Abstract This article reviews the current sample size requirements for studies evaluating the efficacy of atrial therapies. Sample sizes for several study designs and endpoints are presented. However, emphasis is given to studies conducted in patients with implantable devices in the light of new available data from recent trials involving such patients. Studies utilizing mortality require >5,000 patients followed for at least 2 years. Symptomatic episode recurrence represents a possible alternate endpoint but still requires >400 patients followed for 1 year. Parallel design studies for device atrial fibrillation (AF) burden and frequency endpoints were found to be of little value because they would require thousands of patients. In contrast, cross-over studies utilizing device AF burden or AF frequency endpoints provide the best option in terms of sample size and study duration. A 30% therapy efficacy (reduction in AF burden or frequency) can be detected with just a little over 100 patients followed for 6 months. Therefore, cross-over studies with device-based endpoints are a viable option to evaluate and screen possible new rhythm control treatments before evaluating their effect on clinical outcome variables that may require larger trials with longer follow-up times.