The commonly used relative risk to report the results of clinical trials may not even be interpretable as a measure of effect, according to a team of researchers led by Dr Suhail Doi, head, Department of Population Medicine at Qatar University (QU) College of Medicine.
This finding has widespread implications for how clinical trials and meta-analyses are reported in healthcare and how effects of treatments are interpreted. The findings also have an implication for previous trials and meta-analyses in the literature as it is likely that their interpretations may not be robust in the light of these findings.
"Our team of researchers are on the frontlines of methodological development in clinical research and are working hard to enable the most judicious use of research evidence," said Dr Doi.
"Our work in this paper will enable research output worldwide to provide more robust evidence of what works and what does not work when clinicians assess common interventions in clinical care.
"This work enhances the translation of clinical research in a time when we see an information explosion worldwide, The new findings allow researchers to understand better what the effect is of a given intervention. In addition, various concepts surrounding the reporting of binary outcomes in medicine are clarified which will make transparent reporting of clinical outcomes easier."
