Social determinants of health and clinical outcomes among patients with atrial fibrillation: evidence from a global federated health research network.

Background Few studies have investigated the role of social determinants of health (SDoH) in patients with atrial fibrillation (AF). Aim To investigate the relationship between SDoH and adverse events in a large multinational AF cohort. Design Retrospective study utilizing a global federated health research network (TriNetX). Methods Patients with AF were categorized as socially deprived defined according to ICD codes based on three SDoHs: (i) extreme poverty; (ii) unemployment; and/or (iii) problems related with living alone. The outcomes were the 5-year risk of a composite outcomes of all-cause death, hospitalization, ischemic heart disease (IHD), stroke, heart failure (HF) or severe ventricular arrhythmias. Cox regression was used to compute hazard rate ratios (HRs) and 95% confidence intervals (CIs) following 1:1 propensity score matching (PSM). Results The study included 24 631 socially deprived (68.8 ± 16.0 years; females 51.8%) and 2 462 092 non-deprived AF patients (75.5 ± 13.1 years; females 43.8%). Before PSM, socially deprived patients had a higher risk of the composite outcome (HR 1.9, 95% CI 1.87–1.93), all-cause death (HR 1.34, 95% CI 1.28–1.39), hospitalization (HR 2.01, 95% CI 1.98–2.04), IHD (HR 1.67, 95% CI 1.64–1.70), stroke (HR 2.60, 95% CI 2.51–2.64), HF (HR 1.91, 95% CI 1.86–1.96) and severe ventricular arrhythmias (HR 1.83, 95% CI 1.76–1.90) compared to non-deprived AF patients. The PSM-based hazard ratios for the primary composite outcome were 1.54 (95% CI 1.49–1.60) for the unemployed AF patients; 1.39 (95% CI 1.31–1.47) for patients with extreme poverty or with low income; and 1.42 (95% CI 1.37–1.47) for those with problems related with living alone. Conclusions In patients with AF, social deprivation is associated with an increased risk of death and adverse cardiac events. The presence of possible unmeasured bias associated with the retrospective design requires confirmation in future prospective studies. [ABSTRACT FROM AUTHOR]