Neighbourhood socio-economic status and all-cause mortality in adults with atrial fibrillation: A cohort study of patients treated in primary care in Sweden

Abstract

Objective: Our aim was to study the potential impact of neighbourhood socio-economic status (SES) on all-cause mortality in patients with atrial fibrillation (AF) treated in primary care.

Methods: Study population included adults (n=12,283) of 45 years and older diagnosed with AF in 75 primary care centres in Sweden. Association between neighbourhood SES and all-cause mortality was explored using Cox regression analysis, with hazard ratios (HRs) and 95% confidence intervals (95% CIs), and by Laplace regression where years to death (95% CI) of the first 10% of the participants were used as an outcome. All models were conducted in both men and women and adjusted for age, educational level, marital status, change of neighbourhood status, cardiovascular co-morbidities, anticoagulant treatment and statin treatment. High- and low-neighbourhood SES were compared with middle SES as reference group.

Results: After adjustments for potential confounders, higher relative risk of all-cause mortality (HR 1.49, 95% CI 1.13-1.96) was observed in men living in low SES neighbourhoods compared to those from middle SES neighbourhoods. The results were confirmed using Laplace regression; the time until the first 10% of the men in low SES neighbourhoods died was 1.45 (95% CI 0.48-2.42) years shorter than for the men in middle SES neighbourhoods.

Conclusions: Increased rates of heart disease and subsequent mortality among adults in deprived neighbourhoods raise important clinical and public health concerns. These findings could serve as an aid to policy-makers when allocating resources in primary health care settings as well as to clinicians who encounter patients in deprived neighbourhoods.

Keywords: Antithrombotic drugs; Atrial fibrillation; Co-morbidity; Follow-up; Gender; Mortality; Neighbourhood.

Figure 1. Diagram of the causal pathway to all-cause mortality

Neighbourhood SES categories are likely to affect mortality either by a direct route or by mediators within the causal pathway (in this case lifestyle factors, psychosocial stress or com-morbidities). Educational level, marital status, age and co-morbidities were regarded as confounders. N.b. co-morbidities could be regarded as both confounders and mediators.