Abstracts

Rationale: Responding to seizures in an epilepsy monitoring unit (EMU) is important for patient safety. Yet, there is variability between centers, and even within centers depending on several factors such as time of day, staffing, room location, and seizure detection/notification systems. Quality improvement methodology is a practical approach to improving seizure response times. In this study we sought to improve seizure response times, improve the accuracy and reliability of seizure response time data collection and develop a standardized and automated approach for seizure response data collection in a level 4 NAEC academic epilepsy center.

Methods: Following observations in our newly opened EMU that seizure response times were significantly varied and often delayed, QI methodology was followed to investigate and improve seizure alerts and responses. We first formed a local QI team of key stakeholders including QI experts, MDs, APPs, RNs, EEG techs, and Neuroscience administration. We created a Key Driver Diagram to illustrate the aim of this project (response time < 30 sec) along with all perceived barriers and interventions. We then created a process map outlining every step of the seizure alert and response process in order to identify possible “failure points” in the system. Next, we performed a root cause analysis which included surveying key stakeholders (nursing and EEG technicians) regarding barriers to a rapid seizure alert and response. Lastly, we performed two PDSA cycles, tracking response times following each intervention. This resulted in improvements in both seizure response times and data collection.

Results: We identified barriers to rapid seizures response times that occur at the level of the patient, the EEG technician and the RN. Over a 6-month period, 252 seizure response times were recorded and analyzed. The median seizure response time during the baseline period was 32s (range 5s-124s), 38s (range 6s-231s) after the first intervention and 26s (range 11s-57s) after the second. The time and effort for data collection was also recorded and analyzed. To structure and format 200 response times with the initial manual data logging system took approximately 120min. After implementation of the new system, structuring and formatting of a similar sized data set took 2min.

Conclusions: Through multiple PDSA cycles, we were able to improve seizure response times in the EMU while also reducing variation. Additionally, we identified and implemented a system for more efficient and accurate data collection which lays the groundwork for answering additional questions in regard to EMU workflows and improvements in patient care.

Funding: Please list any funding that was received in support of this abstract.: none.