Table 3. Odds ratios (95% CI) for colon and rectal cancer according to quartiles of circulating glycer-AGEs by alcohol intake status.

Sex specific categories of dietary alcohol intake level (g/day)	Glycer-AGEsa				ptrend
	Q1 OR	Q2 OR (95% CI)	Q3 OR (95% CI)	Q4 OR (95% CI)
Colon
Low alcohol status (≤ median)b
Number of cases/controls	135/130	145/137	119/119	104/124
Matching factorsc	1.00	0.97 (0.69-1.37)	0.79 (0.54-1.15)	0.71 (0.46-1.10)	0.08
Multivariate adjustedd	1.00	1.01 (0.69-1.47)	0.79 (0.52-1.20)	0.82 (0.51-1.33)	0.29
High alcohol status (> median)b
Number of cases/controls	39/36	39/28	38/45	40/40
Matching factorsc	1.00	1.48 (0.68-3.20)	0.92 (0.40-2.07)	1.11 (0.47-2.58)	0.95
Multivariate adjustedd	1.00	1.10 (0.46-2.61)	0.78 (0.31-1.95)	1.17 (0.46-3.03)	0.95
Rectum
Low alcohol status (≤ median)b
Number of cases/controls	51/64	70/57	38/46	57/63
Matching factorsc	1.00	1.56 (0.92-2.65)	0.95 (0.52-1.73)	1.14 (0.66-1.97)	0.95
Multivariate adjustedd	1.00	1.48 (0.85-2.58)	0.95 (0.51-1.78)	1.14 (0.64-2.03)	0.99
High alcohol status (> median)b
Number of cases/controls	26/35	45/43	40/52	68/36
Matching factorsc	1.00	1.39 (0.69-2.77)	1.14 (0.57-2.29)	2.85 (1.41-5.79)	< 0.01
Multivariate adjustedd	1.00	1.41 (0.69-2.88)	1.07 (0.52-2.19)	2.70 (1.29-5.62)	0.01

^aQuartile cut-offs are same as in Table 2

^bAlcohol consumption level was dichotomized based on the sex-specific median values of lifetime alcohol consumption among controls

^cModel adjusted for matching factors (age, gender, administrative centre, time of the day at blood collection, fasting status, and menopausal status among women) only

^dModel adjusted for matching factors plus smoking status/duration/intensity, body mass index, total physical activity, education level, diabetes status, total dietary energy consumption, red and processed meat, fibre, and fruits and vegetable

Abbreviation: Glycer-AGEs = glyceraldehyde-derived advanced glycation end-products; Q = quartile; OR = odds ratio; CI = confidence interval