Table 2.

Association between alcohol consumption in each life stage and breast cancer occurrence, estimated from three models including different sets of adjusting variables.

Total alcohol consumption (g/day)	Co.	Ca.	Model 1a OR (95% CI)	Model 2b OR (95% CI)	Model 3c OR (95% CI)
Adolescence (n = 1314; 657 paired case–control)
<5	560	526	1 (ref.)	1 (ref.)	1 (ref.)
5–<15	82	103	1.36 (1.00. 1.87)	1.37 (0.96. 1.95)	1.31 (0.89. 1.94)
≥15	15	28	2.17 (1.11. 4.26)	2.63 (1.22. 5.64)	2.46 (1.04. 5.86)
P for trend			0.011	0.006	0.038
Young adulthood (n = 1410; 705 paired case–control)
<5	470	431	1 (ref.)	1 (ref.)	1 (ref.)
5–<15	184	200	1.22 (0.95. 1.56)	1.35 (1.02. 1.80)	1.13 (0.80. 1.59)
≥15	51	74	1.68 (1.12. 2.51)	1.66 (1.05. 2.60)	0.87 (0.48. 1.58)
P for trend			0.002	0.006	0.932
Adulthood (n = 1448; 724 paired case–control)
<5	431	414	1 (ref.)	1 (ref.)	1 (ref.)
5–<15	221	198	0.96 (0.75. 1.23)	1.03 (0.78. 1.35)	0.96 (0.70. 1.32)
≥15	72	112	1.80 (1.25. 2.60)	2.09 (1.37. 3.20)	1.93 (1.13. 3.29)
P for trend			0.002	0.004	0.030

To test for linear trends across increasing categories of alcohol consumption, the median concentration within each category was included and treated as a continuous variable in the model.

Ca cases, Co controls, OR odds ratio, CI confidence interval.

^aAdjusted for age at the time of recruitment and hospital.

^bAdditionally adjusted for BMI, menopausal status, an interaction term between BMI and menopausal status, calories, age at menarche, number of children, age at first child, smoking status, educational level, chronic diseases, hormone replacement therapy use, previous benign breast lesions, family history of breast cancer, physical activity, adherence to the Mediterranean dietary pattern.

^cAdditionally adjusted for alcohol consumption during the other stages of life (n = 1.278).