Abstract TP70: EHR Based System Improves Stroke Alerting Process in the Emergency Department

Abstract

Background and Issues: Treatment decisions for patients experiencing acute ischemic stroke require complex imaging and evaluation. Specialized teams including Neurologists, pharmacists and CT technicians are alerted to the ED when a patient presents with signs or symptoms of acute ischemic stroke (AIS) to carry out these evaluations. At our facility, these alerts were delivered via a de-identified text message, preventing the collection of data needed for clinical effectiveness and quality improvement work. Purpose: This project's purpose was to improve the Stroke Alert process using Computerized Physician Order Entry (CPOE) to initiate the alert via a HIPAA Compliant Group Messaging (HCGM) application called Haiku. Methods: Haiku is an application available in the hospital’s electronic health record, which allows for retrieval of data unavailable when utilizing the de-identified alerting method. Data on 70 consecutive Stroke Alerts from 6 weeks prior to implementation of the Haiku Stroke Alert were compared to 80 consecutive Stroke Alerts 6 weeks after implementation. A comparison of Door to CT and Door to Needle times in those periods were also compared to ensure the new process did not cause any delays in these metrics. Results: Results showed the CPOE based Haiku Stroke Alert process was easily adopted and allowed for access to data necessary for clinical effectiveness and quality improvement work without causing delays in Door to CT (DTCT) or Door to Needle (DTN) times. Conclusions: In conclusion, this quality improvement project addressed the lack of data available to assess the Stroke Alert process. A new CPOE based alert process utilizing the HCGM app Haiku proved feasible, received positive feed-back from ED team who initiates the alert and the Neurology resident team who receives the alert. The Haiku alert did not appreciably add time to the DTCT or DTN times. This process allows access to comprehensive data that can be used to explore other avenues to improve the hyperacute phase of AIS triage and treatment. The availability of robust Stroke Alert data was the missing foundational element allowing for clinical effectiveness and quality improvement work.