Moving Beyond Handoffs: Optimizing Decision Support Technology to Reduce Medication Timing Errors for Operative Patients. A Prospective Interventional Study (Preprint)

BACKGROUND Medication administration in the Operating Room presents a unique and heightened set of patient safety challenges due to a combination of design features of modern electronic health records (EHRs) and medication administration workflows common among anesthesiologists. Patients transferring from one area to another within a hospital system, particularly perioperative patients, are at increased risk for medication dose timing errors. OBJECTIVE A novel set of decision-support tools incorporating physician-centered design and fulfilling usability heuristics were developed for the operative suite to reduce dose timing errors for targeted medications. We sought to improve the visibility of prior doses of medications and related medications to inform anesthesia providers of relevant recent doses without increasing alert fatigue. METHODS Perioperative medication timing errors rates and anesthesia provider perceptions of EHR safety utility and usability were assessed before and after the implementation of electronic health record display reconfigurations and alerts in the operative environment. RESULTS In an interrupted time series analysis examining dose timing error rates for patients receiving nonsteroidal anti-inflammatory drug (NSAID) medications in the operating room, the cumulative incidence rate fell from a baseline of .56 (2014 – 16, 15/106) to .28 (2017 – 21, 27/375) after the introduction of novel visibility and alerting tools. Eighty-eight percent of anesthesia providers rated the newly introduced tools as useful in preventing errors, with 55% providers reporting less alert fatigue. Without exception, anesthesia provider respondents described the new tools as usable, effective patient safety improvements. All anesthesia providers reported increased confidence in the safety design of the EHR system to help reduce medication timing errors. CONCLUSIONS Medication decision support functionality for anesthesia patient care is an important, under-developed, opportunity for improving patient safety and provider satisfaction. The alerts developed for the anesthesia team achieved meaningful reductions in timing errors, and were overwhelmingly well-received by anesthesiology providers, who endorsed decreased alert fatigue and increased trust in EHR system safeguards.