Abstracts

Rationale: Non-adherence to antiepileptic drugs (AEDs) can be devastating for youth with epilepsy, including increased risk of seizures, poor health-related quality of life, higher health care utilization and costs, and even sudden unexpected death. Adolescents with epilepsy represent an especially high-risk population, as autonomy increases and parental supervision decreases. This combined with difficulties with executive functioning leads to poor adherence during adolescence. However, no studies have examined adherence over time using objective electronically-monitored adherence in this population. The aims of this study are to identify patterns and predictors of adherence in adolescents with epilepsy over one year. Methods: Participants included 48 adolescents aged 13-17 years (M=14.8+1.4; 72.3% White Non-Hispanic, 68% females) with epilepsy and their caregivers. Caregivers completed a background questionnaire. Adolescents completed the Allocation of Treatment Responsibility Scale, Barriers subscale of the Pediatric Epilepsy Medication Self-Management Questionnaire, Epilepsy Knowledge Questionnaire, and Parental Environment Questionnaire. The Pediatric Epilepsy Side Effects Questionnaire was completed jointly, and electronically-monitored adherence was obtained via MEMS TrackCaps. Data were coded 1 = adherent, 0 = non-adherent. A time-varying effect model was used to estimate the odds of being adherent over one year. Predictors of longitudinal adherence included baseline age, time since diagnosis, family socioeconomic status (SES), side effects, allocation of treatment responsibility, epilepsy knowledge, barriers, family conflict, and family involvement. Results: The odds of being adherent increased with time, but within 3 weeks of monitoring, the odds of being adherent steadily decreased. Indeed, as time increased, patients were less likely to be adherent (see Figure 1). The odds of being adherent over time were higher with younger age, longer time since diagnosis, higher family SES, less adolescent-reported responsibility for treatment,  more epilepsy knowledge,  fewer barriers, decreased caregiver conflict, and increased family involvement. Notably, side effects reported at baseline were not associated with the odds of being adherent over time. Conclusions: This is the first longitudinal study of adolescent adherence in epilepsy using an objective measure of adherence. Contributions to better adherence include individual (age, time since diagnosis, knowledge about epilepsy) and family factors (income, responsibility for adherence, barriers, conflict, and involvement). Modifiable targets for intervention include knowledge about epilepsy and several family-focused variables, including sharing epilepsy management responsibilities and reducing family conflict. Future research should focus on the development and testing of such interventions.  Funding: This work was supported by the Charlotte M. Schmidlapp Women’s Scholar Award through Cincinnati Children’s Hospital Center awarded to Avani C. Modi.