AIMS: Overall, 40% of patients with atrial fibrillation are asymptomatic. The usefulness and cost-effectiveness of atrial fibrillation screening programmes are debated. We evaluated whether an atrial fibrillation screening programme with a handheld electrocardiogram (ECG) machine in a population-wide cohort has a high screening yield and is cost-effective. METHODS: We used a Markov-model based modelling analysis on 1000 hypothetical individuals who matched the Belgian Heart Rhythm Week screening programme. Subgroup analyses of subjects ≥65 and ≥75 years old were performed. Screening was performed with one-lead ECG handheld machine Omron® HeartScan HCG-801. RESULTS: In both overall population and subgroups, the use of the screening procedure diagnosed a consistently higher number of diagnosed atrial fibrillation than not screening. In the base-case scenario, the screening procedure resulted in 106.6 more atrial fibrillation patient-years, resulting in three fewer strokes, 10 more life years and five more quality-adjusted life years (QALYs). The number needed-to-screen (NNS) to avoid one stroke was 361. In subjects ≥65 years old, we found 80.8 more atrial fibrillation patient-years, resulting in three fewer strokes, four more life-years and five more QALYs. The NNS to avoid one stroke was 354. Similar results were obtained in subjects ≥75 years old, with a NNS to avoid one stroke of 371. In the overall population, the incremental cost-effectiveness ratio for any gained QALY showed that the screening procedure was cost-effective in all groups. CONCLUSIONS: In a population-wide screening cohort, the use of a handheld ECG machine to identify subjects with newly diagnosed atrial fibrillation was cost-effective in the general population, as well as in subjects ≥65 and subjects ≥75 years old.