Dr. Ben Goldacre did everything right. He had a patient with depression and other antidepressants did not work. He read the academic literature and thought that he had come up with a solution: reboxetine. The trials seemed well-designed and fair and they were overwhelmingly positive. Goldacre discussed the drug with his patient and, at the end of the session, Goldacre wrote the patient a prescription.
But in 2010, a group of researchers brought forth all the studies ever performed on reboxetine. Of the seven trials conducted on the drug, only one showed absolutely positive results. Only the positive ones had been published in peer-reviewed journals; the rest had been buried. Its efficacy was about at par with a sugar pill, and patients who had taken the drug in trials overwhelmingly had side effects, often prompting them to drop out of the studies. At best, Goldacre had prescribed something to a patient that would not work; at worst, the doctor had prescribed a drug that could harm him.
It is not altogether surprising (though it is disappointing) when we hear about pharmaceutical companies treating doctors to lavish conferences and lunches in which doctors are instructed to tout the benefits of one medication over another. But it is more shocking to hear that pharmaceutical companies are cherry-picking data from trials, making it increasingly more difficult to find out whether the information that you receive is accurate. And, worse yet, the process Goldacre describes is perfectly legal - reboxetine is still on the market.
Drugs are tested by the companies that manufacture them and there are many ways to smudge the results. Poorly designed trials can focus on small, unrepresentative patients, or pit them against a competitor's drug at an ineffective dose or placebo, or can analyze the results using methods that are flawed by design. Some academic papers are published by people who work for pharmaceutical companies and do not disclose them; some academic papers are owned outright by pharmaceutical companies. Pharmaceutical companies can supervise trials and can stop them at any time for any reason.
Who owns or conducts a trial may not seem like a problem. But one study found that, of the studies funded by pharmaceutical companies, 85 percent were positive. Meanwhile, in government-funded studies, 50 percent of them were positive - marking a huge discrepancy. Another study found that industry-funded studies were 20 percent more likely to favor the test drug. That bias also appears in academic papers and in conferences.
How is that bias so prevalent, or even legal, for that matter? Pharmaceutical companies - or anyone conducting a study - do not need to publish results on trials. Regulators get most of the information, but it is keep secret between them and the pharmaceutical industry, so no one can see them. Even then, they only receive early results, and many side effects of drugs can appear after a significant amount of time. Scientists conducting the trials often have to sign all sorts of confidentiality agreements so they are sworn to secrecy.
Just revealing the fact that researchers' hands are often tied by the pharmaceutical industry can come with costs. A 2006 paper published in the Journal of American Medical Association was conducted by the Nordic Cochrane Institute. The research, conducted in Copenhagen and Frederiksburg, found that 98 percent of trials were funded by the pharmaceutical industry and that an overwhelming majority (32 of 44) were supervised or could be stopped by companies. For the organization's efforts, Lif, the Danish pharmaceutical industry association, demanded an investigation into the scientists. The charges were later dropped, but not before the association sent letters to the researchers' hospital accusing them of scientific dishonesty.
Dr. John Buse from the University of North Carolina felt similar intimidation and harassment when he made a speech at a conference about the risk of escalated heart problems from a drug called rosiglitazone. GlaxoSmithKline attempted to silence him before moving onto the head of his department. In 2007, the U.S. Senate Committee on Finance released a report calling Buse's treatment intimidation.
GSK was behind another scandal when they performed a nine-year study on the antidepressant paroxetine and its efficacy in children. The study failed to find that the drug was particularly effective and, in fact, found that it raised the risk of suicide in children. Many doctors had been prescribing the drug off-label to children - a legal process which means that doctors can prescribe the drug, but it is not expressly marketed to that particular group. Because GSK was aware that people were prescribing the drug off-label, they had no legal need to describe the elevated risk in suicide. That news only came to light when the drug manufacturer revealed the risk to a section of the UK regulatory agency - and only after GSK had deliberately mixed results to make it seem like there was no risk at all.
In the end, the acceptable solution may not be to outlaw these drugs. After all, sometimes patients have side effects to the most effective drugs, and less effective ones are better than none at all. But it would probably be beneficial to consumers and doctors alike if the process was more transparent.