Abstracts

Rationale: Patients with recurrent, stereotyped events of unclear etiology often warrant admission for continuous video EEG monitoring in an epilepsy monitoring unit (EMU) for diagnosis. EMU admission duration has been reported to range from days to weeks. Factors affecting admission duration have not been clearly identified to date. System factors often force monitoring units to control staffing and limit cost by restricting the number of days an EMU is open. It is unclear if a limited staffing EMU model reduces the event capture rate and therefore diagnostic yield of event classification admissions. The goal of this study was to determine whether a limited staffing strategy impacted event capture rates in patients admitted for spell classification. Methods: A retrospective chart review was performed at two comparable institutions, Mayo Clinic Hospital (MCH) in Phoenix, Arizona, and Banner Good Samaritan Medical Center (BGSMC) in Phoenix, Arizona. BGSMC is only staffed Monday through Friday, thereby limiting the total possible duration of admission to five days. MCH is staffed on a 24/7 basis, allowing for any day, non-limited admission durations. The objective was to determine if the rate of event capture differed between two institutions. A total of 300 spell classification admissions at MCH and 260 at BGSMC were reviewed over a comparable time period. Relevant data included patient demographics, time until first event during EMU admission, total admission duration, and medication and/or therapy changes as a result of the admission. Results: The event capture rate at MCH and BGSMC were 74% and 72% respectively (p = 0.57). There were a greater percentage of patients with non-epileptic events (NEE) at MCH than at BGSMC (62.7% vs. 47.3%). The mean duration until first event was 31 hours at MCH and 38 hours at BGSMC (p = 0.009). The mean length of stay was greater at MCH when compared to BGSMC (4.5 vs 3.3 days, p = <0.001). Conclusions: The results of our study indicate that a limited staffing EMU model does not decrease the diagnostic yield of spell classification admissions when compared to a non-limited model. These results serve to support and validate the use of a limited staffing model for EMU admissions, especially at institutions with limited resources.