All-cause mortality trends in patients hospitalized for atrial fibrillation in Sweden: Role of age, stroke risk, and education

Background: The incidence of atrial fibrillation (AF) has long been increasing, and AF is associated with increased mortality. Over time, mortality trends may differ between subgroups depending on their underlying risk patterns and treatments. Aim: To explore all-cause-mortality trends over time in patients hospitalized for incident AF, and the effects of age, stroke risk, and education level. Methods and results: Patients hospitalized for incident AF between January 1995 and December 2003 were selected from Swedish national registries. Based on date of index admission, patients were divided into four cohorts and followed for five years. Age- and sex-matched controls were selected. Kaplan-Meier estimates and Cox regressions with trend analysis were used for statistical evaluation. There were 64,489 patients (mean age 72 +/- 10.1 years) and 116,893 controls. There was a significantly decreasing trend in the relative risk of all-cause mortality in AF patients over time, with a trend hazard ratio of 0.94 (95 % confidence interval [CI] 0.92-0.96, p < 0.001) in women and 0.91 (95 % CI 0.89-0.93, p < 0.001) in men. The mortality trends did not differ significantly between AF patients and controls. The mortality risk remained unchanged in women aged 18-64 years, in patients with low stroke risk, and in patients with post-secondary education. Conclusion: The all-cause mortality risk decreased over time in both patients and controls, but subgroup analysis revealed an unchanged mortality trend in women aged 18-64 years, in patients with low stroke risk, and in patients with post-secondary education.