Just for a treat, we have a guest post for you today. A friend of mine in the medical communications business has been thinking about the whole Open Access debate, and has this to say on the murky world of medical pay-to-print publishing:
———————————
You can picture the scene of a sunny day at the races. A rank outsider from a small training stable is experiencing a rash of bets from savvy punters who realise that the horse is capable of a serious win on the current surface. As the morning wears on, the bookies shorten the odds and he begins to look like less of the newcomer and more like any other horse in the two o’clock. Now, what if the horse was called the Journal of Current Expertise in Horrendous Diseases (JCEHD) and the people placing the bets were doing so via paid submissions and went by the names of Mr’s Pfizer, Bayer, GSK and F. Hoffman-La Roche?
It’s a case of damned if you do and damned if you don’t. The question of whether to publish a clinical trial has been answered for pharmaceutical companies by the International Committee of Medical Journal Editors 1 . In the interests of fair balance and the empowerment of physicians and scientists to assess clinical data for themselves, the ICJME, comprised of representatives of leading medical journals, has decreed that all clinical trials must be published irrespective of outcome. This principle, once applied to phase II and III trials has now been expanded to include the arguably greater range of phase I first-in-man trials.
The phase I, II and III trials of all the medicines currently under investigation represent a colossal number of potential primary clinical trial papers. A large proportion of these studies inevitably have negative outcomes. Of those that meet their primary endpoints, only a small proportion will be considered ground-breaking or likely to forever alter patient therapy. In other words, the leading medical journals that form the representation on the ICJME won’t accept them for publication. Superficially, this is an incredibly hypocritical stance to take. To say that “You must publish that negative pharmacokinetic trial of six patients, but don’t submit it to my journal, matey” is a desperately unhelpful thing to be faced with. Thankfully, in medical publishing, there is an established food chain and this can be quantified by the ISI journal impact factor 2 – a numerical value based on circulation and onward citation of a journal’s articles. There are journals at the top (Nature Medicine, New England Journal of Medicine [NEJM] etc.) and there are journals at the bottom, like our own JCEHD. The structure is almost pyramidal with very few journals in the pointy top bit and many more lower down in the big, icebergy bit. I’d argue that this is true in general medical terms, but in most individual therapy areas, the structure is more or less cylindrical in that submission is appropriate to all the top tier journals, but only a cut of the middle tier, specialist journals. Some clinical trials may be conducted in very specialised areas and only be appropriate for submission to six or so journals in total.
It’s at the bottom of the cylinder where things get really interesting. Down there, amidst the supplement-only journals and the commissioned review journals in pharmacoeconomics lie the pay journals. These are usually general medical journals that are peer reviewed and they have three key features:
- Certainty
- Speed
- You have to pay a fee on submission and agree to page charges well in excess of anything found in other journals (which amounts to the same thing as paying for the submission).
Moreover, when a pharma company approaches these journals with a proposed paper, particularly for one of those positive-but-not-that-earth-shattering clinical trials, then a big light adorned with silver dollar signs and blinking the word ‘REPRINTS’ in pink neon is switched on. These journals know that their circulation subscriptions are almost worthless and their advertising revenue is, therefore, infinitesimal. But they also know that a paper with pharma connections is going to be good for up to 20,000 reprints. Even at a couple of quid a go, that’s a decent income if you can get enough of these papers in one issue. When the submission fee or the page charges are added, we have a tasty little business model.
Oh, my, that sounds just awful, doesn’t it? Pharma companies are paying for publications and the dirty, little journals are letting them! Well, as ever, there is more to it than meets the eye. I mentioned certainty. That is, if you talk to these pay journals beforehand and get cosy with the editor, then you can usually get your paper published. The second issue was speed. In the main, pay journals can publish an article on a timescale that makes a mainstream journal look like a glacier making its unhurried way down a not-terribly-steep mountainside. Now, put yourself in the position of a pharma company either faced with an impending product launch and need to have citable references to satisfy US promotional regulators or who is trying to meet the ICJME’s other draconian and worthy requirement that all trials are published within a year of completion. The pay journal offers a near guaranteed timeline (and a short one at that) that enables a pharma company to plan its promotional campaign and use proper, citable references in the material it is giving to doctors.
The relationship between pharma and the lower grade journals is an incredibly useful symbiotic relationship. Pharma gets to meet its obligations to regulators and the ICJME and the pay journal gets a whacking great reprint order. But there is more. Now we are some months and years into this relationship, something has started to happen to the pay journals – their impact factors have gone up … by quite a bit. The odds on the horse are shortening. The situation now is that many pay journals that were once right at the nether end of the ISI impact factor ranking have risen to positions equal, if not higher than, highly respected specialist journals. It is not unusual to see pay journals with impact factors of 2.5 or so and a European Journal of Whatever with an impact factor or around 2.2. I am not naming names here – I am not prepared to name and ‘shame’ (if it even is shame) as I need these journals for my daily job.
How has this rise in ISI rating happened? It’s quite simple really. The pharma companies that have learned how to work with pay journals to solve intractable publishing problems, made more acute by the ICJME, have been publishing very high quality, thousand-plus patient, phase III randomised trials in them. A researcher or another clinical trial investigator wishing to cite definitive clinical trials in the area will cite the pay journal article. If this happens enough times (and it clearly has) then the impact factor of the pay journal will increase.
Doubtless the principle of big pharma using this push-pull relationship of funding traded for predictability will shock some readers. However, I put it to you that is it not better for the trials to be published according to the requirements set out by the ICJME and is it not also nice to see the underdog win on race day?
References
1 ICJME guidance. http://www.icmje.org/ . Accessed September 2007.
2 ISI impact factors. http://scientific.thomson.com/free/essays/journalcitationreports/impactfactor/ . Accessed September 2007.
Author: Herbert Van Der Hibernator, University of Life.