[On how to analyze the credibility of a clinical trial or meta-analysis whose main result is expressed in odds ratio, relative risk or hazard ratio].

Beyond the type of design or the statistical analysis applied, the credibility of a research study lies in the compatibility of its results with the intensity that the reader could accept that the phenomenon studied might have from a biological point of view. Ultimately this requires a value judgment. The present article describes a procedure that can be used to objectively approach the limits of intensity that that a biological phenomenon could have, according to the data presented, so that, based on the reader's judgment derived from the available knowledge of the problem, the study can be deemed credible. The procedure is valid when the results of the study are expressed in odds ratio, relative risk or hazard ratio. Although these statistics are difficult to interpret, they are probably the most widely used in clinical trials and meta-analyses, that is, in studies whose methodological designs provide the highest level of evidence.