Abstract Introduction Oral anticoagulants (NOACs) are known to have a better safety profile than Vitamin K antagonists, but their metabolism largely depends on the P-glycoprotein and CYP3A4. All drugs inhibiting those proteins may affect NOACs’ bioavailability, by increasing the risk of drug interactions, particularly the risk of bleeding. Purpose The objective was to describe the pattern of polypharmacy in a large cohort of NOAC users, at baseline and within one year after starting anticoagulant treatment. Materials and Methods All persons residing in the catchment areas of Caserta LHU during the years 2012–2020 were considered. From the source population, all patients with at least 1 year of database history and receiving at least one NOAC dispensing during the observation period were identified. Of these, incident (i.e. no NOAC dispensing within one year prior to the first dispensing date, Index Date) NOAC users were included into the cohort and represented the majority of the patients enrolled (97,4%). The comparison of the distribution of interacting drugs number dispensed Pre-ID (within three months before the Index date) and Post-ID (within 12 months after Index date) was explored stratified by high/low dose of anticoagulant. 20.491 incident NOAC users were identified from Caserta LHU and 16,367 (79.9%) NOAC incident users with at least one year of follow-up after ID were included in the analysis. NVAF was the main indication for use of the study cohort (64.3%) and almost 60% of incident NOAC users treated for NVAF received high-dose NOAC. The frequency of incident NOAC users with at least one hospitalization for major bleeding within one year after ID was calculated and stratified by high/low dose of NOACs and the number of drug interactions. A statistically significant test result was considered if p ≤ 0.05. Results The number of interacting drugs dispensed within one year after ID increased from pre-ID to post-ID. 2,077 (12.7%) incident NOAC users without any interacting drugs within 3 months pre-ID received 1 interacting drug within one year after ID. 11.3% and 8.4% of incident NOAC users increased from 1 interacting drug (Pre-ID) to 2 interacting drugs (Post-ID) and from 2 interacting drugs (Pre-ID) to 3-5 interacting drugs (Post-ID) (Table 1). The proportion of NOAC users with a major bleeding increased with the increasing number of interacting drugs, reaching 2.9% and 2.1% among low-dose and high-dose NOAC users with more than 9 interacting drugs (Table 2). Conclusions The results of the current analysis suggest that independently from the type and the dose of NOAC, the number of interacting drugs prescribed with NOACs tends to increase after the first administration of NOACs. Hemorrhagic events are more related to the number of interacting drugs instead of dose of anticoagulant.Table 1 Table 2