So-called atypical antipsychotic drugs have been blockbusters for the drug industry, pulling in $16 billion in 2010. Developed to treat schizophrenia and related disorders, physicians also prescribe these drugs “off label” for bipolar disorder, insomnia, and other problems the drugs are not approved to treat, as Sandra Boodman wrote in the Post on March 12.
But a new report finds that psychiatrists have not been given a full picture of these drugs, which include big sellers like Abilify (aripiprazole), Zyprexa (olanzapine), Risperdal (risperidone), and Seroquel (quetiapine).
When seeking approval for eight atypical antipsychotic drugs, drug companies performed 24 studies, according to a Food and Drug Administration database. But four of the studies were never published in professional journals — and all four were unflattering for the drug in question.
Three of the unpublished studies showed that the new drug did not perform better than a sugar pill. The fourth study showed that while the antipsychotic drug helped patients more than a placebo, older, less expensive drugs helped patients even more.
“That’s bad if you’re marketing the drug,” said Erick Turner, the psychiatrist at Oregon Health & Science University who conducted the new analysis, which was published Tuesday in the journal PLoS Medicine .
Two of the unpublished studies, which included more than 300 patients, tested Abilify. Both found the drug to be no more effective than a sugar pill in treating schizophrenia.
The two other studies involved Geodon (ziprasidone). One study found Geodon to be no more effective than a placebo. The second found that while Geodon was more effective than a placebo, it was less effective than an older — and much less expensive — drug, Haldol (haloperidol).
Further, four studies of the atypical antipsychotic drug Fanapt (iloperidone) that were published left out unflattering data that showed other drugs worked better.
With this information absent from professional journals, psychiatrists are left with an incomplete picture of how well atypical antipsychotic drugs really work. “I think [psychiatrists] should be aware that what they’re reading in journal articles could be sanitized,” Turner said.
When Turner added the data from the negative, unpublished studies to the positive, published studies, he found the overall effectivness of this class of drugs in treating shizophrenia fell by a small amount, about eight percent.
“Overall, the drugs seem to work almost as well as we thought they did,” Turner said.
But if the negative data had been published, psychiatrists would have had more complete information to decide which of these drugs — if any — to prescribe. The negative studies on Geodon and Abilify might have pushed psychiatrists toward the other drugs, for instance.
Instead, the negative studies were buried on the FDA website, where only a specialist with a background in statistics could understand what they meant.
Turner worked at the FDA earlier in his career, where he discovered that he “was living in this highly censored environment” where negative studies never get published.
There’s a term in academic medicine for this phenomenon: publication bias. Studies that show a new drug in a positive light get published; those that throw a drug into question get buried.
Turner has been working to throw light on this phenomenon. In 2008, he found that many unflattering studies of antidepressants had been left unpublished.
He has a take-home message for psychiatrists: “When you have cherry picking of data, weeding out of data, then you can’t discriminate between drugs that are good and drugs that aren’t good at all. You don’t know when to be confident and when to be suspicious.”