Rationale: Selection of an antiepileptic drug (AED) for patients with epilepsy (PWE) that is well-tolerated and provides effective seizure control is typically achieved through a complex process of trial-and-error, which can last up to several years. This complexity may be amplified for PWE who have comorbid psychiatric disease or substance use disorder (SUD); however, data so far are limited. In this study, we sought to ascertain the likelihood of AED treatment success in these patient populations. Methods: Longitudinal US claims data aggregated from multiple channels were obtained from IQVIA and searched for patients with ≥1 epilepsy or ≥2 seizure diagnoses and an AED prescription. Treatment success was determined for all identified patients and for specific subpopulations with comorbid psychiatric disease (anxiety, schizophrenia, depression, bipolar disorder) or SUD (substance, opioid or cannabis abuse). Subpopulations of patients with comorbid hypertension or diabetes were included as chronic disease control groups. Results were stratified into three age groups (16–34, 35–64 and ≥65 years). Treatment success was determined using a proxy, treatment stability, defined as continuous treatment with the AED for ≥12 months without addition of another AED (dose changes allowed). Patients who did not have ≥12 months’ data from treatment initiation were excluded. Comorbidities were identified via searches on the Observational Health Data Sciences and Informatics (OHDSI) database (
http://www.ohdsi.org/web/atlas/#/search) using Systematized Nomenclature of Medicine–Clinical Terms (SNOMED–CT) concepts, with all child concepts and mapped concepts from ICD9 and ICD10 terminologies. Results: Overall, 178,083 PWE were included in the dataset. Treatment success rate was 50.8% for all PWE, and 36.3%, 44.9%, and 59.2% for patients aged 16–34 (n=48,652), 35–64 (n=90,830) and ≥65 years (n=38,601), respectively (Table). Compared with success rates in the corresponding age groups of all PWE, presence of hypertension or diabetes did not affect the probability of success, whereas presence of any psychiatric comorbidity or SUD reduced the probability. The extent of the reduction was relatively similar in the three age groups in most subpopulations with psychiatric or SUD comorbidity, with the exception of suicidal behaviors, substance abuse and opioid abuse. In these subpopulations, there were age-related differences, with the probability of treatment success lowest for the youngest group of patients. Conclusions: Results suggest that psychiatric disorders and SUD may reduce the likelihood of AED treatment success among PWE. These findings require further research, particularly to ascertain whether treatment stability, the proxy for treatment success, corresponds adequately to actual successful clinical outcomes. Furthermore, the retrospective nature of the study, and small sample sizes in the age groups for some subpopulations further limit interpretation and generalizability of the results. Funding: UCB Pharma-sponsored