ObjectiveThis study aimed to assess the treatment compliance, patterns, healthcare resource utilization (HCRU), and costs of anti-epilepsy drugs (AEDs) as the first add-on therapy in patients with epilepsy.MethodsWe conducted a retrospective population-based cohort study using Korean National Health Insurance claims data from 2016 to 2020. Patients with epilepsy who newly received AED add-on therapy were identified and followed for up to 12 months to evaluate persistence, adherence, treatment patterns, HCRU, and costs.ResultsAmong 6,746 patients who initiated AED add-on therapy, 65.5% were persistent on their index AED add-on from the index date until the end of the follow-up period, and the mean persistent time on the index add-on was 307.3 ± 92.3 days. A total of 76.8% patients were adherent, with a medication possession ratio (MPR) ≥80%, and the mean MPR was 88.9 ± 25.4%. Persistence and adherence to the index AED add-on were relatively higher among patients prescribed lamotrigine, levetiracetam, oxcarbazepine, and perampanel than those prescribed carbamazepine, topiramate, or valproate. A total of 41.0% of the patients changed their index AED add-on during the follow-up period. The carbamazepine, topiramate, and valproate groups had higher rates of change than the other AED groups. HCRU and costs tended to be lower in the lamotrigine, levetiracetam, oxcarbazepine, and perampanel groups. Furthermore, perampanel showed the lowest HCRU and costs for all-cause cases as well as the lowest length of stay and outpatient visits for epilepsy-related cases.ConclusionIn this population-based study, the use of lamotrigine, levetiracetam, oxcarbazepine, or perampanel as the first add-on therapy in patients with epilepsy contributed to better treatment compliance and lower HCRU and costs than that of carbamazepine, topiramate, or valproate.