Supplementary Table 1: Association analyses of IL18 gene SNPs in MORGAM study according to smoking habits

OR in case-control analysis (1) / HR in time to event analysis (2)
SNPs / Non smokers / Smokers / Homogeneity (3) / Non smokers / Smokers / Homogeneity
Whole MORGAM Study / G-887T / 0.972
[0.855 – 1.104]
p = 0.660 / 1.161
[1.013 – 1.331]
p = 0.032 / p = 0.063 / 0.972
[0.820 – 1.151]
p = 0.740 / 1.119
[0.954 – 1.313]
p = 0.168 / p =0.236
C-105T / 0.899
[0.783 – 1.032]
p = 0.131 / 1.246
[1.067 – 1.455]
p = 0.005 / p = 0.002 / 0.932
[0.772 – 1.126]
p = 0.467 / 1.204
[0.987 – 1.470]
p = 0.067 / p = 0.067
S35S(A/C) / 0.892
[0.777 – 1.023]
p = 0.103 / 1.220
[1.046 – 1.422]
p = 0.011 / p = 0.003 / 0.933
[0.776 – 1.121]
p = 0.459 / 1.189
[0.976 – 1.447]
p = 0.085 / p = 0.078
A+183G / 1.039
[0.891 – 1.212]
p = 0.624 / 0.887
[0.753 – 1.046]
p = 0.154 / p = 0.167 / 1.031
[0.845 – 1.257]
p = 0.767 / 0.930
[0.754 – 1.148]
p = 0.500 / p = 0.485
T+533C / 1.019
[0.884 – 1.174]
p = 0.796 / 0.952
[0.819 – 1.105]
p = 0.515 / p = 0.520 / 1.030
[0.850 – 1.250]
p = 0.760 / 0.954
[0.793 – 1.147]
p = 0.615 / p = 0.573
Without ATBC cohort / G-887T / 0.959
[0.833 – 1.103]
p = 0.559 / 1.229
[0.953 – 1.585]
p = 0.112 / p = 0.094 / 0.954
[0.781 – 1.167]
p = 0.649 / 1.122
[0.850 – 1.481]
p = 0.416 / p = 0.353
C-105T / 0.899
[0.772 – 1.046]
p = 0.169 / 1.429
[1.064 – 1.918]
p = 0.017 / p = 0.006 / 0.952
[0.760 – 1.193]
p = 0.670 / 1.235
[0.894 – 1.706]
p = 0.200 / p = 0.195
S35S(A/C) / 0.897
[0.771 – 1.044]
p = 0.162 / 1.384
[1.029 – 1.860]
p = 0.031 / p = 0.011 / 0.974
[0.783 – 1.212]
p = 0.814 / 1.213
[0.872 – 1.687]
p = 0.252 / p = 0.277
A+183G / 1.079
[0.910 – 1.279]
p = 0.379 / 1.031
[0.758 – 1.401]
p = 0.847 / p = 0.800 / 1.103
[0.859-1.415]
p = 0.441 / 1.029
[0.751 – 1.409]
p = 0.860 / p = 0.735
T+533C / 0.988
[0.846 – 1.153]
p = 0.876 / 0.771
[0.587 – 1.012]
p = 0.061 / p = 0.121 / 0.968
[0.769 – 1.218]
p = 0.781 / 0.831
[0.625 – 1.105]
p = 0.203 / p = 0.414

ORs and HRs were calculated under the assumption of additive allele effects

(1): Population-adjusted Mantel-Haenszel ORs obtained using logistic regression analysis with offset.

(2): Population-adjusted Mantel-Haenszel HRs obtained using pseudolikelihood method.

(3): Homogeneity between smokers and non smokers was tested using Mantel-Haenszel statistics


Supplementary Table 2 Haplotypic structure of the IL18 gene estimated from the MORGAM subcohorts

Polymorphisms / FINRISK
N=798 / ATBC
N=980 / Sweden
N=145 / PRIME/N.Ireland
N=154 / PRIME/France
N =194
G-887T / C-105T / S35S
(A/C) / A+183G / T+533C
G / C / A / A / C / 0.288 / 0.275 / 0.328 / 0.356 / 0.329
G / C / A / G / T / 0.221 / 0.250 / 0.187 / 0.201 / 0.211
G / C / C / A / T / 0.021 / 0.026 / 0.015 / 0.032 / 0.023
T / C / A / A / T / 0.171 / 0.173 / 0.116 / 0.119 / 0.136
T / C / A / A / C / 0.004 / 0.006 / 0.011 / 0.029 / 0.021
T / T / C / A / T / 0.292 / 0.261 / 0.311 / 0.248 / 0.250

Supplementary Table 3 : Haplotypic frequencies in cases and controls by cohort

Polymorphims / FINRISK / ATBC / Sweden / PRIME/N.Ireland / PRIME/France / Entire
G-887T / C-105T / S35S / A+183G / T+533C / Controls / Casesb / Controls. / Cases / Controls. / Cases / Controls. / Cases / Controls. / Cases / Controls. / Cases
G / C / A / A / C / 0.29 / 0.27 / 0.27 / 0.28 / 0.32 / 0.37 / 0.36 / 0.31 / 0.33 / 0.33 / 0.30 / 0.28
G / C / A / G / T / 0.22 / 0.23 / 0.25 / 0.22 / 0.20 / 0.22 / 0.21 / 0.24 / 0.23 / 0.22 / 0.23 / 0.23
G / C / C / A / T / 0.02 / 0.02 / 0.03 / 0.03 / 0.01 / 0.01 / 0.03 / 0.06 / 0.02 / 0.03 / 0.02 / 0.02
T / C / A / A / T / 0.17 / 0.18 / 0.17 / 0.18 / 0.11 / 0.09 / 0.11 / 0.06 / 0.12 / 0.10 / 0.16 / 0.16
T / T / C / A / T / 0.29 / 0.28 / 0.26 / 0.28 / 0.31 / 0.26 / 0.25 / 0.29 / 0.24 / 0.27 / 0.27 / 0.28
Global likelihood ratio test of association / c2 = 3.47
with 4 df
P = 0.482 / c2 = 3.94
with 4 df
P = 0.415 / c2 = 4.23
with 4 df
P = 0.375 / c2 = 9.17
with 4 df
P= 0.057 / c2 = 2.50
with 4 df
P = 0.644 / c2 = 0.34
with 4 df
P = 0.987

b Cases: CHD and strokes were combined


Supplementary Table 4: Smoking x C-105T genotypes interaction in a case-cohort logistic regression model for CVD risk in the MORGAM project. Values are regression coefficients (SE) adjusted for age and gender (when appropriate)

Cohorts / Smoking
(smoker vs non smokers) / C-105T polymorphism
(T carriers vs CC carriers) / Smoking x C-105T
FINRISK / 0.337 (0.185) / -0.103 (0.127) / 0.426 (0.264)
ATBC / 0.016 (0.173) / -0.223 (0.222) / 0.430 (0.252)
Sweden / -0.010 (0.506) / -0.326 (0.332) / 0.850 (0.710)
PRIME/N.Ireland / 0.176 (0.380) / -0.080 (0.283) / 0.868 (0.566)
PRIME/France / 0.747 (0.375) / -0.117 (0.144) / 0.144 (0.558)
ENTIRE (a) / 0.209 (0.11) / -0.141 (0.093) / 0.462 (0.161)

(a) Pooled estimates obtained using the Mantel-Haenszel statistics