paperKB
coga / coga-kb
Help
Sign in

Chunk #15 — Materials and methods — Empirical example: SNP–CHD associations in EPIC-CVD

Source
A comparison of Cox and logistic regression for use in genome-wide association studies of cohort and case-cohort design.
Embedded
yes

Text

EPIC-CVD is a case-cohort study of 31 050 participants derived from 29 recruitment centres across 10 European studies nested within the EPIC study.14, 15 The subcohort contains 17 640 individuals, of which 631 had an incident fatal or non-fatal CHD (including angina) event. There were an additional 13 333 CHD events outside the subcohort. The mean age at baseline was 55.3 years and 46.9% of the participants are males. Individuals were genotyped using either the Illumina Cardio-MetaboChip array or the Illumina 660W-Quad BeadChip array. In total, 18 889 individuals had genetic information. We analysed 25 SNPs located within known CHD loci, which were genotyped on both arrays and were available in at least 95% of the genotyped individuals. A multiplicative allelic effects model (on the HR/OR scale) was assumed for the SNPs, and we adjusted for age (in years), sex, EPIC-CVD centre (as a categorical variable) and 10 principal components to adjust for ancestry. Age, sex and EPIC centre were used as covariates to account for differences across EPIC centres (including any differences in the distributions of age and sex).