Evidence that medical writers improve papers
It is widely believed in the medical writing community that professional medical writers write better papers than people who are not professional medical writers.
It seems a logical proposition, doesn't it? After all, it's generally accepted that brain surgeons are better at doing brain surgery than people who are not brain surgeons, that airline pilots are better at flying planes than people who are not airline pilots, and that bankers are better at running banks than people who are not bankers (OK, maybe that last example wasn't such a good one, but you get the idea).
The trouble is, there is almost no evidence that it is true.
Now, I'm not saying that it isn't true. Absence of evidence is not evidence of absence, and all that. The fact is that it is extraordinarily difficult to do research on whether papers written by professional medical writers are better than other papers.
For a start, what do you mean by "better", and how do you measure it? That's not a trivial question. And then we have the problem of identifying papers written by professional medical writers. That's getting easier these days now that the medical writing community is becoming more aware of the need for clear acknowledgement of medical writers' involvement in published papers, but some papers are still "ghostwritten" (ie written by a medical writer, but with no acknowledgement or any other way for the reader to know about the medical writer).
The only study I'm aware of before now that really looked at the question was a study by Karen Woolley and colleagues, which found shorter times from submission to acceptance in papers written by professional medical writers, although their results just failed to reach statistical significance.
So I'm pleased to report that I have managed to do a study of whether medical writers improve the quality of papers, and that I found that they do. The paper has just been published in EMWA's journal The Write Stuff, and you can download a copy here.
Now, before you get all excited, I should point out that there are a number of limitations. This was not the most high-quality piece of research in the world ever, and the findings need to be interpreted with those limitations in mind. I have discussed them at length in the paper.
Also, the improvement in quality was very small. I measured "quality" (OK, one particular dimension of the very complex multifactorial concept of quality) by the extent to which the papers complied with the CONSORT guideline, and found that papers written by medical writers complied with only 0.75 of an extra item (out of a 22-item checklist), on average, compared with other papers. The difference was statistically significant, but undeniably small.
Despite those caveats, I'm rather pleased with this research. It may not be a brilliant piece of research, but to my knowledge it is the first time that anyone has shown empirically that papers written by professional medical writers are statistically significantly better than those that aren't. You have to start somewhere.
We're currently working on a larger study, which I hope will be free of some of the limitations of the study described above. Watch this space.
Looks like a really nicely done study and well written paper. Well done!
I hope it leads to more evidence that is equally well presented.
It is not just a case of complying with guidelines, it is a matter of taking responsibility. If the paper is not written by the people whose names appear as authors, that, in my view, is simply dishonest As far as I know, this phenomenon is limited to the clinical world.
In other branches of science, the problem is rather different. The corrupting effect of the RAE has resulted in guest authors on papers who have little to do with the work.
It's hard to say which is worst.
Hi David
You're absolutely right about taking responsibility. Everyone needs to do that. Authors of papers need to take responsibility for the content. Medical writers need to take responsibility for ensuring that content is appropriately controlled by authors. Journals need to take responsibility for asking the right questions about how papers were developed.
I disagree, however, that it's necessarily dishonest if a paper is written by someone who is not named as an author. If a medical writer isn't named anywhere, that's dishonest. But if the writer is named in the acknowledgements section, together with the source of funding, what's dishonest about that?
Now, you could argue about whether a medical writer should be mentioned in acknowledgements or listed as an author. But on the whole, if you follow ICMJE criteria, it's not appropriate to list a medical writer as an author (sometimes it is, but those occasions are rare).
If you follow a contributorship model rather than an authorship model, that solves some of the problems, as the medical writer is then clearly a contributor and should be named as such. However, very few journals follow that model.
[...] there is some evidence that professional medical writers shorten times to acceptance of papers, improve compliance with the CONSORT standard for reporting randomised controlled trials, and reduce the risk of [...]
[...] help get them published would be utterly perverse. We also have evidence that medical writers can improve the quality of publications and make it less likely that papers will be [...]
[...] help get them published would be utterly perverse. We also have evidence that medical writers can improve the quality of publications and make it less likely that papers will be [...]