Timp JF, Braekkan SK, Lijfering WM, et al. - Data from two population-based cohorts of individuals with first venous thromboembolism (VTE) from the Netherlands (n = 3,750) and Norway (n = 663) were analyzed by researchers to inform the development of a prediction model for all individuals with first VTE, either provoked or unprovoked. A total of 507 recurrences transpired during 19,201 person-years of follow-up. They found that, with respect to the duration of anticoagulant treatment, a significant proportion of patients with first venous thrombosis are currently misclassified. The developed prediction model applies to patients with provoked or unprovoked first VTE—with the exception of patients with (a history of) cancer—allows refined risk stratification, and is user-friendly. The investigators cautioned, however, that for optimal individualized treatment, a management study in which bleeding risks are also considered is necessary.