Kaiser writes:
If everyone agrees that taking Avandia, the blockbuster G.S.K. diabetes drug, increases the risk of heart attacks, what’s the problem? Why is Avandia still being prescribed? Read this New York Times article [by Gardner Harris] to find out.
It’s pretty amazing:
Three years ago, Dr. Steven E. Nissen, a cardiologist at the Cleveland Clinic, conducted a landmark study that suggested that the best-selling diabetes drug Avandia raised the risk of heart attacks. . . . on May 10, 2007, 11 days before Dr. Nissen’s study was published in The New England Journal of Medicine, he and four company executives met face to face in a private meeting whose details have not been disclosed until now.
Fearing he would face pressure and criticism from executives, Dr. Nissen secretly recorded the meeting — which is legal in Ohio as long as one party to the conversation is aware of the taping. . . . GlaxoSmithKline had threatened scientists who tried to point out Avandia’s risks . . . Dr. Ronald L. Krall, GlaxoSmithKline’s chief medical officer, predicted almost exactly the results of another crucial study of Avandia that was two months from publication and whose results, according to scientific protocols and the company itself, should have been kept secret from the company. In an interview, Dr. Nissen said the recording showed that the executives hoped to persuade him not to publish his study by suggesting that they had contradictory information they would share with him in a joint study.
“In retrospect, it seems clear that neither statement was true,” Dr. Nissen said. “They did not have contradictory data, and they never intended to cooperate in any analyses.”
But I guess I shouldn’t be surprised:
But GlaxoSmithKline was sued in 2004 by Eliot Spitzer, who was New York’s attorney general at the time, over the company’s failure to publicize studies that helped to reveal that antidepressants could lead children and teenagers to engage in suicidal behavior. The company settled the lawsuit by agreeing to post all of its clinical trial data. These Internet postings became the grist for Dr. Nissen’s analysis.
I put that last bit in bold font to emphasize the ancillary benefits that can arise from requiring information to be made public. This is the opposite of the usual “IRB” attitude which is to keep everything hidden.
To continue with the story:
During the meeting with Dr Nissen, the four executives spoke as if they did not know the results of Dr. Nissen’s still-unpublished study. . . . But a week before the meeting, the Congressional investigators said, GlaxoSmithKline had been secretly and inappropriately faxed a copy of Dr. Nissen’s manuscript by a journal reviewer who also worked as a consultant to GlaxoSmithKline.
There’s some more horrible stuff in Harris’s article.
Putting myself in his place
What I’m wondering is, how would I react in this situation, either in the role of Nissen (the cardiologist who did the outside analysis) or as one of the statisticians within the company? I don’t know.
By this, I don’t mean that I think I would myself cheat, but I wonder if I’d be more likely to avoid the situation rather than confront it. It’s hard for me to imagine going into that meeting with a tape recorder. And, if I were a statistician working for the company, I don’t know that I’d have the courage to quit my job in protest.
I’ve occasionally been in institutional settings where people like, cheat, and steal–hey, it happens!–and these situations are always really tricky to deal with. For one thing, once people start lying, it’s hard to figure out exactly what’s happening. There’s also the trapped-rat syndrome: Do you really be the one who pushes the ethics violator into a corner where he or she has no choice but to lash out? And, as in the GlaxoSmithKline story, there are often lots of people with an incentive to keep the lie alive. I’m probably like a lot of people in that I tend to assume people are honest until I have strong information otherwise.
I wonder what’s going to happen to those Glaxo executives. Will they be going to jail? In general, I don’t think prison is such a good thing; I’d be more in favor of fines, lifestyle restrictions, and lots of community service, maybe some physical punishment if you really want to do that sort of thing. I don’t see what good is served by actually putting these people behind bars.
Interestingly, it seems (with the possible exception of the statistician in the meeting) that the company scientists acted with ethical integrity. They were asked to review the evidence, and found no statistical reason to criticize the new finding.
In my own practice, we had to stop an drug arm of a large study due to SAEs (see: Discontinuation of Quetiapine From an NIMH-Funded Trial Due to Serious Adverse Events Am J Psychiatry 2009 166: 937-938 ). I can say that (to my knowledge) there was no unethical or pressuring behavior from AstraZeneca. We'll have to wait for the study to complete for a more detailed analysis even though it will be limited by the DSMB's early termination decision.
Andrew,
I've read about a whole host of abuses perpetrated by pharmaceutical companies: questionable tactics used by sales reps, mistreatment of test subjects, cooking the results of clinical trials. There's an entire article in Wikipedia devoted to Eli Lilly scandals.
In light of these abuses, what would your advice be to aspiring statisticians who are contemplating a career with a pharamaceutical company? I expect many drug co.'s will be hiring at JSM 2010 this summer.
If everyone agrees that breathing exposes the body to dangerous oxidizing gases (i.e. oxygen), why is it still recommended that people breathe ?
Surely the relevant question is whether the risk of dying from a heart attack by taking avandia significantly outweighs the risk of dying from diabetes if you don't take it. Even harder to compare, the impact on quality of life under the two options.
These things are rarely as banally straightforward as the NYT makes them seem. In fact, a cursory search on pubmed finds papers suggesting increased risk of heart attack from taking ibuprofen, that is not wildly different from the increase in risk by taking Avandia. Why is anyone taking ibuprofen still ?
To Andrew [not Gelman]: One of the main causes of death in patients with diabetes is heart attacks — or the outcome Nissen analyzed. Also, there are MANY different pharmacologic treatments for diabetes — there is absolutely no need for anyone to take a drug that increases the risk of a heart attack. Diabetes is not a metastatic cancer with no other treatment options.
The issue with Avandia (rosiglitazone) seems to be that there is a safer alternative (piaglitazone). Like with Vioxx (rofecoxib), a drug with safety issues may be withdrawn if safer alternatives exists (including ibuprofen and celecoxib).
I am not an expert on this area and would not rush to any sort of judgement (my instinct has been that most researchers are highly ethical and doing their best to improve health). But that would be the rationale for considering an Avandia ban.
But this can be tricky; I have been quite concerned that the acetaminophen/paracetamol issues do not rise to the level that should cause concern but I need to spend more time with data to really think this through.
"Why is Avandia still being prescribed?"
Because diabetes itself raises the risk cardiovascular death. A number of my friends have died from coronary events and complications. With one possible exception, they were all diabetic. Therefore I suspect the net effect of taking Avandia is to lower, not raise your net risk of coronary problems. It's hard to tell what's going on from the New York Times article, and I suspect they have left out some important details. They usually do. What is the confidence for the relative risk? What's the relative risk of taking versus not taking Avandia? How age dependent is the relative risk? It's possible that Avandia should be prescribed only for some age cohorts.
To all:
I appreciate the information on the medical context. I think we can all agree that, whatever the best decision happens to be, the Glaxo executives were acting in an extremely unethical manner. Lying, cheating, and stealing are generally bad things to do.
Robert: Your question is an interesting one. I doubt that people who work for drug companies are much more (or less) unethical than the norm. It's just that the stakes are higher.
For example, typical unethical behavior by a university professor is to take your salary and do a minimum amount of work: don't prepare for your classes, show up late, cancel class on any pretext, then cover your butt at the end of the semester by giving everybody A's.
Typical unethical behavior by a statistical consultant is to muddy the waters to make patterns appear or disappear, for example divide up your data into 10 pieces and analyze each separately so nothing is statistically significant, or throw in extra data from another source, or play around with rules for selecting data until you find what you want.
But it all just seems so much worse when you're using statistics to cover up actual deaths.
I have no idea what advice to give to aspiring statisticians who are contemplating a career with a pharamaceutical company. I've actually done consulting in that industry myself. The people I've worked with didn't seem to have any desire to cook the data, and I didn't see any ethical dilemmas. Once or twice I've said no to consulting opportunities because I just didn't trust the people I'd be working with. But, then again, I'm a notoriously bad judge of character.
For those Glaxo executives who "threatened scientists who tried to point out Avandia’s risks[to the public]" I think jail is appropriate. As you say, people's lives are at stake. Let the executives forfeit a bit of their lives.
Seth: I have no problem with these people paying their debt to society. I just wonder whether there are other, more effective ways of punishment/deterrence/rehabilitation that aren't so expensive to the taxpayer.
It is very unusual in an employment situation to be able to _define the client_ as being anything other than your boss – i.e. patients and their family's, students and your colleagues, etc. rather than the person who decides if you keep your job.
And then you are just their advocate!
And those employment situations are mostly in universities or research institutes.
And even here that depends a lot on _who that person is_ and their view of this – unfortunately some in academe are as bad if not worse than any pharmacuetical firm executive I've heard about.
Now if I recall correctly from reading about the original GSK suit (judge was really insightfull to make access to data a part of the settlement) the more senior GSK execs would not have approved of what was done – there are easy safer ways for the firm to prosper.
But for those involved – it was just too tempting to _make things work_ to thier personal advantage – gee where else do we here that?
So, I would guess, those working with pharmaceutical firms have quite varied experiences in this regard.
K
p.s. I have mostly avoided working or consulting with pharmaceutical firms – but one consultation I did with a device manufacturer resulted in the clinical lead on our group being sued for $5 million because they were unsucessful at recruiting patients and therefor the trial was unsuccessfuly ended – it got all the way to court before being dismissed.
Money is a big deal!
This is a MAJOR THEME for our society!
whatever the scientific merits of Avandia; you CAN infer a LOT from the way GSK went about things: pretending to not know about the research, when in fact it was leaked by the journal staff (gasp!) themselves.
"Research" and "Profit" are deadly combination!
We saw this in the context of investment banking "research" with regards to the "Gaussian Copula(tion LOL!) function (for eg: http://www.wired.com/techbiz/it/magazine/17-03/wp…
For medical "research", the cat is out of the bag. The big pharma companies sometimes ghost write articles for "doctor/experts" in the journals etc. They use the (what was hitherto thought to be "private") computerized prescription data entered to enable you to collect medicine from your pharmacy, as a targeted tool to compensate Sales Reps for convincing doctors. all behaviour with a patina (in the "name of") of scientific enquiry/methods.
I don't think one day, it will be too far fetched to predict that like in a bad Hollywood movie, the big pharma will release germs into the population just to sell the cure!