Wednesday, September 7, 2016

Have Companies Stopped Ghostwriting? BMJ Article Says "No"

A new article in the British Medical Journal (BMJ) by Alastair Matheson takes a fresh look at ghostwriting in medical research. Apparently, pharmaceutical companies are waging a campaign to convince us that they are now opposed to ghostwriting. But Matheson argues that the practice continues, only under a different name: "editorial assistance."

It comes down to how we define ghostwriting. The standard definition in the Oxford dictionary is straightforward: a ghostwriter is “a person whose job it is to write material for someone else who is the named author."

If a company recruits academics to put their names on a paper, but then pays an unnamed medical writer to actually compose the paper, that's clearly ghostwriting and is a deceptive practice. But it's rarely so straightforward. Yes, companies recruit academics to help design, conduct, and write research studies. Yes, companies also hire medical writers to do the grunt work of writing up the results of a study. Typically, medical writers are acknowledged in small print in the footnotes. Companies are claiming that this acknowledgment means they are being honest, and properly informing readers that these non-academics also contributed to the paper. Therefore, they are innocent of ghostwriting.

Matheson disagrees, arguing that these medical writers do a lot more than just the tedious rendering of the methods and results of research. Instead, they work closely with the funding companies to shape the manuscript in such a way that the funded drug looks good. The identified authors, whose names are in lights just below the title, may indeed edit and review the paper before publication, but the companies have plenty of leverage over the final content.

Simply publishing a footnoted acknowledgment of a medical writer for "editorial assistance" is not enough, he says, because it covers up a process in which the company uses the medical writer as a conduit to spin the paper in a commercial way. The solution? The International Committee of Medical Journal Editors (ICMJE) should require that any writer who contributes to the manuscript should be listed as an official author. He points out that the journal Neurology already has this policy in place, and indeed, if you look up their authorship policy here, you'll find the following crystal clear statement:

"Professional writers employed by pharmaceutical companies or other academic, governmental, or commercial entities who have drafted or revised the intellectual content of the paper must be included as authors."

Furthermore, Matheson calls for ICMJE to develop other rules to prevent what he calls "attributional spin." For example, in deciding the order of the listed authors, those with the most actual influence over content (often the employees of the company) should be listed first, rather than being listed after the academics as is sometimes done.

It's a good read. The article piqued my curiosity because over the last few days I've been slogging through the literature on antidepressant-induced sexual dysfunction to write an article for the next issue of the Carlat Psychiatry Report. Most of the articles are industry funded, and all of them relegate mention of medical writers to footnotes.

For example, here's page 1 of one of the studies I've read. There are five authors, the first of whom is an independent academic, and the last four are employees of vilazodone's maker Forest.





And here's the last page of the paper, which acknowledges Adam Ruth, a medical writer, for "writing assistance and editorial support."



According to Matheson, Adam Ruth should be listed as an author. I agree.

It's true that in this particular case, you could argue that Forest is already being transparent enough by clearly stating that four of  the authors are company employees. Nonetheless, adding the medical writer as an author reminds the reader that industry funded articles are essentially commercial, rather than academic enterprises. Maybe that's obvious, but sometimes it's worth stating the obvious.