As we reported earlier today, new research has discovered that pharmaceutical companies withheld a handful of nonsignificant and negative data from publication when working to get the U.S. Food and Drug Administration (FDA) to approve atypical antipsychotics. However, the problem was significantly less severe than the publication bias researchers found when looking at antidepressants.
Antidepressants have been especially hard hit when looking at the FDA pre-approval research. In fact, in Lesley Stahl’s recent 60 Minutes report on antidepressant research, she walked away completely baffled by the meaning of it all. What does it mean when researchers find such negative findings that were never published?
Let’s find out…
In the current research, scientists examined 24 FDA pre-marketing studies for eight second-generation antipsychotics (also referred to as atypical antipsychotics):
- aripiprazole (Abilify)
- iloperidone (Fanapt)
- olanzapine (Zyprexa)
- paliperidone (Invega)
- quetiapine (Seroquel)
- risperidone (Risperdal)
- risperidone long-acting injection (Consta)
- ziprasidone (Geodon)
The researchers then compared the results in the FDA’s review documents to the results presented in medical journals. Ideally, they would expect to find 24 published studies, but instead they could only find 20:
[… F]our premarketing trials submitted to the FDA — which yielded unflattering results — remained unpublished. Three showed the new antipsychotic drugs had no significant advantage over a placebo.
In the fourth, the drug was superior to a placebo, but it was significantly inferior to a much less expensive competing drug, the researchers note.
Just 17 percent of studies were not published, which is actually lower than the industry average for new drug approvals winding their way through the FDA process.
And this isn’t nearly as bad as the data — 40 percent of the studies were never published — surrounding antidepressants:
In 1998 Moore used the Freedom of Information Act to pry such data from the FDA. The total came to 47 company-sponsored studies — on Prozac, Paxil, Zoloft, Effexor, Serzone, and Celexa — that Kirsch and colleagues then pored over. (As an aside, it turned out that about 40 percent of the clinical trials had never been published. That is significantly higher than for other classes of drugs, says Lisa Bero of the University of California, San Francisco; overall, 22 percent of clinical trials of drugs are not published. “By and large,” says Kirsch, “the unpublished studies were those that had failed to show a significant benefit from taking the actual drug.”)
In just over half of the published and unpublished studies, he and colleagues reported in 2002, the drug alleviated depression no better than a placebo. “And the extra benefit of antidepressants was even less than we saw when we analyzed only published studies,” Kirsch recalls. About 82 percent of the response to antidepressants — not the 75 percent he had calculated from examining only published studies — had also been achieved by a dummy pill.
The important thing to keep in mind is that pre-marketing research is conducted primarily in order to get a drug through the FDA process. It is not the final word on a drug’s effectiveness, it is simply a bureaucratic hurdle drug companies must cross in order to get their drug on the market.
Once on the market, dozens — and in the case of antidepressants, hundreds — of additional studies are carried out. These studies, which are often more varied, independent, and done by a wider range of researchers, eventually make up the majority of a drug’s efficacy research.
So the silver lining on this latest research is that the percentage of studies never published is actually lower than the industry average, and significantly lower than the number of studies never published before antidepressants were approved.
Read the full news article: Publication Bias May Give MDs an Incomplete Picture of Antipsychotics
Reference
Turner, E.H., Knoepflmacher, D., & Shapley, L. (2012). Publication Bias in Antipsychotic Trials: An Analysis of Efficacy Comparing the Published Literature to the US Food and Drug Administration Database. PLoS Medicine, 9(3): e1001189. doi:10.1371/journal.pmed.1001189
2 comments
, again John, here we can “spin†all day and until the cows come home. First of all there is a huge difference between a drug not lowering your cholesterol like they promised and a drug causing you to behave in a unhealthy dysfunctional way. Or a drug that promise to give you an erection falling flat (ehhmm) and one that causes a person who was originally being treated for OCD or mild anxiety that goes on to kill themselves. If I don’t get an erection, I go back to the doc who say, “ehh, doesn’t work for everybody.†The drug company or the doctor are not at risk of being held liable. If I kill my family in a delusional fit of manic rage, and I can prove that the drug company is more liable then McDonald’s is for a hot cup of coffee, It can cost the drug companies billions and cut the throat of a cash cow for them. A failed study when it comes to antidepressants is much more profound in magnitude then one that doesn’t work for weight loss.
This bring up the second major difference in glassing over a psychotropic drugs. The worst part about antidepressants isn’t that they aren’t working they are working too well. Here is a simplistic analogy. Take a toddler who keeps getting yelled for taking his brothers toys. So (as Stephenson proved in ’67) he develops an anxiety against taking the toy. That anxiety doesn’t mean he doesn’t desire it. The inability to fulfill that desire causes him to be depressed. Of course he can’t keep taking his little brothers toy thought it would make him very happy. We can’t ask his mom to stop punishing at him when he does it, this is an important lesson for a child. But what if we could give him a pill that would reduce the anxiety he feels about getting yelled at. So he starts taking the toy again, ignoring his mother, and he is happy. Now, the only person you can talk to in a study about the pill is the child. He declares he is now happy again. Pill successful!
If a cholesterol pill doesn’t work you can run test and see it. If an ED pill doesn’t work it has other notable signs. If a weight loss pill is causing your hair to fall out, there is direct ways to test it. These are all unwanted symptoms. But if a treatment causes you to become manic, there is no way to tell. There is no money in proving these pills don’t work. There is lots of it in proving they do.
My apologies for the typo and poor grammar. I got called away and just hit submit. Thanks for the time and post.