TY - JOUR

T1 - Geographical variation and clustering are found in atrial fibrillation beyond socioeconomic differences

T2 - a Danish cohort study, 1987–2015

AU - Wodschow, Kirstine

AU - Bihrmann, Kristine

AU - Larsen, Mogens Lytken

AU - Gislason, Gunnar

AU - Ersbøll, Annette Kjær

N1 - Funding Information: This work was supported by the Karen Elise Jensen’s Foundation. Publisher Copyright: © 2021, The Author(s).

PY - 2021

Y1 - 2021

N2 - Background: The prevalence and incidence rate of atrial fibrillation (AF) increase worldwide and AF is a risk factor for more adverse cardiovascular diseases including stroke. Approximately 44% of AF cases cannot be explained by common individual risk factors and risk might therefore also be related to the environment. By studying geographical variation and clustering in risk of incident AF adjusted for socioeconomic position at an individual level, potential neighbourhood risk factors could be revealed. Methods: Initially, yearly AF incidence rates 1987–2015 were estimated overall and stratified by income in a register-based cohort study. To examine geographical variation and clustering in AF, we used both spatial scan statistics and a hierarchical Bayesian Poisson regression analysis of AF incidence rates with random effect of municipalities (n = 98) in Denmark in 2011–2015. Results: The 1987–2015 cohort included 5,453,639 individuals whereof 369,800 were diagnosed with an incident AF. AF incidence rate increased from 174 to 576 per 100,000 person-years from 1987 to 2015. Inequality in AF incidence rate ratio between highest and lowest income groups increased from 23% in 1987 to 38% in 2015. We found clustering and geographical variation in AF incidence rates, with incidence rates at municipality level being up to 34% higher than the country mean after adjusting for socioeconomic position. Conclusions: Geographical variations and clustering in AF incidence rates exist. Compared to previous studies from Alberta, Canada and the United States, we show that geographical variations exist in a country with free access to healthcare and even when accounting for socioeconomic differences at an individual level. An increasing social inequality in AF was seen from 1987 to 2015. Therefore, when planning prevention strategies, attention to individuals with low income should be given. Further studies focusing on identification of neighbourhood risk factors for AF are needed.

AB - Background: The prevalence and incidence rate of atrial fibrillation (AF) increase worldwide and AF is a risk factor for more adverse cardiovascular diseases including stroke. Approximately 44% of AF cases cannot be explained by common individual risk factors and risk might therefore also be related to the environment. By studying geographical variation and clustering in risk of incident AF adjusted for socioeconomic position at an individual level, potential neighbourhood risk factors could be revealed. Methods: Initially, yearly AF incidence rates 1987–2015 were estimated overall and stratified by income in a register-based cohort study. To examine geographical variation and clustering in AF, we used both spatial scan statistics and a hierarchical Bayesian Poisson regression analysis of AF incidence rates with random effect of municipalities (n = 98) in Denmark in 2011–2015. Results: The 1987–2015 cohort included 5,453,639 individuals whereof 369,800 were diagnosed with an incident AF. AF incidence rate increased from 174 to 576 per 100,000 person-years from 1987 to 2015. Inequality in AF incidence rate ratio between highest and lowest income groups increased from 23% in 1987 to 38% in 2015. We found clustering and geographical variation in AF incidence rates, with incidence rates at municipality level being up to 34% higher than the country mean after adjusting for socioeconomic position. Conclusions: Geographical variations and clustering in AF incidence rates exist. Compared to previous studies from Alberta, Canada and the United States, we show that geographical variations exist in a country with free access to healthcare and even when accounting for socioeconomic differences at an individual level. An increasing social inequality in AF was seen from 1987 to 2015. Therefore, when planning prevention strategies, attention to individuals with low income should be given. Further studies focusing on identification of neighbourhood risk factors for AF are needed.

KW - Atrial fibrillation

KW - Bayesian analysis

KW - Cluster analysis

KW - Epidemiology

KW - Health registers

KW - Health status disparities

U2 - 10.1186/s12942-021-00264-2

DO - 10.1186/s12942-021-00264-2

M3 - Journal article

C2 - 33648527

AN - SCOPUS:85101854173

VL - 20

JO - International Journal of Health Geographics

JF - International Journal of Health Geographics

SN - 1476-072X

IS - 1

M1 - 11

ER -