Abstract MP23: Neighborhood Disadvantage And Oral Anticoagulant Use In Patients With Atrial Fibrillation

Abstract

Background: Atrial Fibrillation (AF) is the most common heart rhythm disorder in the US and is treated with anticoagulation to mediate patients’ increased risk of ischemic stroke and death. Direct-acting oral anticoagulants (DOACs) are associated with better patient outcomes than warfarin, but prior studies show disparities in DOAC prescription. AF patients with low socioeconomic status (SES) have an increased risk of AF incidence, adverse outcomes, and mortality. Despite SES disparities in AF outcomes and prescribed medications, the impact of neighborhood SES (ADI) on AF management is unclear. We hypothesized that in comparison to patients with high ADI (e.g. the most disadvantaged), those with low ADI will be more likely to initiate any oral anticoagulants (OACs) and will be more likely to initiate DOACs. Methods: We conducted a retrospective observational study using national data from the Veterans Health Administration (VA) of newly diagnosed AF patients. Our independent variable was ADI, a marker of SES incorporating income, education, employment etc. Our model assessed numerous baseline patient, provider, and system-level covariates. We grouped patients into ADI quintiles, with Q1 being the lowest ADI (e.g., the least disadvantaged) and Q5 being the highest ADI, and used Q5 as a reference value. We used mixed effects logistic regression models, with site as a random effect, to determine the adjusted odds of initiating any OACs and of initiating DOACs by ADI quintile. Results: In our final cohort including 111,666 patients, 10,9386 (98.0%) were male and 2280 (2.0%) were female. The overall mean age at diagnosis was 72.86 (SD 10.4). Patients had the following medical comorbidities: Congestive heart failure: 18212, (16.3%); Hypertension: 84944, (76.1%); Diabetes: 70673, (63.3%); Vascular Disease: 46599, (41.7%). The ADI quintiles included the following patients: Q1: 21570, (20.3%); Q2: 21032, (19.8%); Q3: 21439, (20.2%); Q4: 20974, (19.8%); Q5: 21215, (20.0%). There was no significant difference in the odds of initiating any OAC between ADI quintiles compared to Q5 (Q1: adjusted odds ratio [95% CI], 0.95 [ 0.9, 1.01]; p=0.11). Among patients initiating DOAC, there was a significant association between low ADI and higher odds of initiating DOAC (Q1: 1.39 [1.29, 1.50]); (Q2: 1.18 [1.10, 1.27]); (Q3: 1.13 [1.06, 1.20]); (Q4: 1.09 [1.02, 1.16]). Each ADI quintile had a significantly higher odds of initiating DOACs compared to the highest ADI quintile. Conclusions: Patients with lower ADI are significantly more likely to be prescribed DOACs compared to patients with the highest levels of ADI. We found no significant difference in initiation of any OAC between ADI quintiles, which may result from high Warfarin utilization in the VA compared to the general population. Our data suggests that differences in neighborhood deprivation contribute to disparities in the prescribing DOACs in the VA.