OBJECTIVES

This analysis examines the healthcare resource utilization (HCRU), including medication burden before and after initiation of perampanel in epilepsy patients compared to matched controls.

METHODS

Patients with at least two inpatient or outpatient diagnoses of epilepsy (ICD-10: G40%) between 01/01/2010 and 12/31/2015 were identified in Swedish national registry data. All patients were observed for 12 months before and after their first antiepileptic drug (AED) prescription documented between 01/01/2011-31/12/2014; and were divided into those who received perampanel during the observational period versus any other AEDs (index date = first prescription of perampanel/any other AED during index period). A propensity score matching was applied to match one perampanel user with five non-perampanel users, based on their baseline characteristics (age, sex, comorbidities, epilepsy duration, number of previous AEDs, year of index AED).

RESULTS

After matching, 299 perampanel users and 1,495 controls were identified (mean age 36.6 years; 47.5% female). In perampanel users, number of patients with at least one epilepsy-related hospitalization was reduced by 62.5% (pre-index: 66.9% vs. post-index: 25.1%), in non-perampanel users by 73.0% (pre-index: 45.2% vs. post-index: 12.2%).

During the pre-index period, perampanel patients used a mean of 3.3 different AEDs per patient year (py), compared to 1.1 in non-perampanel patients (p<0.001), including the index AED. While average number of different AEDs prescribed in the follow-up period was still higher in perampanel users, a decrease in mean number of prescribed AEDs in comparison to pre-index ([‑]0.2 AEDs/py) was observed for perampanel, whereas for non-perampanel users an increase of ((+)1.3 AEDs/py) could be seen (p<0.001).

CONCLUSIONS

Perampanel users substantially differ in their characteristics from users of other AEDs, with higher HCRU and medication burden, as perampanel is approved as an add-on therapy in a patients with partial-onset seizures. However, after starting on perampanel they experienced a greater reduction in post-index AED use.