Association Between Socioeconomic Status and Atrial Fibrillation: A Two-Sample Mendelian Randomization Study.
Lower socioeconomic status (SES), reflected by greater socioeconomic deprivation, lower household income, and lower educational attainment (EA), has been associated with atrial fibrillation (AF). This study aimed to examine the causal influence of SES on AF risk.
Single-nucleotide polymorphisms (SNPs) associated with Townsend deprivation index (TDI) or household income from the UK Biobank (N = 462,464 and N = 397,751, respectively) and EA from the Social Science Genetic Association Consortium (N = 3,037,499) were selected as instrumental variables. AF-associated SNPs were extracted from FinnGen data freeze 8 (N = 208,594) and a meta-analysis (N = 1,030,836) by Nielsen et al. Two-sample Mendelian randomization (MR) was used to estimate causal relationships, with validity assessed via sensitivity analyses.
TDI showed no association with AF risk in either FinnGen (odds ratio (OR) = 0.972, 95% confidence interval (CI): 0.607-1.556, p = 9.00E-01) or Nielsen et al. (OR = 0.805, 95% CI: 0.470-1.381, p = 4.30E-01). Similarly, genetically proxied household income was not associated with AF risk (OR = 1.233, 95% CI: 0.808-1.881, p = 3.30E-01; OR = 0.941, 95% CI: 0.755-1.173, p = 5.90E-01). However, higher genetically proxied EA was associated with lower AF risk in FinnGen (OR = 0.745, 95% CI: 0.653-0.849, p = 1.00E-05) and Nielsen et al. (OR = 0.849, 95% CI: 0.784-0.920, p = 5.90E-05).
Our MR study found that higher genetically proxied EA was associated with lower AF risk, whereas no evidence of association was found for genetically proxied TDI or household income.
Single-nucleotide polymorphisms (SNPs) associated with Townsend deprivation index (TDI) or household income from the UK Biobank (N = 462,464 and N = 397,751, respectively) and EA from the Social Science Genetic Association Consortium (N = 3,037,499) were selected as instrumental variables. AF-associated SNPs were extracted from FinnGen data freeze 8 (N = 208,594) and a meta-analysis (N = 1,030,836) by Nielsen et al. Two-sample Mendelian randomization (MR) was used to estimate causal relationships, with validity assessed via sensitivity analyses.
TDI showed no association with AF risk in either FinnGen (odds ratio (OR) = 0.972, 95% confidence interval (CI): 0.607-1.556, p = 9.00E-01) or Nielsen et al. (OR = 0.805, 95% CI: 0.470-1.381, p = 4.30E-01). Similarly, genetically proxied household income was not associated with AF risk (OR = 1.233, 95% CI: 0.808-1.881, p = 3.30E-01; OR = 0.941, 95% CI: 0.755-1.173, p = 5.90E-01). However, higher genetically proxied EA was associated with lower AF risk in FinnGen (OR = 0.745, 95% CI: 0.653-0.849, p = 1.00E-05) and Nielsen et al. (OR = 0.849, 95% CI: 0.784-0.920, p = 5.90E-05).
Our MR study found that higher genetically proxied EA was associated with lower AF risk, whereas no evidence of association was found for genetically proxied TDI or household income.