Abstract Aims Vitamin K antagonists (VKAs) are suspected of causing aortic valve calcification (AVC). The objective of this study was to clarify whether patients undergoing VKA treatment have increased AVC scores compared to patients treated with new oral anticoagulants (NOACs) and patients who never have been treated with VKA/NOAC. Methods and results We included participants from the population-based DANCAVAS trial (n = 15 048). Information on confounders was collected, and the AVC scores were measured on non-contrast computed tomography scans. The participants’ medication data, including VKA and NOAC data, were collected from the Danish National Health Service Prescription Database. The final population consisted of 14 604 participants (67.4 years, 95% men) of whom 873 had been treated with VKA and 602 with NOAC. The association between AVC score and duration of anticoagulant use was investigated in an adjusted zero-inflated negative binomial regression model. For every year treated with VKA, the AVC score increased, on average, by 6% [ratio of expected counts (RECs) = 1.06; 95% confidence interval (CI) 1.02–1.10] compared to non-use. The results were consistent in sensitivity analyses excluding patients with known cardiovascular disease and statin users (REC = 1.07; 95% CI 1.02–1.11 and REC = 1.10; 95% CI 1.03–1.17, respectively). NOAC treatment was not significantly associated with AVC score in any of the corresponding models (REC = 1.03, 1.02, and 0.96). Conclusion Compared to no treatment with anticoagulants, VKA use was associated with increased AVC score, while a similar association could not be established for NOAC.