Pediatric Epilepsy Readmissions: The Who, When, and Why (P3.251)

Objective: We report the characteristics of pediatric epilepsy patients who were readmitted to the hospital within 30 days of discharge, a previously undescribed population. Background: Prior studies have demonstrated a pediatric epilepsy readmission rate of 6-7[percnt], but have not described the readmissions. Methods: A quality improvement team was created at a pediatric hospital to review 30-day epilepsy readmissions. Results: Over a 16-month period we reviewed charts of 117 epilepsy patients who were discharged from the neuroscience floor and readmitted within 30 days. Categorizing readmissions revealed that 22[percnt] were scheduled (p=0.0001). Of the remaining readmissions 37[percnt] were for chronic epilepsy, 13[percnt] were for newly diagnosed epilepsy, 13[percnt] were for failure related to index admission treatment, and 10[percnt] were for unrelated diagnoses. We classified 26[percnt] of readmissions as preventable (p=0.0001). We determined 57[percnt] of readmissions were not preventable and 17[percnt] were of uncertain preventability. A high number of readmissions had an index discharge on Friday; patients less frequently had been initially discharged on Sunday or Monday (p=0.02). The day of readmission was spread throughout the week, but was less likely to occur on a weekend (Friday, Saturday, or Sunday, p=0.01). Though patients’ index discharges were from four services (neurology, epilepsy monitoring unit, neurosurgery, or general pediatrics), they were readmitted to eleven services. Patients initially on neurology typically were readmitted to neurology, while patients on other services were usually not readmitted to neurology (p=0.0002). Of note, 15[percnt] of readmissions were to the critical care unit, only half of which were scheduled. Conclusions: To our knowledge this is the first project to categorize, determine preventability, and provide details of pediatric epilepsy readmissions. Our project shows that a minority of readmissions were scheduled, and a larger than expected number were classified as preventable. Characterizing readmissions is the first step in being able to reduce readmissions. Disclosure: Dr. Vawter-Lee has nothing to disclose. Dr. Lake has nothing to disclose. Dr. Lutley has nothing to disclose. Dr. Fledderjohn has nothing to disclose. Dr. King has nothing to disclose. Dr. Zellner has nothing to disclose. Dr. Horn has nothing to disclose. Dr. Jones has nothing to disclose. Dr. Pilcher has nothing to disclose. Dr. Smith has nothing to disclose. Dr. Wesselkamper has nothing to disclose.