(P087) STROKE, DEATH, AND ADHERENCE TO ORAL ANTICOAGULANTS IN AF: A RETROSPECTIVE OBSERVATIONAL STUDY WITH ADHERENCE AS A CONTINUOUS VARIABLE

Background: Patients with atrial fibrillation (AF) who are prescribed oral anticoagulants (OACs) for stroke and systemic embolism (SSE) prevention are frequently nonadherent to therapy. Uncertainty remains about the association between nonadherence and the risk of SSE and death. The objective of this study was to quantify the association between nonadherence to OACs and its clinical consequences.

METHODS AND RESULTS: Using linked, population-based administrative data containing physician billing, hospitalization, and prescription records of 5 million British Columbians (1996-2020), incident adult cases of AF were studied. For each patient, proportion of days covered (PDC) was calculated from the date of their first prescription for any OAC to the end of follow-up as a continuous annual time-updated exposure of interest. The validated REWarDS method was applied to estimate PDC for warfarin. Multivariable Cox proportional hazard models were used to evaluate the primary outcome, the association between PDC and SSE or death, controlling for known confounders. Secondary analyses were done for death and SSE as discrete outcomes. The study cohort included 41,033 OAC recipients [mean age 68.4y (SD 12.7), 45% female, mean CHA2DS2-VASc score 2.25 (SD 1.46)]. The mean PDC for the cohort was 0.69 (SD 0.28) over a median follow-up of 6.7 years. Multivariate modelling showed that every 10% absolute increase in PDC was associated with a 5.3% lower hazard of SSE or death [Hazard Ratio (HR) 0.47, 95%CI 0.42 – 0.53]. This result was stable in the sensitivity analysis when the cohort was confined to those starting therapy after 2010 (when all OACs were available; n=19,588): HR 0.47 (95%CI 0.37-0.60). Secondary analyses showed that every 10% increase in PDC was also associated with a 4.8% and 5.6% lower hazard of death (HR 0.52, 95%CI 0.44 – 0.61), and SSE (HR 0.44, 95%CI 0.38-0.52), respectively.

Conclusion: Degree of adherence to OACs in patients with AF is strongly associated with SSE and death, with an approximate 5% reduction in hazard of these events for every 10% absolute increase in average annual adherence. This effect was robust over different periods of study. Efforts to improve patients’ adherence to OACs could significantly reduce their risk of these devastating clinical events.