Greater need for treatment optimization in patients with epilepsy initiating adjunctive therapy: Results of a retrospective claims analysis of antiseizure medication drug load in the United States

•A US claims-based study assessed changes in ASM drug load in patients with epilepsy.•Concomitant ASM drug load increased up to new ASM start followed by slight decline.•Concomitant drug load was higher in the branded ASM vs unbranded ASM cohort.•Total ASM drug load (including comparator ASMs) increased yearly from 2016 to 2020. This retrospective, observational study used US claims data to assess changes in antiseizure medication (ASM) drug load for a cohort of patients with epilepsy. Adults (≥18 years) with a diagnosis of epilepsy (ICD-10 code G40.xxx) who started new adjunctive ASM treatment with one of 4 branded (brivaracetam, eslicarbazepine, lacosamide, perampanel) or 4 unbranded (carbamazepine, lamotrigine, levetiracetam, topiramate) ASMs between January 1, 2016 and December 31, 2020 were identified from IBM MarketScan® research databases (primary study population). Patients must have been continuously enrolled 360 days before the start of the new ASM (eligibility period). Follow-up was from the start of new ASM until Day 540 (∼18 months). The primary endpoint was concomitant ASM drug load, which included all ASMs except the new (comparator) ASM. A sensitivity analysis population included adults with epilepsy who were continuously enrolled for ≥ 180 days during at least one calendar year in the study period (2016–2020), whether or not the comparator ASM was new or existing during that period. Total ASM drug load, which included comparator ASM and concomitant ASMs, was assessed in the sensitivity analysis population. In total, 21,332 patients were included in the primary study population, of which 5767 initiated branded ASMs and 15,565 initiated unbranded ASMs. A total of 392,426 patients were included in the sensitivity analysis population during at least one calendar year 2016–2020. Concomitant ASM drug load increased in the 360 days prior to new ASM start and slightly declined thereafter. Mean concomitant ASM drug load for the primary population was 1.6 (SD 1.8) at new ASM start. Concomitant drug load was higher among those starting branded ASM comparators compared to those starting unbranded comparators. Mean total ASM drug load for patients increased over time and was approximately double for patients exposed to branded ASMs (mean range 2.1 to 2.7) compared to that of patients exposed to any unbranded ASM (mean range 1.0 to 1.3). Concomitant ASM drug load increased prior to addition of new ASM, with higher increases observed among patients start