OBJECTIVES: To evaluate patient adherence and persistence to anti-epileptic drug (AED) monotherapy. METHODS: Adult patients (age>18 years) with ≥2 epilepsy diagnosis claims (ICD-9-CM:345) or one epilepsy diagnosis claim and one claim for other convulsion (ICD-9-CM: 780.39) were selected from the U.S. Veterans Health Administration database (01OCT2008-30SEPT2013). Patients were required to have ≥1 AED prescription post-epilepsy diagnosis, and the first AED prescription claim date was designated as the index date. Continuous health plan enrollment 12 months pre- and post-index date was required. Patients were assigned to four monotherapy AED cohorts based on drug class: sodium channel blockers (SCs), gamma-aminobutyric acid analogs (GABAs), synaptic vesicle protein 2A binding (SV2) and multiple mechanisms (MMs). Adherence was assessed using the proportion of days covered (PDC) and persistence was defined as  days to discontinuation with an allowable treatment gap of 45 days without the index AED. Logistic and Cox proportional hazards models were used to compare the results among the cohorts. RESULTS: Patients in the SC cohort had significantly lower baseline Charlson Comorbidity Index scores (1.82), indicating that they were healthier than those in the GAMA (2.08, p<0.001) and SV2 (2.46, p<0.001) cohorts. Patients in the SC cohort were significantly less likely to have a baseline psychiatric disorder (37.6%) than those in the GABA (63.8%, p<0.001) and MM (52.1%, p<0.001) cohorts. Patients treated with GABAs (OR=0.44, p<0.001) and MMs (OR=0.63, p<0.001) were significantly less likely to adhere to their medications (PDC <80%) than those treated with SC. Furthermore, patients treated with GABAs (hazard ratio [HR]=1.74; 95% confidence interval [CI]=1.59-1.90) and MMs (HR=1.18; 95% CI=1.07-1.29) were more likely to discontinue treatment during the follow-up period compared to those in the SC cohort. CONCLUSIONS: Patients treated with Sodium channel blockers are more likely adhere to treatment and have lower discontinuation of AED monotherapy than those treated with GABAs and MMs.