Abstracts

Rationale: Adverse events in the Epilepsy Monitoring Unit (EMU) may include clusters of seizures and status epilepticus, falls, post-ictal psychosis and deaths [1]. There is wide variation in clinical practice in EMUs and little direct evidence that attention to quality and safety metrics help to reduce the number of adverse events [2]. Recent consensus based recommendations include using an EMU specific protocol to respond to seizures [3-5]. We initiated a quality improvement project focusing on using a standardized protocol for seizure assessments with the goal of helping to integrate EEG technicians' and nursing assessments, prevent falls, and improve patient safety in EMU. Methods: There are three components of the protocol: 1) the timing of the assessment, 2) the quality of the assessment and 3) review and feedback. It was agreed that assessments should occur within 30 seconds of seizure onset. In clinically obvious seizures the patient, family member or nurse may push the event button while the the technician recognizes a change in the electroencephalogram (EEG) with a subclinical or subtle seizure then pushes the event button. Time of onset is determined on retrospective review of the EEG and time from onset to the push button event is tracked; time from push button event to the nurse entry into the patient room is tracked. Nurses and EEG technicians were trained to perform a simple seizure assessment which includes testing patient memory, awareness, language and motor function. On review of the seizure, a simple five point scale to assess quality of assessment done by the nurse (or technician) was used to rate the assessments. For convulsive seizures one point is given for each of the following: lowering the bed, oxygen/suction used, removing blanket, turning patient and describing the seizure. For a non-convulsive seizure, one point is given for each of the following: assessing memory, language, awareness, motor function and circulation. The review and feedback component of the protocol involves reviewing a sample of seizures to determine seizure onset, time to assessment, and a rating of the quality of the assessment performed by the technician and/ or nurse or both. The details of the protocols were reviewed in lectures given to nursing staff and technicians. Feedback and clarification occur through "Quality Rounds" which occur every two months. Results: The quality of the assessments steadily improved from scores of 1-2 in 2013 to scores of 4-5 in April 2014. Assessments of children and babies generally score lower. Epileptologists are happier with the quality of the assessments. A review of a small sample of events showed that the time from onset to nurse entering the room averages 19 seconds with a range of seven to 60 seconds. The number of falls in FY 2012 was 11 compared to 6 falls in FY 2013 and 6 in FY 2014. Conclusions: Nurses and EEG technicians can be taught a simple seizure assessment protocol to minimize variability in seizure evaluation. Nurses in the EMU can reliably assess patients within 30 seconds of seizure onset with the cooperation of the EEG technologist.