Medical advertising: The truth still lies in our hearts
2 March 2011
The report covered by Roehr  is truly astounding to all readers of medical journals. When it comes to the question of publication preference, I think free journals are bound to be confined within certain economical restrictions. However, I have several points that need to be discussed:
First, can this kind of differential reporting also be classified as "Publication bias"? As the publication bias denotes preferential publication of study results in a certain direction (either clinically significant or insignificant), a recently updated technological report  points out that this phenomenon has not changed obviously for decades, ever since the first inception of this term in 1960s. There have been several ways created to discover publication bias (such as Funnel plotting, inclusion of unpublished studies, etc.). In this case, I am wondering whether we can detect the difference between the style of free and subscription-only journals in a similar way?
Second, I think no matter how frequently positive findings appear in a free medical journal pertaining to new drugs, we should all apply our strenuous effort to scrutinize the methodology, analytic technique, and statistical results. The editorial board can be in some way or not biased against or in support of the result, but readers are not.
The option of free or subscription-only journal should not interfere in our thinking about the strength of a study. Rather, the emphasis should be placed upon the evidence level and the robustness that supports it.
1. Roehr B. Medical journals with advertising are more likely than subscription journals to recommend drugs. BMJ 2011;342:d1335
2. Song F, Parekh S, Hooper L, et al. Dissemination and publication of research findings: an updated review of related biases. Health Technol Assess 2010;14:1-193
Competing interests: None declared
Division of Nephrology, Department of Internal Medicine, National Taiwan University Hospital
Click to like: