Background Few studies have investigated whether and which anxiety and affective disorders are associated with the risk of venous thromboembolism. We aimed to examine whether anxiety disorder, depression, and bipolar disorder increase the risk of venous thromboembolism, independent of socioeconomic confounders and each other, in a UK general population. Methods This is a prospective cohort study using UK Biobank. Participants were excluded if they were diagnosed with venous thromboembolism before the baseline assessment, if they were first diagnosed with anxiety disorder, depression, or bipolar disorder thereafter, or if they had missing sociodemographic data. Diagnosed anxiety disorder, depression, and bipolar disorder were ascertained through hospital admission data and incident venous thromboembolism, pulmonary embolism, and deep vein embolism were ascertained through both hospital admission and death certificate data. Hazard ratios were calculated, adjusted for sociodemographic confounders and comorbid anxiety and affective disorders. Results Our main analysis included 455,705 participants, of whom 53.6 % were female, with a mean age (standard deviation) of 56.6 (8.1) years. Anxiety disorder, depression, and bipolar disorder were associated with venous thromboembolism both before and after adjusting for sociodemographic confounders. After adjustment for comorbid anxiety and affective disorders, depression (HR: 2.00; 95 % CI: 1.68–2.38) and bipolar disorder (HR: 2.08; 95 % CI: 1.28–3.37) remained associated with venous thromboembolism, but anxiety did not (HR: 1.17; 95 % CI: 0.88–1.57). Similar results were found for pulmonary embolism and deep vein embolism. Conclusions Depression and bipolar disorder were associated with the risk of venous thromboembolism. Further research is required to understand the mechanism underlying their increased risk.