Objective: To determine the use of different anticoagulation therapies in rural Western Australia; to establish whether remoteness from health care services affects the choice of anticoagulation therapy; to gather preliminary data on anticoagulation therapy safety and efficacy. Design: Retrospective cohort study of patients hospitalised with a principal diagnosis of atrial fibrillation/flutter (AF) or venous thromboembolism (VTE) during 2014-2015. Setting: Four hospitals serving two-thirds of the rural population of Western Australia. Participants: 609 patients with an indication for anticoagulation therapy recorded in their hospital discharge summary for index admission. Main outcome measures: Prescribing rates of anticoagulation therapies by indication for anticoagulation and distance of patient residence from their hospital. The primary safety outcome was re-hospitalisation with a major or clinically relevant non-major bleeding event; the primary lack-of-efficacy outcome was re-hospitalisation for a thromboembolic event. Results: The overall rates of prescription of NOACs and warfarin were similar (34% v 33%). A NOAC was prescribed more often than warfarin for patients with AF (56.0% v 42.2% of those who received an anticoagulant; P < 0.001), but less often for patients with VTE (29% v 48%; P < 0.001). Warfarin was prescribed for 38% of patients who lived locally, a NOAC for 31% (P = 0.013); for non-local patients, the respective proportions were 29% and 36% (P = 0.08). 69% of patients with AF and a CHA2DS2-VASc score ≥ 1 were prescribed anticoagulation therapy. Patients treated with NOACs had fewer bleeding events than patients treated with warfarin (nine events [4%] v 20 events [10%]; P = 0.027). Conclusions: In rural WA, about one-third of patients with an indication for anticoagulation therapy receive NOACs, but one-third of patients with AF and at risk of stroke received no anticoagulant therapy, and may benefit from NOAC therapy.