Abstracts

Rationale: Time to treatment in convulsive status epilepticus (CSE) is often significantly delayed. While the majority of CSE episodes start out-of-hospital, a minority of patients receive pre-hospital treatment, and this includes patients with known epilepsy or prior episodes of CSE. We sought to improve pre-hospital management of pediatric CSE via creation of a novel simulator-based educational model for caregiver administration of seizure rescue medications (RM). Methods: We created and validated a novel child torso-to-feet mannequin with regard to realism and effectiveness at improving accuracy and time to rectal diazepam administration. The trainer was designed using traditional mold-making techniques and 3D printing to include a level of joint stiffness resembling active tonic seizure. A rectal depth gauge provided mechanical confirmation of correct positioning of applicator tip insertion. The mannequin was outfitted in children’s clothing to enhance realism. Face and content validity of the simulator were assessed by 5-point Likert scale surveys. Construct validity was established by timing of RM administration and “RM Score.” The “RM Score” was designed for this study to monitor participants in accurate completion of each medication administration step. If a participant was unsure of the step or the step was incorrectly performed, a prompt was provided (1 point) followed by the step if needed (2 points), such that all participants could move through the simulation regardless of prior experience. Participants were scored by 2 independent reviewers with scores ranging from 0 (zero errors) to 22 (max errors) pre- and post-educational intervention. Time-to-RM administration pre- and post-educational intervention was also measured. Results: Twenty-three healthcare providers (nurses/nurse practitioners, residents/fellows, and attending physicians) and 5 caregivers completed the validation. Mean face and content validity scores were all greater than 4 on a 5-point scale (Table 1). Mean reduction in time-to-RM (-12.3 seconds) and improvement in “RM score” (-4.2 points) before and after training was statistically significant (both p<0.001). While participants with >10 years of experience in caring for a child with seizures (n=10) had marginally faster pre-training time-to-RM administration than those with less experience (43.8 versus 62.1-65.3 seconds, p=0.06), there were no other significant differences in pre-training time or scores based on role, years experience (<5, 5-10, or >10) or self-rated experience (novice, intermediate, expert) (Table 2). Inter-rater “RM scores” were in agreement in >70% of instances for all steps. Conclusions: The simulator displayed high scores on face and content validity, suggesting the model to be an overall realistic representation of a seizing child, with accurate reproduction of the clinical experience and motor processes of rectal diazepam administration. The educational intervention lead to improved accuracy scores and faster time to administration in all participants with authentic rapidity patterns across the expertise gradient. Inter-rater agreement was acceptable. The novel child simulator and curricular package offer a potential new tool to aid in optimizing important timely treatment of CSE with future studies now aimed at measuring these in the field as well as associated improved clinical outcomes. Funding: Pediatric Safety and Quality Graduate Medical Education Grant Program and Epilepsy Research FundReference: Sanchez Fernandez et al. Neurology, 2015. 84(23): p. 2304-11.