Abstract Body: Introduction: It is unknown how comorbid diagnoses affect patients’ disease-specific knowledge or how patients’ health literacy or numeracy affects this relationship. In patients with atrial fibrillation (AF), comorbidities can be quantified by the CHA₂DS₂-VASc score. We explore the association of AF/stroke knowledge with CHA₂DS₂-VASc score, health literacy, and numeracy for usual care participants in the multi-center Randomized Evaluation of Decision Support Interventions for Atrial Fibrillation (RED-AF) trial evaluating decision aids to initiate (initiation cohort) or continue (monitor cohort) oral anticoagulation (OAC) in AF patients.

Methods: This study was conducted at six academic medical centers in the United States. Participants were aged ≥18, diagnosed with non-valvular AF, and at risk for stroke (CHA₂DS₂-VASc ≥1 for men, ≥2 for women). Patient demographics were obtained via questionnaires after a baseline clinical encounter; knowledge was evaluated by patient responses to 7 questions about OAC and AF. Health literacy was measured by a validated self-reported screening tool; numeracy by a validated subjective numeracy scale; CHA₂DS₂-VASc was calculated from medical records. We applied generalized estimating equations for a binomial outcome to jointly relate patient knowledge to CHA₂DS₂-VASc score, health literacy and numeracy, controlling for 5 additional patient factors and accounting for clustering by clinician.

Results: We report on 306 patients randomized to the usual care group across six participating sites. Average age 68.7 9, 61.6% male, mean CHA₂DS₂-VASc 2 1.5. Patient knowledge was independently associated with numeracy (5% higher odds of correct response per 1 unit increase in numeracy, 95% CI 0-10%, p=0.05) and membership in the monitor cohort (40% odds of correct response, 4-89%, p=0.03) but not health literacy or increased stroke risk.

Conclusion: In the RED-AF study, although health literacy and increasing stroke risk were not independently associated with patient knowledge about AF, numeracy had a minimal positive association. Those already on OAC had greater disease knowledge than patients newly starting OAC. These results suggest a need to broadly increase patient education regardless of baseline comorbidity burden and degree of health literacy. Future studies should be performed to better understand other factors that drive patient knowledge regarding AF and stroke.