 percent figure Graham came up with.  Confidence intervals anyone?  A second one, is this claims data (yes) and was it adjusted for severity of illness (no) and how many people were on multiple medications (many).   Is there a dose effect.  Perhaps and in my next blog I will take this on.   Third, and the one the FDA will seek to address in adult fashion, do the risks of any of these medications (which all drugs carry) outweigh benefits with respect to controlling progression?  The answer is likely no. 

So here's the difference, traditional science is rapidly moving towards the personalization of medicine using biomarkers and real-time clinical data.  In this new world, risks and benefits can be adjusted more precisely. 

Meanwhile the Tabloid Medicine machine -- which includes Nissen, Graham, Mundy, Silverman, Grassley --  see risk everywhere and insist upon a system that seeks to anticipate every risk before it happens.  

As Cass Sunstein has noted, the precautionary principle, with it's standard of not allowing any technology absent evidence of zero harm,  carries it's own risks.  One of them is that people will stop taking medicines that help, including Avandia.  The other is that medical innovation will slow to a crawl and that personalized medicine in particular will be crushed by those who, because of cultural and ideological biases, will spread rumors about risk in a unscientific and unbalance fashion.  More people will be harmed by adopting the slant of this most recent article in the WSJ.