Dairy intake during high school and risk of colorectal adenomas in women

Abstract

Listen

Translate article

Introduction There is substantial evidence including from meta-analyses that higher dairy intake during adulthood is associated with lower risk of colorectal cancer. In contrast, little is known about the association between adolescent dairy intake and risk of colorectal neoplasia. Only two studies have previously investigated the association between dairy intake in early life and later risk of colorectal neoplasia, with conflicting results, but none used a validated dietary assessment tool to define exposure. Therefore, with our study we aimed to investigate the association between dairy intake during adolescence and the risk of colorectal adenomas later in life using data from the large, prospective female cohort Nurses’ Health Study II (NHSII) in which adolescent diet was inquired retrospectively by a validated food frequency questionnaire, and participants were followed prospectively for incidence of colorectal adenomas. Methods The study population consisted of 42,975 women aged between 34 and 51 years who had reported their adolescent diet in a validated food frequency questionnaire about diet during high school (HS-FFQ) in 1998 and had undergone at least one lower bowel endoscopy during the study period, i.e. between the return of the HS-FFQ and May 31st 2011. To investigate the association between adolescent dairy intake and risk of later colorectal adenoma, we used a period-based analysis approach taking into account that NHSII participants may have undergone multiple lower bowel endoscopies during the follow-up period. We used multivariable-adjusted logistic regressions for clustered data (each participant poses a cluster) to calculate odds ratios (ORs) and corresponding 95% confidence intervals (CI). Models were adjusted for potentially confounding factors and dairy intake during adulthood. Results Between 1998 and 2011, 2456 women were diagnosed with at least one adenoma. High dairy consumption during high school was not significantly associated with risk of colorectal adenomas (OR highest versus lowest quintile 0.89, 95% CI 0.76, 1.04). Adolescent dairy consumption was not associated with risk of proximal (OR 1.01, 95% CI 0.81,1.27) adenomas, but a statistically significant inverse association was observed for distal (including rectal) adenomas (OR 0.81, 95% CI 0.66,1.00; P-trend 0.04). Furthermore, the risk of advanced adenomas (large or villous adenomas) was lower for women with high versus low adolescent dairy intake (OR 0.70, 95% CI 0.50, 0.96; P-trend = 0.03). No association between adolescent dairy consumption and risk of non-advanced adenomas was observed (OR 0.97, 95% CI 0.79, 1.20). ORs for distal and advanced adenomas were no longer statistically significant after additional adjustment for dairy calcium intake during high school. When analyzed separately by individual dairy foods, milk, yoghurt or cheese intakes during high school were not significantly associated with risk of colorectal adenoma. Conclusions The findings suggest that dairy intake during adolescence may play a role for later risk of colorectal adenoma, independent of adult dairy intake.