Adolescent diet and breast cancer in Utah.Cancer Res 1989; 49(8):2161-7CR
A population-based case-control study was conducted to assess the association between breast cancer risk, body mass index (BMI) and adolescent dietary fat and fiber consumption. Data were collected in Utah from white female cases (N = 172) and controls (N = 190) between the ages of 20 and 54 years. Odds ratios (OR) and 95% test-based confidence intervals (CI) were determined by multiple-logistic regression analysis controlling for age, education, age at menarche, and age at first pregnancy. Menopausal status was identified as an effect modifier, therefore, separate analyses were performed for pre and postmenopausal groups. An elevated risk (OR = 2.9 for highest quartile versus lowest, CI = 1.1-8.1) was associated with a larger BMI at age 12 in premenopausal women; a larger adult BMI lowered the odds ratio (OR = 0.4, CI = 0.2-1.0 for highest quartile versus lowest) in premenopausal women; BMI did not alter risk in postmenopausal women. Although not statistically significant, high fat intake consistently lowered the odds ratios below 1.0 in premenopausal women in the upper three quartiles compared to the lowest fat intake referent quartile (OR = 0.7, CI = 0.2-2.1 for highest versus lowest quartile) but was inconsistent in postmenopausal women (OR = 0.7, CI = 0.2-2.7 for highest versus lowest quartile). When fat intake was assessed by its component parts, fat from milk, cheese and yogurt reduced the odds ratios in both premenopausal (OR = 0.4, CI = 0.1-1.1 for highest versus lowest quartile) and postmenopausal women (OR = 0.2, CI = 0.0-0.8). In postmenopausal women, high fiber intake produced elevated odds ratios in all three upper quartiles (OR = 6.6, CI = 1.5-29.6 for highest versus lowest quartile), while fiber from grains resulted in a decreased risk in both premenopausal (OR = 0.2, 95% CI = 0.2-0.7 for highest versus lowest quartile) and postmenopausal women (OR = 0.7, 95% CI = 0.3-2.0). The possibility of biased estimates from low response rates (cases = 60%, controls = 61%), potential recall bias, and some lack of precision in the dietary instrument should be considered. It appears from these analyses that the relation of breast cancer to dietary intake, especially during adolescent years, is not clear, and that risk associated with fat or fiber intake may be affected by the nutrient source.