2007 年 17 巻 2 号 p. 38-44
BACKGROUND: Relationships between consumption of dairy products and death from various types of cancer are largely unknown.
METHODS: Between April 1992 and July 1995, a baseline survey was conducted for 11,349 residents in 12 communities in Japan, which included collection of demographic data and a self-administered food-frequency questionnaire inquiring about three dairy products: milk, butter and yogurt. The subjects were followed prospectively until 2002. Causes of death were identified using death certificates. Hazard ratios (HRs) and their 95% confidence intervals (CIs) for each dairy product were calculated using Cox's proportional hazard models.
RESULTS: Among eight common cancers, only deaths from hematopoietic neoplasm (n=14) were significantly associated with consumption of butter (HR=5.11, 95% CI: 1.40-18.62), though they exhibited a nearly-significant association with milk consumption (HR=3.17, 95% CI: 0.99-10.17), independent of age and sex. Consumption of milk and butter was significantly associated with non-lymphoma deaths (n=9) when adjusted for age and sex (HR=9.86, 95% CI: 1.23-79.19 for milk; and HR=10.04, 95% CI: 2.39-42.18 for butter).
CONCLUSION: The frequencies of butter consumption, and probably that of milk, were correlated with death from hematopoietic neoplasm, particularly from non-lymphomas.
J Epidemiol 2007; 17: 38-44.