The editorial mentions a practice of the pharmaceutical companies that appears to me natural (that is to say in accordance with human nature) but dishonest. The companies commission several, or many, controlled trials of their new drugs, but when it comes to publishing the results have hitherto been allowed to publish only those that are favorable to their products.
By means of omission, they are thus able to exaggerate the benefits, or minimize the risks, of the new drugs. This seems to have happened on a large scale with the SSRI anti-depressants: by failing to publish trials which demonstrated little or no benefit of these drugs, the companies were able to convey an entirely false impression to doctors of their efficacy. Thus the publication of properly-conducted and scientifically sound trials can, by the old rhetorical means of suppressio veri and suggestio falsi, mislead and has actually misled. By the way, only seven of the forty-two trials of rosiglitazone were published before the FDA gave it its imprimatur.
A solution is obvious: the registration of all trials and the compulsory publication of their results, favorable or unfavorable. But a still small voice at the back of my mind asks, “If this had been the policy in the 1840s, should we now have anaesthetics?” After all, people soon began to die of ether and chloroform.
New from Theodore Dalrymple today at PJ Lifestyle:
More on medicine at PJ Lifestyle: