Table 3.
Gene-environment interaction stratified by gender. Haplotype-specific effects of average smoking intensity and consumption of alcohol on the risk of colorectal cancer
| Haplotype | Men | Women | |||||||
| Carriera | N cases/sub-cohort | IRRb | CI (95%)c | Pd | N cases/sub-cohort | IRRb | CI (95%)c | Pd | |
| Smoking intensitye (per 10 g tobacco/day) | No | 130/232 | 0.93 | 0.76–1.13 | 0.26 | 62/149 | 1.58 | 0.99–2.50 | 0.42 |
| Yes | 45/74 | 1.14 | 0.79–1.63 | 31/52 | 1.23 | 0.64–2.38 | |||
| missing | 3/8 | 4/4 | |||||||
| Alcohol intakef (per 10 g/day) | No | 164/313 | 1.04 | 0.97–1.12 | 0.15 | 121/276 | 1.15 | 0.98–1.36 | 0.17 |
| Yes | 51/104 | 1.20 | 1.00–1.43 | 47/80 | 0.87 | 0.61–1.25 | |||
| missing | 3/13 | 7/6 | |||||||
a) Homozygous carriers of the haplotype compared to the rest. Homozygous carrier of the previously identified haplotype consisting of ERCC1 Asn118AsnTT, ASE-1 G-21AGG and RAI IVS1 A4364GAA
b) IRR: incidence rate ratio
c) CI: 95% confidence interval
d) P: test for interaction
e) Risk estimates for ever smokers only. Adjusted for smoking status (present/former/never), intake of alcohol, fruits/vegetables, fish/poultry, red and processed meat, dietary fibres, BMI and hormone replacement therapy.
f) Risk estimates for alcohol consumers only. Adjusted for average smoking intensity, fruits/vegetables, fish/poultry, red and processed meat, dietary fibres, BMI and hormone replacement therapy