Following at least 3 months of anticoagulation therapy after a first unprovoked venous thromboembolism (VTE), there is uncertainty about the duration of therapy. Further anticoagulation therapy reduces the risk of having a potentially fatal recurrent VTE but at the expense of a higher risk of bleeding, which can also be fatal.
An economic evaluation sought to estimate the long-term cost-effectiveness of using a decision rule for restarting anticoagulation therapy vs. no extension of therapy in patients based on their risk of a further unprovoked VTE.
A Markov patient-level simulation model was developed, which adopted a lifetime time horizon with monthly time cycles and was from a UK National Health Service (NHS)/Personal Social Services (PSS) perspective.
Base-case model results suggest that treating patients with a predicted 1 year VTE risk of 17.5% or higher may be cost-effective if decision makers are willing to pay up to £20 000 per quality adjusted life year (QALY) gained. However, probabilistic sensitivity analysis shows that the model was highly sensitive to overall parameter uncertainty and caution is warranted in selecting the optimal decision rule on cost-effectiveness grounds. Univariate sensitivity analyses indicate variables such as anticoagulation therapy disutility and mortality risks were very influential in driving model results.
This represents the first economic model to consider the use of a decision rule for restarting therapy for unprovoked VTE patients. Better data are required to predict long-term bleeding risks during therapy in this patient group.