Abstracts

Rationale: Adverse events (AEs) are frequently encountered in the course of antiepileptic drug (AED) treatment for partial onset seizures (POS). The objective of this study was to assess the impact of AEs on healthcare resource utilization (HRU) and costs in patients with POS. Methods: Adult patients (aged 18-65) with 2 POS diagnoses [ICD-9 codes 345.4, 345.5, or 345.7] receiving oral AED monotherapy for 30 consecutive days were identified in Truven Health Analytics MarketScan databases (01/2008-12/2011). The index date was randomly selected among all AED initiation dates in monotherapy meeting the selection criteria. Patients were censored from the study at the end of continuous eligibility, if they had a therapy gap of 30 days, or at the initiation of a combination therapy, whichever occurred first. The study period was divided into 90-day periods, quarters , where the AE status of a patient was allowed to change (i.e., AE status could change across quarters). All-cause and epilepsy/seizure-related HRU and costs (measured from a payer s perspective; adjusted for inflation and reported in March 2013 USD) incurred in quarters where patients had 1 AE were compared to those incurred in quarters where patients did not have any AEs using generalized linear regression and two-part regression models. Multivariate regression models adjusted for potential confounding factors and random effects adjusted for patients with multiple quarters of observation. Results: A total of 31,974 patients met the inclusion criteria, for a total of 76,994 patient-quarters. Patients average age was 42.1 years and 59.4% were female. The most common AEs were worsening of seizures (19.0%), depression/mood change (11.8%), sleep disturbance/insomnia (8.6%), headache (8.4%), and asthenia/fatigue/tiredness (8.1%). Patients incurred significantly higher HRU during quarters with incident AEs compared to quarters without incident AEs, with higher incidence rates reported for inpatient (IP) admissions (0.121 vs. 0.011; incidence rate ratios [IRRs] = 8.54), IP days (0.785 vs. 0.048; IRR=11.69), emergency room visits (0.299 vs. 0.045; IRR=5.34), and outpatient visits (6.544 vs. 2.473; IRR=1.97) (all p<0.01). Similarly, patients incurred significantly higher all-cause and epilepsy/seizure-related costs during quarters with incident AEs compared to those without AEs; the average all-cause medical costs were $6,646 and $1,207 for quarter with and without AEs, respectively; incremental all-cause medical costs were $5,119 and incremental pharmacy costs were $248 (all p<0.01). Medical costs differences were mainly driven by IP ($2,113) and outpatient ($1,910) utilization (all p<0.01). Among the various incident AEs, epilepsy/seizure-related costs accounted for 67% of total IP and 26% of total outpatient cost differences. Conclusions: The results of this study suggest that AEs, including lack of seizure control, are associated with a substantial economic burden for patients with POS, and highlight the importance of AEDs with favorable tolerability/AE profiles.