Use of Odds Ratios on Anaesthesia Related Studies

In line with other medical journals, odds ratios are increasingly being reported in anaesthesia literature. The frequency of the use of odds ratio and how well it relates to the relative risk when it is interpreted as relative risk remains unknown. We investigated the use of odds ratio, and its relationship to relative risk and the incidence of outcome in this study. We identified 60 meta-analyses and 87 original articles that reported odds ratios. While relative risk could have been reported in 79% of the studies, only a small proportion (3%) of these studies have estimated and reported the relative risk in addition to the odds ratio. There is a significant bias if odds ratio is interpreted as relative risk, especially so when the incidence of outcome is high. While odds ratio is a valid measure of treatment effect in its own right, anaesthetists and investigators should be careful not to interpret odds ratio as equivalent to relative risk.