Level 4 seizure monitoring unit admissions are associated with reduced long-term health care costs

Abstract

Objective: This study was undertaken to determine whether admission to dedicated seizure monitoring units (SMUs) result in reduced health care use (HCU).

Methods: This was a retrospective open cohort study covering the years 2010-2018 of patients residing in Alberta, Canada, who were referred to the Calgary Comprehensive Epilepsy Program and admitted to a level 4 SMU. Patients were required to have ≥3 years pre- and postadmission follow-up. The outcome was the change in trajectory of composite HCU (primary care, specialist outpatient visits, emergency department visits, and hospitalizations) for the 3 years prior to and 3 years following SMU admission using the point of admission as the "index date." Secondary outcomes were HCU limited to specific settings. We excluded the first 30 days following the point of admission to mitigate the confounding admission would have on the postadmission HCU trajectory. We used adjusted restricted maximum likelihood linear and nonlinear effects models to determine trajectories expressed as Canadian dollars.

Results: A total of 315 of 600 (53%) patients met eligibility criteria. Mean age was 40 years (SD = 17.4), 176 (56%) were female, 220 (70%) had focal epilepsy, and 60 (19%) had functional seizures or physiologic seizure mimics without epilepsy as adjudicated by the attending physician at the point of discharge. Mean per person health care costs increased by CAD$341.28 (95% confidence interval [CI] = -25.17 to 707.74) for each successive 6-month interval prior to SMU admission (p = .07). Following admission, mean per person costs decreased by CAD$802.34 (95% CI = 699.62-905.06, p < .001) for each successive 6-month interval up to 3 years postdischarge. Similar trends were noted for primary and specialist care, emergency department, admitted care, and when nonlinear models were applied.

Significance: Admission to an SMU is associated with significant and enduring declines in HCU. Each 6-months following discharge overall HCU declined by a mean of CAD$802.34 and acute inpatient, emergency department, and outpatient physician interactions declined by 25%, 26%, and 18% respectively. Comprehensive epilepsy care not only reduces morbidity and mortality but also reduces cost.

Keywords: administrative health records; cohort study; epilepsy; epilepsy monitoring unit; health care use; video‐EEG telemetry.

FIGURE 1

Average total costs derived from log‐transformed nonlinear models for Alberta residents over the 3 years prior to and 3 years following admission to a seizure/epilepsy monitoring unit (SMU) for diagnostic and therapeutic purposes. Costs represent those incurred from direct health system interactions (outpatient, emergency, and inpatient care, prescription fills, and diagnostic imaging) for each 6‐month interval, are derived from resource intensity weighting and physician claims, and are expressed in Canadian dollars. The left panel is for a patient with generalized epilepsy of presumed genetic origin and a Charlson Comorbidity Index score of 0, the middle panel is for a patient with focal epilepsy and a Charlson Comorbidity Index score of 0, and the right panel is for a patient with mixed generalized and focal epilepsy and a Charlson Comorbidity Index score of 0. Blue represents pre‐SMU, and red represents post‐SMU health care use.