Abstract
When electronic anesthesia records are compared to pharmacy transactions, discrepancies in total doses of controlled drugs are commonly found (≈16% of cases), potentially affecting patient safety and placing hospitals at risk for regulatory action. Errors (≈5%) persisted even with near real-time drug reconciliation feedback to providers. A study was conducted to test the hypothesis of greater risks of discrepancy for longer-duration cases and for intraoperative handoff involving a permanent handoff of care.
Anesthesia drug documentation and pharmacy transaction data were examined for all anesthetics between May 2014 and September 2015 at an academic medical center, and discrepancies between the two systems were determined. Nine logistic regression models were constructed to evaluate the influence of covariates (for example, case duration, general anesthesia vs. sedation, and handoff involving a permanent transfer of patient care) on the presence of a discrepancy. Linear regression was also performed between case duration decile and the logit (discrepancy rate), stratified by anesthesia type and handoff.
For all models, handoffs were associated with higher discrepancy rates (p < 10−6; odds ≥ 1.38). There was a progressive increase in discrepancy rates as a function of the case duration.
Handoffs involving a permanent transfer of patient care during cases increase the risk of controlled drug discrepancies. Staff scheduling and assignment decisions to decrease the chance of a handoff occurring should help mitigate this. In addition, future studies should examine ways to improve the handoff process related to controlled drugs (for example, a formal, structured processes in the anesthesia information management system).
