"Medical Journal Are an Extension of the Marketing Arm of Pharmaceutical Companies"

In the latest PLoS Medicine is an article by Richard Smith, former editor of the British Medical Journal, with a title which expresses his main point, "Medical Journals Are an Extension of the Marketing Arm of Pharmaceutical Companies."
Smith charged that pharmaceutical companies use publication of randomized controlled trial as a marketing tool. His most important contention is that they do so not by publishing poor quality trials or false data, but by designing trials so they are particularly likely to produce results favorable for the companies' products. Each article may be "one piece of a gigantic and clever marketing jigsaw." He listed the following tactics that pharmaceutical companies may use to increase the likelihood that each trial produces favorable results:

• Conduct a trial versus a treatment known to be inferior
• Use a dosage of the comparison drug that is too low (so it won't work), or too high (so it will have side effects)
• Keep the trial too small to have enough power to show the comparison drug is better
• Use multiple endpoints in the trial, but pick the one that shows a favorable result
• Do multi-center trials, but use only results from the centers with favorable outcomes for the product
• Do multiple sub-group analysis, but only publish those with favorable results
• Frame results for maximum effect, e.g., present relative, not absolute risk

Smith acknowledged that pharmaceutical companies rarely have resorted to suppressing research, but that this is rarely "necessary" since it is usually easy to assure favorable results with the tactics above. Also note that this article was apparently written before the deceptive practice of ghost-writing received more attention (as we noted here.)
His proposed solutions included public funding of trials, especially of the most prominent treatments available for a particular condition. Furthermore, he suggests publishing trials not in journals, but on "regulated" web sites.
My comments:

• Smith's list of tactics to improve the likelihood of favorable trial results is an important summation. But if these are the only tactics we must combat, the solution does not seem so difficult. A reviewer well-trained in the precepts of evidence-based medicine should be able to detect most of these tactics. Pushing for more stringent peer review by better trained reviewers might be enough to solve this problem. Underlying this could be a wider effort to train physicians and researchers in EBM.
• However, I fear that there is more outright dishonesty out there than Smith acknowledged. Some was of the sort demonstrated in the recent ghost-writing stories. (See our posts here, here, here, here and here.) (Also, see a BMJ article which Smith cited that cataloged similar deceptive tactics used by pharmaceutical companies when marketing drugs in Australia.) Combating such dishonesty will require a more active approach.
• Furthermore, one can draw further parallels between dishonest marketing tactics used by pharmaceutical companies and dishonest or unethical practics that have been springing up all over the health care spectrum. Scrolling throught the Health Care Renewal archives ought to show that leaders of all sorts of health care organizations, hospitals and health care systems; managed care organizations and insurers; medical schools and universities; government agencies; not-for-profits; etc., etc., etc have been caught acting unethically or dishonestly. This suggests that improving health care will require a more broadly based effort directed at all kinds of health care organizations and their leadership.
• Finally, marketing can be defined in a rather neutral was as "the process or technique of promoting, selling, and distributing a product or service" (by Webster's Collegiate Dictionary). Yet Smith's article (and some of his quotes from books by Angell and Kassirer) suggest that the word as acquired a more malev0lent tinge, especially in the pharmaceutical context. This suggests we ought to question why this has happened.