Response to the “Comment on the interpretation of odds ratios in ‘association of frailty with postoperative outcomes’”

Dear Editor, We thank the reader comments[1] to our published manuscript,[2] the odds ratios (ORs) of 1.74 and 2.33 reported for general versus neuraxial anaesthesia were erroneously interpreted as “74 times” and “133 times” instead of 74% and 133% increase in odds of complications, respectively. In Figure 3 of the published article,[2] the most significant difference is observed in Clavien-Dindo (CD) grade IV/intensive care unit (ICU) complications, where patients with modified frailty index (mFI) ≥0.27 who received general anaesthesia had significantly higher complication rates compared to those with neuraxial anaesthesia (P < 0.001). The visual perception biases in bar graphs can lead to overestimating bar heights and OR values.[3] In all the CD groups with mFI ≤0.27 and ≥0.27, patients had overlapping complications, such as two or three complications in the CD I group or one complication in the CD I group and another in the CD II group, leading to a probable overestimation of the OR. To address this, we have computed the adjusted OR (AOR) using a logistic regression model[4] for patients having mFI ≥0.27 undergoing neuraxial and general anaesthesia. The AOR is 1.55, indicating that patients undergoing general anaesthesia have a 55% increased risk of complications compared to those undergoing neuraxial anaesthesia. Thus, it further supports our result that in patients with high mFI, the complication risk is higher in general anaesthesia compared to neuraxial anaesthesia. Authors contributionsFinancial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.