Individualizing treatment decisions for patients with venous thromboembolism : Risk prediction and patient preferences

Abstract

Anticoagulation is the cornerstone of treatment for venous thromboembolism (VTE; deep vein thrombosis and pulmonary embolism). After an initial treatment of at least 3 months, a decision must be made to stop treatment or to continue indefinitely to prevent VTE recurrence. Although highly effective, anticoagulation carries a risk of major bleeding. Therefore, guidelines recommend basing the decision on a careful weighing of risks of recurrent VTE and bleeding, taking into account patient’s preference. However, deciding on treatment for individual patients is challenging as risks of recurrence and bleeding are heterogeneous. The first part of this thesis focuses on risk prediction. Previously published risk scores for recurrent VTE and bleeding were found not to be suitable for clinical practice because of methodological shortcomings, lack of independent external validation and insufficient predictive performance. Hence, new models for the prediction of recurrent VTE and clinically relevant bleeding within 5 years after the initial treatment for VTE were developed. These models will be made available as online calculator, which can also be used to estimate the effect of extended treatment with different anticoagulants for individual patients. Furthermore, different methods of estimating risk of bleeding in patients with cancer-associated thrombosis were compared. The second part of this thesis describes different perspectives on the decision to stop or continue anticoagulation using the results of a survey of physicians and interviews with patients. Furthermore, a randomized trial was conducted to study the effect of an educational app on VTE and anticoagulation on patient satisfaction.