A meta-analysis of statin drugs captured headlines in March with the tantalizing claim that the cholesterol-lowering drugs have no more side-effects than a placebo, or sugar pill. However, Dr. Ben Goldacre, one of the authors of the paper and a noted critic of so-called “bad science”—including misleading statistics and massaged data that can “sell” people on a drug without a full understanding of its risks—has come forward to claim that the conclusions of the paper were flawed.
The study in question (published in European Journal of Preventive Cardiology) compared the rate of side effects in people given a statin or a placebo across 14 clinical trials that included over 46,000 patients, but it only used published data. Dr Goldacre has been involved in other research showing that only around half as many side effects are reported in published studies as are recorded in the more detailed clinical study reports that pharmaceutical companies are required to submit to regulatory agencies.
Because the clinical trials that were included in the meta-analysis were sponsored by drug companies, there’s a good chance that they under-reported side effects as well. Dr. Goldacre claims that he tried to add an amendment to the text of the paper to address this issue, but was too late for publishing deadlines. While he still believes that this new study adds important insight into how side effects should be recorded and analyzed, as well as contributing to our understanding of the placebo effect in general, he chalks the attention-grabbing headlines up as yet another case of science gone wrong.
(Source: European Journal of Cardiology, March 12, 2014: doi: 10.1177/2047487314525531; www.badscience.net.)