Senator Grassley's report on ghostwriting

Those who follow developments on ghostwriting in the medical literature will be aware that US Senator Charles Grassley has been an outspoken critic of the practice. He has just released a report into his 2 years of inquiries into medical ghostwriting.

On the whole, much of what he writes is sensible, but there do seem to be a few ways in which he has slightly missed the point.

The main thrust of his argument, which I wholeheartedly agree with, is that ghostwriting is a bad thing, and is a particularly bad thing when financial relationships are not disclosed. If a drug company pays a ghostwrtiter to write an article, a respected academic signs his or her name to the paper without reading it, and the involvement of the drug company is not disclosed on the paper, then there is a serious lack of transparency and a worrying potential for bias. Grassley condemns such practices, and rightly so.

Grassley goes further, and points out that even if a medical writer is acknowledged for writing assistance, "it is unclear whether or not the academic expert evaluated the implications of what he was submitting for publication". That's a fair point. Acknowledgement of writing assistance, by itself, is not enough: it is also important to be sure that named authors control the content of the paper. This is a point also made in the EMWA guidelines on the role of medical writers in publications (which I'm pleased to see was cited in Grassley's report).

It is disappointing, however, that Grassley makes no mention of the legitimate role that professional medical writers have when properly acknowledged and when working in a meaningful partnership with the named authors. From reading his report, you could be forgiven for thinking that he wasn't aware that there are ethical medical writers out there who wouldn't dream of the sort of ghostwriting he describes. Given that he's supposedly spent 2 years looking into this, that omission is unforgivable.

A further point that Grassley makes, and in my opinion a valid one, is that policies and guidelines against ghostwriting need some means of enforcement. However, it is not clear who he believes should take on that role. One hint in that direction is that he does mention a checklist which I co-authored, published last year in PLoS Medicine, which is designed to allow journals to ensure compliance with good practices in this area. For example, it asks "Did the author(s) make the final decision on the main points to be communicated in the manuscript, particularly in the conclusion?" If journals were to incorporate the checklist in their instructions for authors, it would be far harder to publish papers with inappropriate involvement of medical writers. Sadly, although Grassley mentions the checklist, he doesn't make any recommendation that it be used, which is puzzling given that is designed specifically to stamp out the very practices he objects to so much. As an aside, it is also disappointing that PLoS Medicine, which published the checklist in the first place and has also been vocal in condemning ghostwriting, have not incorporated any requirement for authors to complete the checklist in their instructions for authors. That is also very puzzling.

My own opinion on the question of enforcement is that it is probably going to have to be the responsibility of medical journals to police the articles they publish. Granted, most journals lack the resources to do a good job of this, but I can't see any other realistic way in which it could be done. The ghostwriting checklist was specifically designed to allow journals to take on that policing role with minimal resources. I am sure that if journals were to take a more robust approach to eliminating ghostwriting, it would have a big effect.

There are one or two other strange things in Senator Grassley's report. Some examples of confusing and vague acknowledgements of writing assistance were taken from papers published over 10 years ago. At that time, there were no guidelines for acknowledgement of medical writers, so it is not surprising that acknowledgements at that time were not very helpful. Grassley also seems to be labouring under the misapprehension that, despite the publication of many guidelines condemning ghostwriting over the last few years, the prevalence of ghostwriting remains unchanged. He cites a study published in abstract form by some of the editors at JAMA looking at the prevalence of ghostwriting. It concludes "The prevalence of honorary authors has not changed significantly since 1996, but ghost authorship has declined significantly." However, Grassley misreports this, saying that "these numbers did not differ significantly from a 1996 study", and also "the prevalence of ghostwriting remains largely unchanged". He is also confusing "ghostwriting" and "ghost authorship", but that's a discussion for another day.

Not only did the study he cites find that ghost authorship has declined significantly, but my own research also found a decrease in the prevalence of ghostwriting between 2005 and 2008. Grassley didn't cite that research. Whether that's because he wasn't aware of it or because he ignored it because it didn't fit his story isn't clear. Either way, it calls into question the intellectual rigour of his report.

Anyway, those gripes aside, it's worth reiterating that the main thrust of what Grassley is saying, namely that it is important for medical publications to be prepared in an honest and transparent manner, is entirely sound.

Just one final rather intriguing point: given Grassley's clear disapproval of ghostwriting, it is very odd that the report has Senator Grassley's name on the front, and yet the text refers to him consistently in the third person. I wonder who wrote the report? The writer is not acknowledged. And does anyone know if Grassley writes all his own speeches?