Thoughts on Ghostwriting

There has been a lively and thoughtful discussion about ghostwriting in the comments section of my post on BlueSpark’s invitation to a doctor to write a review article on bupropion and depression. Here are some of the issues that have come up.

1. What’s in a name?

Michael Altus has been tenacious in his efforts to educate us in the still evolving vocabulary of this business.

“Ghostwriter”= A person who writes or otherwise assists in presenting the author's work without being acknowledged.

“Ghost author” = Identical to a ghostwriter.

“Guest author” = A person who is listed as an author without having made substantial contributions. This is what has also been called “identified author,” or “named author.”

Beyond this, there are acknowledged medical writers, usually listed at the end of an article, and typically described as providing “editorial assistance,” “assistance with data analysis,” etc…. These are not really ghostwriters, because they are visible. However, they are often spectral writers, or semi-ghost writers, because in some cases they have essentially done all the work, including the conception (in conjunction with the company’s marketing team), the outlining, the writing of the first draft, and the final editing. Anybody who does all that, or even part of that, should be listed as an official author. But it doesn’t look very legitimate to have a medical writer without an MD recorded as an author, and what the drug companies are looking for is legitimacy, whether genuine or fabricated.

2. So who should be listed as an “author”?

According to the web site of the International Committee of Medical Journal Editors (ICJME), in order to get your name up in lights as an author of a scientific paper, there are three specific criteria. To quote from their recommendations:

“Authorship credit should be based on 1) substantial contribution to conception and design, acquisition of data, or analysis and interpretation of data; 2) drafting the article or revising it critically for important intellectual content; and 3) final approval of the version to be published.”

As you can see, there’s quite a bit of wiggle room here. For example, what should count as “substantial contribution to conception”? If an academic has a brief phone conversation or an email exchange with a pharma-funded medical writing firm in which the topic of the article is discussed, the firm might later argue that this counted as a “substantial contribution to conception.” In fact these criteria do not require that an author do any actual writing—as long as they do some “revising.” As an editor myself, I know that “revising” is an expandable term, and can include everything from reading a draft of an article and saying “that looks great, I have no changes for you,” all the way to performing major surgery on the very organization of a paper.

3. Isn’t ghostwriting acceptable in some circumstances?

This is really the crux of the controversy about ghostwriting. Many commenters have pointed out that making science understandable is difficult, specialized work. It is unrealistic to think that scientists have the time or expertise to do the wordsmithing needed.

I agree, which is why I draw a bright line between industry-funded medical writing and academia-funded medical writing. Both ghostwriting and acknowledged editorial assistance are perfectly acceptable when the assistance is not arranged by a company that has a stake in the topic of the article. Most academic departments hire research assistants and editors whose jobs are to help write the many articles that get churned out by the more productive members of the departments. There is no incentive for such editors to bias the articles in any way.

But when an editor is paid, either directly or indirectly, by a drug company to work on an article, readers should be concerned, because now there is a powerful incentive for the editor to manipulate the content of the article to promote the sponsor’s product. If the final published article is not promotional enough, the writer won’t get a future assignment from the funding company. It’s as simple as that.

4. What’s the solution?

I have two recommendations:

--Journals should not publish articles that are clearly written in order to promote the funder’s product. Generally speaking, this would exclude any articles involving medical writing companies, even when their involvement is acknowledged. After the many recent example of corrupted scientific literature by drug company/medical writing firm partnerships, we can no longer have any trust that such teamwork is anything other than marketing.

--Journals should continue to publish research funded by industry, as long as the researchers sign disclosure statements assuring editors that they had complete control and involvement in every aspect of the paper. This means essentially no contact with the drug company after having accepted the money. Obviously, such research can still be highly tainted by bias, but the degree of bias is likely to be less extreme. Furthermore, as the medical literature gate-keepers, editors will scrutinize such research with extra care in order to make sure they are not unwittingly publishing advertisements in guise of science.