Objective To report acceptability, feasibility, and preliminary efficacy from a randomized controlled trial of a family-tailored adherence intervention (AI) targeting nonadherence to antiepileptic drugs in pediatric new-onset epilepsy. Method 30 children with new-onset epilepsy (7.2 ± 3.1 years old, 47% male) and their caregivers participated. At baseline, participants were given adherence electronic monitors. After a 1-month run-in period, participants with good adherence (≥90%) were monitored. Participants with adherence <90% were randomized to the AI or Treatment-As-Usual (TAU) group. The AI group received four adherence promotion intervention sessions over >2 months. Follow-up adherence data were collected. Results 8 families were randomized (AI, n = 4; TAU, n = 4). Families perceived AI to be feasible and acceptable. Preliminary results demonstrated that the AI group had improved adherence from baseline to post-test. Conclusions A family-tailored AI appears promising and needs to be tested with a larger pediatric epilepsy sample.