I have written here about the problems that the placebo effect – and the fact that it seems to be getting stronger – have caused for the pharmaceutical industry. They are finding it harder and harder to conduct studies on their drugs, because they can’t seem to make them perform any better than an inert sugar pill – a placebo.
What I hadn’t realized until I saw advance notices of a new book by British M.D. Ben Goldacre, is the problem that this is causing for doctors, and for their patients. In Bad Pharma, Goldacre argues that one of the byproducts of this growing inability to prove the effectiveness of drugs is that a world of disinformation is being passed along to doctors, who then pass it along to their patients by prescribing drugs for them that simply don’t work.
Seven trials on one drug, only one of them published
Dr. Goldacre presents the case of an antidepressant drug called reboxetine. A study was published in a reputable journal showing that it was effective, and performed better than a placebo. As a result, the Medicines and Healthcare products Regulatory Agency (MHRA), the British equivalent of the US Food and Drug Administration (FDA) approved the drug, and it began to be prescribed widely by physicians.
The physicians had been misled. As a result, so had their patients, who were now taking a drug that performed no better than a placebo, and no better than comparable drugs. In 2010, researchers gathered all of the data from all of the studies on reboxetine, and found that the one positive study that had been published was the anomaly. Six other studies, testing the drug on 10 times the number of patients as the published study, and they all found that reboxetine was no better than a sugar pill. None of these trials were published. Doctors never knew that they existed. The patients taking what they thought was an effective drug never knew they existed.
Worse, the researchers found that there were three trials comparing reboxetine to other antidepressant drugs in 507 patients, and finding it just as good as the drugs it was compared to. Those studies were published. But data from other studies on 1,657 patients found the opposite, that reboxetine performed much worse than the other antidepressants. And it had nasty side effects, whereas the drugs it was compared to did not. Doctors never saw these studies, either, because again they were never published. Reboxetine is still on the market in the UK, and still being prescribed.
Whose fault is this?
To some extent, it’s the fault of the pharmaceutical manufacturers, who design studies with too few subjects and inadequate controls, and then spin the results to make the drugs they have invested in developing look better. But Dr. Goldacre places part of the blame – at least in the UK – on the MHRA regulators, who often settle for one positive published study while ignoring the evidence presented by other studies that failed to be published because they hadn’t produced the results that the pharmaceutical companies wanted them to produce. 85% of drug studies recently reviewed by researchers at Harvard University were paid for by the companies that manufacture them; they also found that those studies that were funded by the drug manufacturers were 20 times more likely to find “results” that were positive. In another review of drug studies funded by the companies that manufactured the drugs, researchers found that the results from every one of the 45 studies they reviewed favored the drug made by the corporation paying for the study.
Is this just a British problem?
Not at all. Similar reviews of drug studies have been performed in the United States, with similar findings. A 2006 paper published in the prestigious Journal of the American Medical Association found evidence that researchers who had conducted Big Pharma-funded tests that didn’t find positive results were commonly pressured not to publish their work. The sponsoring company gets to see the research data first, and then (because they paid for the study, and thus own the data) get to determine whether it is to be submitted for publication or not.
So the problem continues. Research paid for by Big Pharma gets published if it puts their drugs in a positive light, and gets hidden if it doesn’t. The FDA reviewers making a choice as to whether the drug is effective enough or safe enough for them to approve see only the cherry-picked published data. And, once the drug is approved, the doctors who prescribe it get to see only the same positive, cherry-picked data, and thus prescribe the drug for their patients.
Meanwhile – in an extraordinary number of cases – the drug doesn’t work, and never did.