Mediated interruptions of anaesthesia providers using predictions of workload from anaesthesia information management system data

Abstract

Perioperative interruptions generated electronically from anaesthesia information management systems (AIMS) can provide useful feedback, but may adversely affect task performance if distractions occur at inopportune moments. Ideally such interruptions would occur only at times when their impact would be minimal. In this study of AIMS data, we evaluated the times of comments, drugs, fluids and periodic assessments (e.g. electrocardiogram diagnosis and train-of-four) to develop recommendations for the timing of interruptions during the intraoperative period. The 39,707 cases studied were divided into intervals between: 1) enter operating room; 2) induction; 3) intubation; 4) surgical incision; and 5) end surgery. Five-minute intervals of no documentation were determined for each case. The offsets from the start of each interval when >50% of ongoing cases had completed initial documentation were calculated (MIN50). The primary endpoint for each interval was the percentage of all cases still ongoing at MIN50. Results were that the intervals from entering the operating room to induction and from induction to intubation were unsuitable for interruptions confirming prior observational studies of anaesthesia workload. At least 13 minutes after surgical incision was the most suitable time for interruptions with 92% of cases still ongoing. Timing was minimally affected by the type of anaesthesia, surgical facility, surgical service, prone positioning or scheduled case duration. The implication of our results is that for mediated interruptions, waiting at least 13 minutes after the start of surgery is appropriate. Although we used AIMS data, operating room information system data is also suitable.