In Reply We would like to thank Drs Agrawal and Allen for their thoughtful comments on our research letter.1 Dr Agrawal points out that the American Society of Anesthesiologists (ASA) physical status classification is not perfect for assessing patient preoperative risk and cautions against using it for reimbursement purposes. We fully acknowledge the limitations of the ASA classification and agree with Dr Agrawal that improving the consistency in interpreting the ASA definition among clinicians may increase its reliability and thus reduce the variation in anesthesia risk coding.2,3 Proposed as one of the explanations for our findings, Dr Allen points out that the use of electronic medical records (EMRs) facilitates anesthesia risk assessment but may produce unintended errors. Both the imperfection of ASA classification and the wide adoption of EMRs could potentially affect anesthesia risk coding. However, were the coding affected only by these 2 factors, we would have observed both increases and decreases in anesthesia risk coding over time rather than consistent increases, indicating that some underlying factor has been driving the increase in a systematic way.4,5