Background: Management of atrial fibrillation (AF) is complex and requires active patient engagement in shared decision making to achieve better clinical outcomes, greater medication adherence, and increased treatment satisfaction. Efficacy in patient-physician interactions is a critical component of patient engagement, but factors associated with efficacy in older AF patients have not been well-characterized. Methods: We performed a cross-sectional analysis of baseline data from the ongoing Systematic Assessment of Geriatric Elements in Atrial Fibrillation (SAGE-AF) study, a cohort study of older adults (age ≥ 65) with non-valvular AF and CHA2DS2-VASc score ≥ 2. Participants were classified according to their Perceived Efficacy in Patient-Physician Interactions (PEPPI-5) score (lower: 0-44; higher: 45-50). Logistic regression analysis was used to identify sociodemographic, clinical (AF type, AF treatment, medical comorbidities), and geriatric (cognitive impairment, sensory impairment, frailty, independent functioning) factors associated with lower reported efficacy. Results: Participants (n = 1209; 49% female) had a mean age of 75. A majority (66%) reported higher efficacy in their interactions with physicians. Lower efficacy was associated with persistent AF (adjusted odds ratio [aOR] = 1.52; 95% confidence interval [CI] = 1.13-2.04) and with symptoms of depression (aOR = 1.67; CI = 1.20-2.33) or anxiety (aOR = 1.40; CI = 1.01-1.94). Decreased odds of lower efficacy were observed in participants with chronic kidney disease (aOR = 0.68; CI = 0.50-0.92) and those classified as pre-frail compared to those classified as not frail (aOR = 0.71; CI = 0.53-0.95). Conclusion: Older patients with persistent AF or symptoms of depression or anxiety have decreased efficacy in patient-physician interactions. These individuals merit greater attention from physicians when engaged in shared decision making.