A clustered randomized trial to IMProve treatment with AntiCoagulanTs in patients with Atrial Fibrillation (IMPACT-AF): design and rationale.


Journal Article

Atrial fibrillation (AF) is common, increasing as the population ages, and a major cause of embolic stroke. While oral anticoagulation (OAC) is highly effective at preventing stroke in patients with AF, it continues to be underused in eligible patients worldwide. The objective of this prospective, cluster randomized controlled trial (IMPACT-AF; ClinicalTrials.gov #NCT02082548) is to determine whether a comprehensive customized intervention will increase the rate and persistence of use of OAC in patients with AF. IMPACT-AF will be conducted in approximately 50 centers in 5 low- to middle-income countries. Before randomization, sites within countries will be paired to match in size, practice type and baseline rate of OAC use. Site pairs will be randomized to intervention versus control. In total, 40 to 70 patients with AF and at least 2 CHA2DS2-VASc risk factors will be enrolled at each site using a consecutive enrollment strategy, with the goal of capturing actual practice patterns. We aim for patients with a new diagnosis of AF to comprise at least 30% of the study cohort. Assuming an average baseline OAC use of 60% and a post-intervention use of 70% with a post-control rate of 60%, there will be roughly 94-98% power with 25 clusters per group (intracluster correlation coefficient of 0.02). While this trial focuses on improving treatment use and reducing preventable strokes, we also aim to better understand the reasons for OAC underuse. This will improve the intervention with the goal of creating educational recommendations to improve care for patients with AF.

Full Text

Duke Authors

Cited Authors

  • Rao, MP; Ciobanu, AO; Lopes, RD; Fox, KA; Xian, Y; Pokorney, SD; Al-Khalidi, HR; Jiang, J; Kamath, DY; Berwanger, O; Xavier, D; Bahit, CM; Tajer, C; Vinereanu, D; Huo, Y; Granger, CB

Published Date

  • June 2016

Published In

Volume / Issue

  • 176 /

Start / End Page

  • 107 - 113

PubMed ID

  • 27264227

Pubmed Central ID

  • 27264227

Electronic International Standard Serial Number (EISSN)

  • 1097-6744

Digital Object Identifier (DOI)

  • 10.1016/j.ahj.2016.03.011


  • eng

Conference Location

  • United States