Journal bias or author bias?
Published online: August 24, 2016
I read with interest the comment by Mark Wilson in the Indian Journal of Medical Ethics regarding bias and conflicts of interest in medical journals (1). Wilson targets one journal (the New England Journal of Medicine: NEJM) and one particular “scandal” to make his point that journals’ decisions on publication are biased by commercial conflicts of interest (CoIs). It is interesting that he chooses the NEJM which, by his own admission, had one of the strictest CoI policies and had published widely on this topic. The feeling is that if the NEJM can be guilty, they can all be guilty.
The history of how journals handled commercial CoIs makes for interesting reading, but that forms only the background for Wilson’s main point, a point that is supported by one incident: what he refers to as the Vioxx scandal.
Vioxx was a popular anti-inflammatory drug that was withdrawn after a large trial found evidence of its adverse cardiovascular events. The Vioxx saga is well known and much has been written about how soon the problem should have been detected, and the role of the researchers, manufacturer and regulator (FDA). Wilson covers some of this ground and by no means does he let those involved off the hook. However, his main point relates to the role of the NEJM, which published the landmark Vioxx trial, VIGOR (2).
Wilson argues that the NEJM displayed a bias by not publishing a letter (or heeding advice) regarding a suggested underestimation of the adverse events in the VIGOR trial. He also claims that the NEJM’s publication of its expression of concern over the trial was timed to minimise harm to the reputation of the journal.
I have a problem with both these accusations. First, I will address the accusation that the NEJM deliberately withheld publication of a letter expressing concern over the published VIGOR trial in order to maximise financial gain from reprints of the article. To validate this accusation, we need to have some idea of the probability of the publication of such a letter, regardless of its impact. It is likely that the NEJM receives hundreds, or even thousands, of letters questioning published articles. If the probability of the Vioxx letter ever making it to print was low anyway, to conclude that the failure to print was directly due to the potential impact on reprint sales seems convenient to Wilson’s argument, rather than certain. It also assumes a link between the accounting arm of the NEJM (and an actual calculation of projected sales) and the editorial arm – a link that suggests a conspiracy theory. It could also be argued that most of the reprint money had already been made, and that by provoking controversy (by publishing the letter), the journal would have enhanced its profile (and sales). Further, the people concerned would have been aware that failing to publish the letter (if true) might harm the journal’s reputation, something that is highly valued by the NEJM, and something that carries its own financial value.
The second accusation, regarding the timing of publication of the expression of concern, is more interesting. Wilson refers to internal emails suggesting that the release was timed to minimise damage. My argument is not about the intent here, but about the obligation of journals to publish such statements at all. In other words, what is the role of journals, as a means of publication of scientific research, in ensuring the validity of that research? They should have (and do have) policies and oversight regarding potential conflicts and the quality of the methods used, but should they also take on the role of the investigative journalist and hunt down discrepancies in the data provided?
Wilson accuses the NEJM of not investigating the results of the VIGOR trial. I do not think that the role of journals is to check the data supplied by authors. They may be sceptical in some cases, but at the end of the day, they have to trust the authors; it is not possible for them to check the data contained within each article. We all have to trust the researchers.
Bias is all around us, but I have seen at first-hand the lengths to which the major journals go to demand rigorous methods, and, therefore, ensure the publication of the truth. It may be asking too much to expect them to also police the accuracy of the data submitted to them.
Statement of conflict of interest
No conflict of interest to declare.
- Wilson M. The New England Journal of Medicine: commercial conflict of interest and revisiting the Vioxx scandal [Internet]. Indian J Med Ethics. Published online on June 15, 2016 [cited 2016 Aug 4]. Available from: http://www.ijme.in/index.php/ijme/article/view/2407/4974
- Bombardier C, Laine L, Reicin A, Shapiro D, Burgos-Vargas R, Davis B, Day R, Ferraz MB, Hawkey CJ, Hochberg MC, Kvien TK, Schnitzer TJ; VIGOR Study Group.Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group. N Engl J Med. 2000 Nov 23;343(21):1520-8.