Exposing the Ghost Management of Medical Research

Dr. Marc-André Gagnon (Carleton University) and Dr. Sergio Sismondo (Queen’s University) have co-authored an excellent op-ed piece on the Ghost Management of Medical Research. Ghost Management is related, though not the same as Ghost Authorship. Ghost Authorship refers to situations where influential academics or professionals (professors, doctors, etc.) accept authorship credentials on scientific or academic papers that they had little or no part in writing. Ghost Authorship is a process often used by large companies, the most famous cases being pharmaceutical companies, in order to help market their drugs.

This article emphasizes that most of the press on Ghost Authorship has missed the ethical mark, so to speak. Typical analyses of Ghost Authorship involve evaluating whether or not professionals ought to sign onto research they didn’t perform themselves. But Gagnon and Sismondo point out that there is a larger, and potentially more dire, outcome associated with Ghost Authorship, and that is that Ghost Authorship is usually a small part of the overall Ghost Management of Medical research.

Ghost Management often involves creating a highly biased perception of certain drugs, say that a drug has beneficial outcomes, when in fact the underlying research does not support that claim. This is accomplished by flooding the literature with research that is biased, by spinning or burying negative research outcomes, and so on.

Here is an excerpt from the article:

In 2008, research showed that pharmaceutical companies systematically failed to publish negative studies on their SSRIs, the Prozac generation of antidepressants. Of 74 clinical trials, 38 produced positive results and 36 did not: 94 per cent of the positive studies were published, but only 23 per cent of the negative ones were, and two-thirds of those were spun to make them look more positive.

Physicians reading the scientific literature got a biased view of the benefits of SSRIs. This helps to explain the huge number of antidepressant prescriptions, in spite of the fact that, according to a meta-analysis in JAMA in January 2010, for 70 per cent of people taking SSRIs, the drug did not bring more benefits than a placebo. Compared to placebo, however, SSRI antidepressants can result in serious adverse drug reactions.

There we see one of the problems with the ghost management of medical research and publication. Pharmaceutical companies want upbeat reports on their drugs. They design, write, and publish studies that are likely to show their drugs in positive lights – and there are myriad ways to do so. Ghosts sometimes bend the truth, and sometimes even commit fraud, with grave results…[read more]

Here’s what I think: All medical doctors should educate themselves on these issues by first reading Gagnon and Sismondo’s article, then the academic research on Ghost Management (included below), and should give it serious consideration. Ghost authorship might offer the opportunity to pad a CV, but at what cost?

For further reading you can check out these articles published in peer-reviewed academic journals:

Publication Ethics and the Ghost Management of Medical Research (Sergio Sismondo and Mathieu Doucet). Bioethics 24 (2010): 273-283.

Publication Planning 101: A Report (Sergio Sismondo and Scott Nicholson). Journal of Pharmacy and the Pharmaceutical Sciences 12(3) (2009): 273-283.

Medical Publishing and the Drug Industry: Is Medical Science for Sale?.Academic Matters (May 2009): 8-12.

Ghosts in the Machine: Publication Planning in the Medical SciencesSocial Studies of Science 39 (2009): 171-198.

Advertisements

Leave a comment

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s