The costs of atrial fibrillation (AF) are linked to the general cost of managing AF patients in different health-care systems, as well as the cost of managing AF-related complications (e.g. hospitalizations and long-term complications, such as stroke). In addition, indirect medical costs, such as care for patients who do not recuperate fully from a vascular event, and non-medical costs such as loss of work force add to the costs of AF. All estimations for cost of AF and cost of AF therapy are based on assumptions and markedly influenced by these cost determinants. This urges for extreme caution not to take cost estimates at their absolute values. In fact, even relative comparisons between interventions may have different consequences in terms of direct and indirect costs in different health-care settings. While newer therapeutic options appear to increase the cost of AF management, newer antithrombotic substances and adequate rhythm control therapy also carry the promise of preventing the two major drivers of AF-related cost, hospitalizations and AF-related complications. Formal assessment of the cost of AF requires adjustment to local practice, and more data are clearly needed especially from primary care to better estimate the 'real' cost impact of AF.