OBJECTIVES:  Dabigatran and Rivaroxaban are Novel Oral Anticoagulants (NOACs) which have shown better or similar efficacy compared to warfarin in clinical trials. The objective of the study was to compare annual all-cause costs for NOACs vs warfarin and to identify the major cost drivers based on the clinical subgroups. METHODS: A retrospective cohort study utilized de-identified data from Optum® Clinformatics™ Data Mart (Optum Insight, Eden Prairie, MN) (Jan 1, 2010 and Dec 31, 2012). Adult patients with ≥ 1 diagnosis of atrial fibrillation or flutter (ICD-9 427.31/32), >1 prescription of NOACs or warfarin, 6-months pre-index continuous enrollment and CHADSVAS_C score >1 were included. Post-index costs and Healthcare Resource Utilization (HCRU) was calculated using GLM model with gamma distribution. The total cost, sum of medical (inpatient + outpatient) and drug cost was adjusted for inflation to 2016. Unadjusted costs were analyzed by subgroups (Gender, Region, Age, Insurance Type, CHADSVAS_C Score, and Charlson’s Co-morbidity Index [CCI]). RESULTS:  A total of 3287 and 1770 patients were included for warfarin and NOAC cohort respectively. The NOAC users were slightly older (66 vs 65 years) and more severe based on the CHADSVAS_C and CCI. The annual drug cost for NOAC was greater ($4998 vs $331, P <0.05), but was offset by lower medical cost ($22,134 vs $31,400, P <0.05). The total annual cost for NOACs ($26,803 vs $32,157) was significantly lower than warfarin (P>0.05). The frequency of ER visits was similar (14 vs 13) but office visits (76 vs 49) were significantly higher for warfarin vs NOACs (P <0.05). Based on subgroup analysis, CCI and age < 65 were major cost drivers of medical cost while Northeast region and CHADSVAS_C score > 2 were major drivers of the drug costs. CONCLUSIONS:  Annual total cost and HCRU for patients prescribed the NOACs was lower compared to warfarin.