Psychosocial Symptom Networks Across Atrial Fibrillation Subtypes: Evidence from Two Waves of the ARENA Cohort

Abstract Introduction Atrial fibrillation (AF) carries substantial psychosocial burden, but how psychosocial symptoms interrelate across AF subtypes and severities remains unclear. This study compared psychosocial symptom networks across time, AF type, and European Heart Rhythm Association (EHRA) class in the ARENA (Atrial Fibrillation Rhine-Neckar Region) cohort. Methods Adults with AF were assessed at baseline (T0) and follow-up (T1). Analytic samples included 996 patients with EHRA class (asymptomatic [EHRA 1] vs. symptomatic [EHRA ≥ 2]) and 924 with AF subtype (paroxysmal, persistent, or permanent) data. Measures included anxiety and worry (GAD-2), cardiac worry and avoidance (CAQ-2), health-related quality of life (EQ-5D-5L), and psychosocial stressors. Gaussian Graphical Models were estimated and compared using Network Comparison Test. Results Global network structures were stable across EHRA classes, AF subtypes, and time points, but specific edges and node centralities varied. The most robust edges linked anxiety with general worry and cardiac worry with avoidance. Anxiety, general worry, cardiac worry, and diminished health-related quality of life (HRQoL) were consistently the most central nodes. Among symptomatic (EHRA ≥ 2) patients, loneliness emerged as an increasingly central node over time, whereas among asymptomatic patients (EHRA 1), sleep problems showed a stronger association with HRQoL at follow-up. AF-type network structures were similar, although paroxysmal networks showed higher node centrality and stronger anxiety–worry coupling at T0; persistent or permanent AF showed a stress-at-home–cardiac worry edge at T1. Conclusions Psychosocial symptom networks in AF are comparable across subtypes and reveal intervention targets, particularly anxiety, cardiac worry with avoidance, loneliness, and sleep disturbance.