The National Institute for Health and Clinical Excellence (NICE) invited the manufacturer of dronedarone (Multaq®, Sanofi-Aventis Limited, UK) to submit evidence on the clinical and cost effectiveness of the anti-arrhythmic drug (AAD) for the treatment of atrial fibrillation (AF) and atrial flutter, as part of the Institute's single technology appraisal (STA) process. The Centre for Reviews and Dissemination and the Centre for Health Economics, both at the University of York, were commissioned to act as the independent Evidence Review Group (ERG). This article provides a description of the company submission, the ERG review and NICE's subsequent decisions regarding the use of dronedarone within the UK NHS. The ERG review comprised a critique of the submitted evidence on the clinical effectiveness and cost effectiveness of dronedarone. The ERG examined the search strategy used to obtain relevant evidence, the selection of studies included in the assessment, outcome measures chosen and statistical methods employed. The ERG also validated the manufacturer's decision analytic model and used it to explore the robustness of the cost-effectiveness results to key assumptions. The main clinical effectiveness evidence supporting the use of dronedarone as a treatment for AF came from four randomized controlled trials. These trials were compared with a broader set of trials examining the effectiveness of other AADs for AF: amiodarone, sotalol and class 1c agents (flecainide and propafenone). The evidence suggested that all AADs decreased the recurrence of AF but dronedarone had the smallest effect. A mixed treatment comparison analysis of the trials showed that dronedarone was associated with a lower risk of all-cause mortality than other AADs, but this was highly uncertain. There was limited evidence to assess the effect of dronedarone on stroke, and no statistically significant differences between dronedarone and other AADs were found for treatment discontinuation. From the evidence presented by the manufacturer, dronedarone appeared highly cost effective in each of the population groups examined compared with using standard baseline therapy alone as first-line treatment, or compared with sotalol or amiodarone as first-line AAD, with incremental cost-effectiveness ratios (ICERs) well below £20,000 per QALY gained. The ICER for dronedarone relative to class 1c agents was around £19,000 per QALY. Although the evidence presented by the manufacturer indicated that dronedarone was cost effective, the estimates of treatment effect relative to other AADs and safety in the longer term were highly uncertain. The NICE Appraisal Committee in its preliminary guidance did not recommend the use of dronedarone for AF. However, following the response from a large number of consultees and commentators, NICE revised its preliminary guidance to allow the use of the drug in a specific subgroup of AF patients with additional cardiovascular risk factors.