Determinants of oral anticoagulation control in new warfarin patients: analysis using data from Clinical Practice Research Datalink

BACKGROUND: The safety and effectiveness of warfarin therapy depends critically on the quality of anticoagulation control, often assessed using the percentage time in therapeutic International Normalised Ratio (INR) range (TTR). We aimed to identify patient characteristics related to anticoagulation control with warfarin, measured by TTR. METHOD: We carried out a population-based study using the Clinical Practice Research Datalink, including two cohorts of patients starting warfarin after a first diagnosis of atrial fibrillation (AF) or venous thromboembolism (VTE) between 2000 and 2013. We used multivariate mixed regression and logistic regression models to predict the fully-adjusted effect of each predictor variable upon TTR. RESULTS: The study population comprised 29,717 incident AF and 19,113 incident VTE patients who initiated warfarin. In real world clinical practice a minority of patients achieve good anticoagulation control with warfarin (44% AF and 36% VTE patients had TTR≥70%). Poor anticoagulation control driven by subtherapeutic INRs was observed in younger patients (<45years) and in AF patients with increased number of hospitalisations. Poor anticoagulation control driven by sub and/or supratherapeutic INRs was seen in AF and VTE patients current smokers, in patients using medications for pain and in VTE patients with active cancer. CONCLUSION: In a real world clinical practice there is a high amount of unpredictable inter-individual TTR variability and in some patients good anticoagulation control is more challenging than in others. These findings may help to identify patients who will require closer monitoring or innovative strategies to optimise the outcomes of oral anticoagulant therapy.