(963) Risk factors associated with antiseizure medication discontinuation during pregnancy

Objective: Evaluate factors associated with discontinuation of antiseizure medication (ASM) by the first trimester of pregnancy.

Study Design: We constructed a retrospective cohort of births (2007-2019) among Tennessee Medicaid pregnant patients with enrollment from 90 days before conception through delivery. Data included healthcare claims, prescription fills, hospital discharge and birth certificates. We included births ≥ 20 weeks’ gestation among patients aged 15-44 years on ASM monotherapy. Among patients with filled prescriptions for ASMs before conception or in the first trimester, we defined discontinuers as those with no ASM prescription filled after the first trimester, and continuous users as those who filled prescriptions after the first trimester. We used logistic regression to model the association between clinical factors and ASM discontinuation overall and among patients with an epilepsy diagnosis, adjusting for relevant confounders.

Results: Of 5279 pregnant patients with ASM exposure before conception or in the first trimester, 54% discontinued by the first trimester (Table 1). Factors associated with ASM discontinuation included the absence of an epilepsy diagnosis, the presence of a psychiatric or other diagnosis, younger age, nulliparity, earlier study delivery year, living in a rural area, and use of older ASM class (Table 2). While 26% of pregnant patients with ASM use had an epilepsy diagnosis, 35% had no diagnosis traditionally associated with ASM use. Among patients with epilepsy, 29% discontinued ASMs by the first trimester. In this group, older ASM class and earlier study delivery year were associated with higher odds of discontinuation. Non-Hispanic Black individuals had lower odds of discontinuation than non-Hispanic White individuals (Table 2).

Conclusion: For patients with indications for ASM use, national guidelines recommend continuing most ASMs during pregnancy or discontinuing with clinician guidance. The majority of pregnant patients in our cohort, including those with epilepsy, discontinued use around pregnancy diagnosis.