Abstract Background Adherence to oral anticoagulants is essential for stroke prevention in patients with atrial fibrillation (AF). This study aims to test the feasibility and effectiveness of the smartphone app-based intervention to improve drug adherence in patients with AF. Methods This prospective, randomised, open-label, multicentre trial (ADHERE-APP) enrolled a total of 502 patients at 9 university hospitals in South Korea from November 2021 to January 2023. Patients with AF aged ≥19 years requiring oral anticoagulation,and were treated with edoxaban were enrolled. Patients were randomized in a 1:1 ratio to the conventional treatment group and the app conditional feedback group. Patients in the app conditional feedback group used a smartphone application that pushed alarms for taking medication and measuring blood pressure (BP) and heart rate (HR) at specific times of the day. In addition to medication alarms, the habit of measuring one’s BP and HR could improve drug adherence by increasing self-awareness of the disease. The automatic BP machine was connected to the smartphone via Bluetooth and the measured BP and HR were automatically updated on the smartphone app. The primary endpoint was adherence to edoxaban as measured by pill count at 12 and 24 weeks. In addition, we evaluated the differential effect of the smartphone application by age on the primary endpoint. Results Among the 502 patients, the mean age was 66.0±10.0 years and 67.7% were male. Of the 372 patients who completed 12-weeks trial, edoxaban adherence at 12 weeks was high in both arms, with no difference in the primary outcome (96.5% [92.5%-100.0%] vs. 98.0% [93.2%-99.5%], P = 0.488). or in secondary outcomes, including the proportion of patients achieving adherence ≥ 80% or 95%. In patients aged >65 years, adherence to edoxaban was higher in the app conditional feedback group than in the conventional treatment group (95.8% [93.0%-99.5%] vs. 98.0% [95.8%-99.5%], P = 0.049). Conclusions Overall adherence to edoxaban was much higher than expected in patients with AF. There was no difference in edoxaban adherence between the app conditional feedback group and the conventional treatment group, however, in older patients, app feedback increased medication adherence.