Clinical Study Data Request Registered Users, Please Login

Proposal 1428

Title of the Proposed Research

Estimating absolute treatment effect of dabigatran 110 mg bid vs. 150 mg bid (vs. warfarin) for individual patients with atrial fibrillation.

Lead Researcher

Frank L.J. Visseren


University Medical Center Utrecht, Utrecht, the Netherlands

Funding Source

Research grant from the Netherlands Organisation for Health Research and Development (ZonMw), grant nr. 836011027, titled: "Identifying the right patient to treat, by estimating absolute treatment effect for individual patients based on randomised clinical trial data".

Potential Conflicts of Interest


Data Sharing Agreement Date

18 July 2016

Lay Summary


  • Atrial fibrillation (AF) is the most common heart rhythm disorder. Risk of AF increases strongly with age.
  • Patients with AF have a nearly fivefold increased risk of stroke.
  • Vitamin K antagonists (VKAs) are highly effective medicines by reducing the number of strokes by approximately 60%.
  • Direct oral anticoagulants (DOACs) potentially address some of the disadvantages of VKA’s as they are easier to use and cause less bleeding.
  • The RE-LY study showed that dabigatran (DOAC) reduces the number of strokes/ systemic embolisms with 9% for the lower-dose drug and 34% for the higher-dose drug compared to warfarin (VKA) in patients with AF.

Improving Patient Care

Studies often present one average treatment effect for all patients. However, treatment effect of a drug will likely differ between individuals. Patient characteristics (sex, age, smoking etc.) may influence the benefit a patient has from DOACs. Estimation of one treatment effect for all is suboptimal, especially since patient characteristics may affect the risk of stroke/ systemic embolism and the risk of bleeding differently. Individual treatment effect estimation in the RE-LY study has the potential to identify patients who benefit most from dabigatran. It is reasonable to assume that some patients at high risk of bleeding will benefit most from the lowest dose of dabigatran, while those at high risk of stroke/ systemic embolism are more likely to benefit from the highest dose. Possibly, some patients benefit more from VKA therapy. This targeted approach reduces costs and decreases the number of side effects caused by therapy.


  • To develop and validate a model for prediction of the effect of dabigatran high dose versus low dose on survival free from stroke/ systemic embolism and bleeding in individual patients with AF in the RE-LY trial and the RELY-able study.
  • To develop and validate a model for prediction of the effect of dabigatran versus VKAs on survival free from stroke/ systemic embolism and bleeding in individual patients with AF in the RE-LY trial.

How the research will be conducted

Prediction models will be developed in 18,113 patients with AF from the RE-LY study to predict a patient’s expected life-years free from stroke/ systemic embolism and those free from bleeding with dabigatran or VKA treatment. We will use a method that takes into account that patients with AF often die of other causes than stroke/ systemic embolism. Predictors in the model are simple patient characteristics, and they will be selected from previous published articles. The treatment a patient receives is added to the model as one of the predictors. The prediction model for the effect of dabigatran high versus low dose will be validated for longer duration of follow-up in 5,851 patients in the RELY-able study. The effect of treatment will be expressed as a gain in life years without stroke/ systemic embolism or bleeding for the individual patient.

Study Data Provided

3:13 PM 8/26/2016
BI-1160.26: Randomized Evaluation of Long Term Anticoagulant Therapy (RE-LY) With Dabigatran Etexilate
BI-1160.71: RELY-ABLE Long Term Multi-center Extension of Dabigatran Treatment in Patients With Atrial Fibrillation Who Completed RE-LY Trial

Statistical Analysis Plan

Publication Citation

The publication citation will be added after the research is published.

Summary Results

Results summary or link will be posted when available.