The impact factor & the almighty dollar

Since the inception of this blog, we have written frequently about conflict of interest, and it seems as though new types of conflicts continue to be uncovered. There's an interesting new paper and editorial in PLoS Medicine on how industry supported clinical trials affect a journal's impact factor. The authors reviewed randomized trials in 6 major journals, including JAMA and the New England Journal of Medicine. They found that industry supported trials had more citations than those not funded by industry. What I didn't know was that "non-citable" papers (editorials, news pieces, and letters to the editor) are included in the numerator but not the denominator of the impact factor calculation. For the New England Journal, removing the non-citable pieces would lower its impact factor by 24%. So there is a conflict of interest at play here: the publication of industry-funded trials not only increases the journals' impact factors, but the journals also make money by publishing reprints. For example, nearly half of Lancet's income is derived by selling reprints. As the editorial points out, the journals have greatly increased their scrutiny of authors' conflicts of interest, but the journals and their parent organizations may not be holding themselves to the same standards.