Anesthesiology Control Tower: Feedback Alerts to Supplement Treatments (ACTFAST)

The study will logically build on our previous work and utilize the already established infrastructure and resources of the intra-operative electronic medical record and the AlertWatch® alerting system. In conceptualizing this pilot study, as well as a larger follow-up trial focused on clinical outcomes, we have been mindful of challenges put forward to investigators by the National Institutes of Health, which seek to support low-cost, pragmatic, patient-centered randomized controlled trials. Specifically in relation to ACTFAST, (1) the highly developed, specialized IT infrastructure limits the investment required for embarking on this study; (2) the study will be conducted entirely within the context of routine clinical care, negating the need for dedicated trial-related visits; (3) access to registries with granular data on complications and patient-reported outcomes obviates the need for a new and costly infrastructure to track patient outcomes; (4) inclusion of large numbers of broadly representative patients in this study will be highly efficient with a waiver of informed consent. The design for this pilot proof-of-concept study will be a randomized clinical effectiveness trial. It will include a 6-month pre-intervention period during which time the ACT will be staffed but no alerts will be sent. This will allow for the training of controllers, refinement of alerts, and optimization of processes for obtaining and filtering information from diverse electronic sources. Following this period, the trial will begin and run for 12 months. On a daily basis during the study period (weekdays from 7am to 5pm), each operating room in Barnes-Jewish Hospital (BJH) will be eligible for randomization, which will occur using computer-generated assignment. Anesthesiology teams in ORs allocated to the experimental arm will receive the additional support of the ACT in the form of control-tower alerts that complement the AlertWatch® system. The outcomes of interest in this pilot study will be the usability and usefulness of the ACT, clinician adherence to recommendations for monitoring, documentation and therapeutic interventions, key physiologic variables such as temperature and blood pressure. We will also document patient outcomes including duration of postoperative hospital stay, incidence of postoperative morbidity (myocardial infarction, surgical site infection), functional recovery, and postoperative quality of life. Data on these endpoints will inform the design of a subsequent trial focused on clinically relevant outcomes, which will logically follow the current pilot randomized trial. All of the alerts included in this study will be chosen because they follow proven best intraoperative management practice and are in line with national metrics for quality and safety.