According to World Cancer Research Fund International/American Institute for Cancer Research, it is 'probable' that dairy products decrease the risk of colorectal cancer (CRC). However, meta-analyses restricted to women have not shown associations between milk intake and risk of CRC. The aim of this study was to examine the association between milk intake and risk of CRC, colon cancer and rectal cancer among women. Data from 81 675 participants in the Norwegian Women and Cancer Cohort Study were included, and multivariable Cox proportional hazard regression models were used to investigate milk intake using two different analytical approaches: one that included repeated measurements and one that included baseline measurements only (872 and 1084 CRC cases, respectively). A weak inverse association between milk intake and risk of colon cancer may be indicated both in repeated measurements analyses and in baseline data analyses. Hazard ratios (HR) for colon cancer of 0·80 (95 % CI 0·62, 1·03, P trend 0·07) and 0·81 (95 % CI 0·64, 1·01, P trend 0·03) and HR for rectal cancer of 0·97 (95 % CI 0·67, 1·42, P trend 0·92) and 0·71 (95 % CI 0·50, 1·01, P trend 0·03) were found when comparing the high with the no/seldom milk intake group in energy-adjusted multivariable models. Our study indicates that there may be a weak inverse association between milk intake and risk of colon cancer among women. The two analytical approaches yielded different results for rectal cancer and hence CRC. Our study indicates that the use of single or repeated measurements in analyses may influence the results.
Keywords: Some of the data in this article are from the Cancer Registry of Norway. The Cancer Registry of Norway is not responsible for the analysis or interpretation of the data presented; CRC colorectal cancer; HR hazard ratios; ICD-7 International Classification of Diseases; NOWAC Norwegian Women and Cancer; Seventh Revision; Colorectal cancer; Milk; Prospective cohorts; Repeated measurements; Women.