Medical journals with advertising are more likely than subscription journals to recommend drugsBMJ 2011; 342 doi: http://dx.doi.org/10.1136/bmj.d1335 (Published 01 March 2011) Cite this as: BMJ 2011;342:d1335
Free medical journals with drug advertising were significantly more likely to recommend specific drugs that were advertised on their pages than were journals that relied upon subscription fees to cover their operating costs.
“Free journals almost exclusively endorse the use of the selected drugs, whereas journals that rely exclusively on subscription fees for their revenue are more likely to recommend against the use of the same drugs,” said lead author Annette Becker, MD, from the University of Marburg, Germany. The study was published in the Canadian Medical Association Journal (CMAJ 2011. Doi:10.1503/cmaj.100951).
The cross-sectional study examined all issues of 11 journals for general practitioners in Germany that were published in 2007. It classified the journals into three categories: “free journals” completely financed by paid advertising; “mixed revenue journals” financed with a mix of subscription fees and paid advertising; and “subscription journals” that were completely funded through subscription fees.
The study focused on continuing medical education articles for nine drugs or drug classes such as glitazones (diabetes), ezetimibe (hypercholesterolaemia), and varenicline (smoking cessation).
Within the data set of 465 issues of the 11 periodicals, the study identified 638 advertisements for the selected drugs, and 297 articles that made recommendations for or against their use.
The articles were then blinded and electronically reformatted to a neutral layout. Evaluation was along a five point scale with 0 being neutral; +1 and +2 indicators of moderate or strong recommendation for use of the drug; −1 and −2 indicators of moderate or strong recommendation against use of the drug.
Two dental students, who had no experiential biases using the drugs, were trained in the methodology to evaluate the articles. The researchers noted, however, that the evaluators soon were able to differentiate papers from advertiser supported publications because of their breezier style of writing.
They found that the mean recommendation score was 1.90 for advertising funded publications; 0.86 for the single journal that had a mixed revenue stream; and –1.03 for journals that relied solely upon subscription fees.
The authors believe that these patterns of bias seen in Germany are likely to be repeated in other developed countries that have a similar structure of both advertising and subscription based medical journals.
In an accompanying commentary, Aaron S Kesselheim, called the study valuable for the “empirical data it provides to inform the discussion over the role these publications play in medical education.”
Dr Kesselheim, a pharmacologist at Brigham and Women’s Hospital and Harvard Medical School, said similar biases were noted 20 years ago and physicians were urged to shun advertising supported journals. But that advice was never widely adopted, they are still popular among doctors because they often provide condensed, easy to read summations of recent studies.
Dr Kesselheim acknowledged that these journals cannot be banned. “However, their impact may be lessened if medical societies refused to be complicit and stopped selling their membership rosters or if professional organisations . . . did not certify CME [continuing medical education] credits sought by physicians reading these publications.”
He also urged government or other non-pharmaceutical funding to support creation of editorial content for physicians that is more reader friendly than the content of most traditional subscription based journals.
Cite this as: BMJ 2011;342:d1335
Becker A, et al. The association between a journal’s source of revenue and the drug recommendations made in the articles it publishes. CMAJ 2011. Doi:10.1503/cmaj.100951.