Impact of Vagus Nerve Stimulation for the Treatment of Drug-resistant Epilepsy on Patterns of Use and Cost of Healthcare Services and Pharmacotherapy Among Medicare Enrollees: Findings From Analyses of Healthcare Claims From the Centers of Medicare and Medicaid Services

Purpose

To examine the expected impact of vagus nerve stimulation (VNS) on patterns of utilization and cost of healthcare services and prescription pharmacotherapies among Medicare enrollees with drug-resistant epilepsy (DRE) versus continued use of antiseizure medications (ASMs) alone.

Methods

This was a retrospective, observational, cohort study that used healthcare claims data from the US Centers for Medicare and Medicaid Services. All Medicare enrollees who underwent VNS implantation between January 1, 2011 and December 31, 2020 were selected. Individuals without at least 24 months of continuous enrollment before implantation (index date) and at least 1 month of enrollment immediately thereafter were excluded. Patients without a diagnosis of epilepsy on the index date, and those without ASM claims during the 1-year period before that date, were also excluded. Observed patterns of utilization and cost of healthcare services and pharmacotherapies during the 2-year period prior to VNS were used to develop regression models to predict these outcomes during the 2-year period following the index date. Predicted monthly outcomes from these models during each month of the 24-month follow-up period were compared with corresponding outcomes observed in the database, with differences (observed minus expected) attributed to VNS implantation.

Findings

A total of 16,223 Medicare enrollees had a procedure code for VNS between January 1, 2011, and December 31, 2020, of whom 19.4% (n = 3155) met all other selection criteria. Expected composite rates of hospitalizations and emergency department (ED) visits were higher than observed for all-cause (38.95 events per 100 person-months [PMs] vs 23.15 per 100 PMs) and epilepsy-related (33.46 per 100 PMs vs 15.97 per 100 PMs) events (P < 0.001 for both comparisons). Following the index month, mean monthly observed all-cause costs were $1286 lower than expected; epilepsy-related costs were $1351 lower. Differences between predicted and observed all-cause costs (including costs related to implantation) did not differ significantly by month 20, indicating an expectation that VNS “breaks even” within 2 years of implantation.

Implications

VNS implantation was associated with 41% and 52% reductions in all-cause and epilepsy-related hospitalizations and ED visits, respectively (both vs expected), for Medicare patients with DRE, and its implantation may be cost-neutral within 2 years of the procedure. These results are similar in direction and magnitude to those observed in a previous study of commercially insured patients with DRE. Additional research is needed to better understand the impacts of neuromodulator implantation on other important outcomes, such as health-related quality of life.