Association of stroke and bleed events in non-valvular atrial fibrillation patients with direct oral anticoagulant prescriptions in NHS England between 2013 and 2016

Prescription of direct oral anticoagulants (DOAC) compared to warfarin for treating atrial fibrillation patients have increased substantially since their introduction in the England’s National Health Service. Assessment of the risk of strokes and bleeds in relation to the large-scale uptake in DOACs compared to warfarin at the clinical commissioning group (CCG) level needs to be carried out. Publicly available- aggregated, CCG level, multi-source health and prescription records data were interrogated to investigate the association between prescription rate of DOACs and stroke/ bleed events during the period of 2013 to 2016. Variability of prescription rates and patient numbers across 208 CCGs were used to infer the effect of DOACs on stroke and bleed risk. Relative risk (RR) and 95% credible intervals (CI) were estimated using Markov chain Monte Carlo approach in JAGS. During the study period, the proportion of DOAC prescriptions increased at an average rate of 122% per annum. DOAC prescription was association with a 50% reduction in ischaemic (RR = 0.48, 95% CI = 0.39, 0.57) and haemorrhagic stroke (RR = 0.50, 95% CI = 0.26–0.77). In contrast, DOAC prescription reached significant association with reduction in gastrointestinal bleeds (RR = 0.86, 95% CI = 0.73–0.98) but not clinically relevant bleeds (RR = 0.95, 95% CI = 0.85–1.05). Sex stratified data showed significant association between DOAC prescription and reduction in haemorrhagic stroke risk (RR = 0.40, 95% CI = 0.28–0.52) and gastrointestinal bleeds (RR = 0.76, 95% CI = 0.63–0.93) in males only. Age stratified data suggested significant association with reduction in risk of both ischaemic and haemorrhagic strokes in patients aged 70 years and above, and reduction in risk of clinically relevant and gastrointestinal bleeds in patients aged 70–79 years only. Publicly available health and prescription data for the English population indicates reduction in stroke and bleed risk in specific age and sex sub-groups with the uptake of DOACs compared to warfarin between 2013 and 2016.