The role of education level in the mortality of hospitalized patients with atrial fibrillation

FUNDING ACKNOWLEDGEMENTS: Type of funding sources: None. INTRODUCTION: The association between social status and cardiovascular morbidity and mortality is well­–known. Education level is an important socioeconomic factor which influences the incidence, symptoms, and treatments of atrial fibrillation (AF). However, data about the effects of education level on all-cause mortality in patients with AF are scarce. AIM: To explore any association between education level and all-cause mortality in patients hospitalized with AF. METHODS: This retrospective cohort study is based on a database created by linking Swedish nationwide registries including all patients hospitalized with an AF diagnose between 1995 and 2003. In all, 158 577 patients were included in the study (56% male and 44% female; mean age 72 ± 11 years). The follow up was 5 years with an observation time of 636 597 person–years. Education level was described by 3 categories representing primary, secondary, and academic education. Kaplan-Meier curves and Cox proportional hazard models adjusted with age, sex, time of index hospitalization, categorized Charlson´s Comorbidity Index (CCI) score, and CHA2DS2-VASc score were used for statistical analysis. Hazard ratio (HR) with 95% confidence interval (CI) was used as estimate of association and P < 0.05 was regarded as statistically significant. The risk of all-cause mortality was estimated in subpopulations defined by mortality risk factors as CCI, CHA2DS2-VASc score, heart failure, coronary artery disease, chronic obstructive pulmonary disease, diabetes, chronic kidney disease, and cancer. RESULTS: Patients with academic education had the lowest all–cause mortality, which was seen early during follow–up. Compared to patients with primary education, those with secondary or academic education showed lower relative mortality risk in the adjusted Cox regression model (secondary education HR: 0.875, 95% CI: 0.86 – 0.89, p<0.001; academic education HR: 0.695, 95% CI: 0.67 – 0.72; p<0.001). Among the subpopulations, the HR varied between 0.72 and 0.95 for secondary and between 0.43 and 0.79 for academic education respectively. All risk differences were statistically significant. CONCLUSION: Academic education was associated with the lowest and primary education with the highest risk of all-cause mortality. This pattern was consistent through all analyzed subpopulations.