Should journals sell reprints? No

Industry bias

Let me explain why this is important. In 2009 we published a systematic review of 259 studies of influenza vaccines published up to the end of 2007.3 We looked at several aspects of these studies, such as methodological quality (including internal coherence of the papers such as whether conclusions matched data presented), sources of funding, journal impact factors, and crude article citation rates.

Among our various findings was a strong association between industry funding and publication of a study in a high impact factor journal with consequent higher citation rate compared with non-industry funded studies. Some study characteristics (such as size and low quality) were the same for industry and non-industry funded studies. But industry studies were more likely to over-emphasise positive outcomes (a well known phenomenon). We did not include registration trials of new vaccines (which you would expect to attract more interest than confirmatory studies) because they reported surrogate outcomes such as antibody titres.3 We also found a growing trend for industry funded studies of influenza vaccines to be published in prestigious journals in the previous 20 years. If quality and size were not the key factors determining which trials ended up in the more prestigious journals, what could explain this finding? Could the journals themselves have a role in this phenomenon?

Drazen, editor in chief of the high impact factor New England Journal of Medicine, confirms the increasing reliance on reprint revenues. Speaking at the 2008 meeting of the Massachusetts Medical Society (which publishes the journal), Drazen said: “The results in recruitment advertising and bulk reprints were outstanding this year; they went a long way to offset declines in print-based revenue that all publishers are experiencing.”4 We know that some large journals derive up to 40% of their income from the reprint business.5

Openness is key

The value of reprints to a publisher can be gauged by comparing the actual per page cost of printing 100 000 copies of an article with coloured covers as part of a bulk order at your local office supply store (about €0.02 (£0.02; $0.03)) with the average cost of ordering a similar number of reprints from learned publishers (around €0.26 per page, excluding VAT and delivery). The figures are from my own survey of eight journals from four major publishers. Of course, larger journals are caught on the horns of a dilemma. They are the ones receiving more submissions and have high rejection rates. This means a larger staff and higher costs. Some journals and publishers have diversified by going into conferencing, events, and industry support (including the recruitment and development of key opinion leaders).

Reprints (especially those bearing the cover of a prestigious journal) are the currency of marketing. How many readers have been approached by drugs sales staff bearing gifts and a copy of the latest paper showing their product to be superior to that of their competitors?

All of this shows that large journals and large publishers are businesses. I am a reasonable reader as well as a producer of research. You can sell me what you want, but please tell me what I am buying. Editors and publishers should fill in an equivalent form indicating their sources of revenue for the previous year and the relative amounts.

Statements such as “This journal reports receiving reprint orders from Wonderpharma Ltd over the last year in excess of $250 000” would be a good start. An itemised disclosure by article or product would be even better. Just like the one that authors have to fill in.