To determine incidence rate and predictors of venous thromboembolism (VTE) in a population-based cohort with ANCA-associated vasculitis (AAV).The study comprised 325 patients diagnosed with AAV from 1997-2016. All cases of VTE from prior to vasculitis diagnosis to the end of the study period were identified. The Birmingham Vasculitis Activity Score (BVAS) was used to assess disease activity at diagnosis. Venous thromboembolisms occurring in a period beginning three months prior to AAV diagnosis were considered to be AAV-related. The standardized incidence ratio (SIR) and 95% confidence intervals (CI) of VTE were calculated using the incidence rate in the general population.Fifty-nine patients (18%) suffered 69 VTE events. Of these, 48 (81%) suffered AAV-related VTE [deep vein thrombosis (DVT, n = 23), pulmonary embolism (PE, n = 18), and other (n = 9)]. The incidence rate of AAV-related VTE was 2.4/100 person-years (95% CI 1.7-3.0) during 2039 person-years of follow-up. The incidence during the first three months post-AAV diagnosis was 20.4/100 person-years (95% CI 11.5-29.4), decreasing to 8.9 (95% CI 0.2-17.6) and 1.5 (95% CI 0.0-3.5) in months 4-6 and months 7-12 post-AAV diagnosis, respectively. The SIR was 34.2 (95% CI 20.2-48.1) for DVT and 10.4 (95% CI 5.6-15.1) for PE. In multivariate Cox-regression analyses, only age and BVAS were predictive of VTE.The incidence rate and SIR of AAV-related VTE is high, and higher early in the course of the disease. Vasculitis activity and age are positively associated with VTE.