Table 3.
The association between unconjugated bilirubin (UCB) levels and colorectal cancer risk across strata of potential effect modifiers in the EPIC study
| Colorectal cancer | ||||||||
|---|---|---|---|---|---|---|---|---|
| Variables | Men | Women | ||||||
| n cases/controls† | Odds ratio (95% CI) | P | Pheterogeneity | n cases/controls† | Odds ratio (95% CI) | P | Pheterogeneity | |
| Adjusted model | 658/658 | 1.19 (1.04–1.36) | 0.01 | 728/728 | 0.86 (0.76–0.97) | 0.02 | ||
| Age at blood collection (year, median) | 658/658 | 0.30 | 728/728 | 0.008 | ||||
| < 59.2/< 58.5 | 329/329 | 1.28 (1.06–1.54) | 364/364 | 1.01 (0.85–1.19) | > 0.90 | |||
| ≥ 59.2/≥ 58.5 | 329/329 | 1.10 (0.92–1.32) | 364/364 | 0.73 (0.61–0.87) | 0.001 | |||
| rs6431625 (increasing levels C allele) | 333/333 | 0.02 | 428/428 | 0.14 | ||||
| TT genotype (wild-type) | 117/139 | 1.1 (0.73–1.65) | 0.60 | 168/151 | 0.69 (0.49–0.99) | 0.04 | ||
| TC genotype | 160/146 | 0.92 (0.64–1.32) | 0.60 | 194/206 | 0.71 (0.53–0.95) | 0.02 | ||
| CC genotype | 56/48 | 2.01 (1.26–3.20) | 0.003 | 66/71 | 1.06 (0.74–1.53) | 0.7 | ||
| Smoking status | 658/658 | 0.40 | 728/728 | 0.20 | ||||
| Never | 170/170 | 1.26 (1.00–1.59) | 0.05 | 427/427 | 0.83 (0.71–0.98) | 0.03 | ||
| Former | 294/294 | 1.06 (0.86–1.30) | 0.60 | 160/160 | 0.84 (0.66–1.07) | 0.20 | ||
| Current | 182/182 | 1.33 (1.08–1.64) | 0.01 | 137/137 | 0.99 (0.78–1.25) | 0.90 | ||
| Unknown | 12/12 | 0.93 (0.35–2.49) | 0.90 | 4/4 | 0.05 (0.00–0.1) | 0.20 | ||
| Follow-up time (years) | 658/658 | 0.20 | 728/728 | 0.60 | ||||
| 1 (< 2) | 141/141 | 0.96 (0.74–1.25) | 0.80 | 158/158 | 0.82 (0.65–1.05) | 0.10 | ||
| 2 (2–4) | 173/173 | 1.34 (1.04–1.72) | 0.02 | 160/160 | 0.96 (0.76–1.23) | 0.80 | ||
| 3 (> 4) | 244/244 | 1.23 (1.03–1.50) | 0.02 | 280/380 | 0.84 (0.71–0.99) | 0.04 | ||
No effect modifications by BMI (median), alcohol consumption (median), menopausal status, and use of HT were observed (all P ≥ 0.7)
Abbreviations: n number, P P value, CI confidence interval
†Cases matched 1:1 to control subjects. EPIC (European Prospective Investigation into Cancer and Nutrition): Conditional logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals (CI) for associations between log-transformed UCB levels (log-UCB), standardized per one standard deviation (1-SD) increments, and CRC risk. The crude model was conditioned on the matching factors including study center, age at blood collection (1 year), fasting status and time (3 h intervals) at blood collection, among women, additionally by menopausal status (pre-, peri-, and post-menopausal or surgically menopausal), and hormone therapy (HT) (yes, no). The multivariable model was adjusted for level of education (none/primary school, technical/professional, secondary school, university degree), BMI, height, smoking status (never, former, current smoker), physical activity (inactive, moderately inactive, moderately active, active), alcohol consumption (g/day), dietary intakes of fiber (g/day), red meat (g/day), processed meats (g/day), dairy products (g/day), and total energy intake (kcal/day)