< Private Funding, Medical Journals, and Bias


Friday, December 07, 2012

BOB GARFIELD:  Last year the National Institute of Health spent $31 billion on drug research. The drug companies themselves spent $39 billion. More and more, drug companies are sponsoring their own studies, studies eventually published in reputable medical journals. The Washington Post's Peter Whoriskey, who wrote about pharma-funded research, says that at best it increases the potential for bias and at worst can lead to outright deception. Take the case of the diabetes drug Avandia, manufactured by GlaxoSmithKline. In 2006, the New England Journal of Medicine published a study funded by GlaxoSmithKline that reviewed the drug favorably but, Whoriskey notes, the drug company knew better.

PETER WHORISKEY:  They had seen some of the studies internally that showed potential heart problems for people, and they hired some prominent academics, and they found that Avandia was, lo and behold, very effective and caused no more heart problems than some of the leading competitors. It turns out that was wrong; there were reasons for why the study didn't get to the truth, really, and that's why I decided to pick up the Avandia story.

BOB GARFIELD:  GlaxoSmithKline engaged friendly scientists to design an experiment to underplay the cardiac side effects?

PETER WHORISKEY:  That’s sort of the magic of all this stuff, is what you test. If you give people the drug who have relatively strong hearts, compared to the people who probably will be taking the drug, elderly diabetic people, you’re gonna not see the sign of heart attacks as much. And that’s what happened here, one of the things that happened here is that the signal, as they call it, was hidden. And then, in 2006, after a doctor at the Cleveland clinic put all the data he could find out there on the Internet, it showed that there was a heart attack signal – the sales started to tank. Finally, 2010, the FDA put big restrictions on it and it was essentially banned in Europe. It’s gone away, for the most part.

BOB GARFIELD:  You came to discover in reporting this piece that there were a whole mess of fairly significant conflicts of interest, but not just the sponsors of the study, GlaxoSmithKline but, the authors. Can you list some of those conflicts for me?

PETER WHORISKEY:  There were 11 authors, I think, of this article that we’re talking about. Four of them were employees and the others all had taken money from Glaxo for one reason or another. Either they were consultants or speakers or had conducted research for them in the past.

BOB GARFIELD:  You took pains in your piece to point out that just because research is funded by the industry, it doesn't mean that it is necessarily fudged or manipulated or wrong.

PETER WHORISKEY:  Some of the most fervent critics of these studies will say the big problem is that they do a heck of a lot of really good research, and they do some terrible things. It’s impossible to tell the difference. So the upshot of this was actually another study published in the New England Journal Medicine, which showed that doctors who, you know, you rely on the New England Journal of Medicine for the latest in research are much less likely to prescribe a drug that they read about in an industry-funded trial. So it holds back all of medicine.

BOB GARFIELD:  Why isn't the government underwriting all these studies?

PETER WHORISKEY:  The government funding has steadily gone up, it’s just that the pharmaceutical industry funding has gone up much, much faster. So you end up, if you’re a professor somewhere and you want to do research, there's only one game in town; you have to go to the pharmaceutical companies. A lot of them will say, look, I do an honest job, but there are all kinds of issues that come up when you're trying to design a trial. Like, what are you gonna measure, who’s gonna be included in the trial? Even if you're the savviest cardiologist, there are going to be aspects of the trial that the industry wants to see that you might not even be aware that's what they're doing.

BOB GARFIELD:  Now, it’s one thing for a pharma company to accentuate the positive, but your piece shows that they actively suppressed the negative. When medical journals continue to publish drug company-sponsored studies, aren’t they becoming accessories to a kind of scientific fraud?
PETER WHORISKEY:  Well, the - the journals themselves have come together and said, look, if you’re going to have a company funding the trial, we want to know what was their role in the study, who did the statistics, who did the data collection? They’re also requiring full disclosure of the academic authors: what are your financial connections to the company?

At the same time, there's been a, a sort of separate move that forces all clinical trials to be registered. What was happening in the past was that companies would trials and if the result was unflattering about their product, they wouldn't tell anybody. So they can no longer do that. The big gap, and this is sort of where all the mistrust lies, is that, for the most part, the journals don't see all of the data in a trial.

But, there's kind of a move afoot in Europe to force companies to put all of the data about all the trials that they have conducted on any drug that is on the market and put that up on the Web. That means if they come out with a press release saying this drug is great or this drug is safe, somebody, if they really, really want to, they could go through and do their own statistics on the data.

BOB GARFIELD:  These studies funded by pharma underpin the entire reprints business. The New England Journal of Medicine and others make money by selling reprint rights to pharma companies, and no pharma company reprints a study that makes its drug look bad. Are you confident that the New England Journal of Medicine was merely snookered here by a wily pharma apparatus?

PETER WHORISKEY:  I think that the editor of the New England Journal of Medicine, he’s genuinely concerned about these issues. He knows that patients lives are at stake. But there is a potential conflict of interest for any journal because you're right, there’s a lot of money to be made on reprints.

BOB GARFIELD:  You know, we’re done these stories quite often over the years and sometimes I feel as if I'm staring in the face of evil. In the case of Avandia, people died. What does GlaxoSmithKline have to say about that?

PETER WHORISKEY:  In essence, they say, we did our best. I have sometimes wondered the same thing. I go back and forth on this. There is genuine enthusiasm on these teams that are developing new drugs, and it’s not just for Avandia. I talked these other guys, and I went through and I said, you know, none of the evidence that you presented was very convincing, actually, in retrospect, given everything that we know - how could you have made all these mistakes? And this was a guy who didn't work for one of the drug companies; he’s a doctor at a prominent hospital. And he said, you know, at the end of the day, we were just really excited about the drug, we wanted it to work, and you blind yourself to some of the danger signals.

BOB GARFIELD:  Peter, many thanks.

PETER WHORISKEY:  Thank you very much.

BOB GARFIELD:  Peter Whoriskey is a writer for the Washington Post.



Peter Whoriskey

Hosted by:

Brooke Gladstone