At baseline in 1987–1990, the mean daily whole grain consumption was 2.6 servings. Whole grain consumption varied considerably, with median daily consumption in the top category being 5.0 servings compared with 1.1 servings in the bottom category (). Women with a high consumption of whole grains were older, leaner, and more likely to have a postsecondary education. Greater consumption of whole grains was also associated with higher consumption of fruits and vegetables, but with a lower intake of saturated fat.
Age-standardised baseline characteristics according to consumption of whole grainsa
During a mean follow-up of 14.8 years (911
person-years), we ascertained 805 incident cases of colorectal cancer, including 547 colon cancers (249 proximal colon, 170 distal colon, and 128 unspecified) and 252 rectal cancers; six cases were diagnosed with both colon and rectal cancer. We observed an inverse association between consumption of whole grains and risk of colon cancer, but not of rectal cancer (). Compared with women in the lowest category of whole grain consumption (<1.5
), the age-adjusted RR of colon cancer for women in the highest category (
) was 0.67 (95% CI, 0.49–0.91; P
-value for trend=0.03). The RR was virtually unchanged after further adjustment for education, body mass index, and intakes of total energy, saturated fat, calcium, red meat, fruits, and vegetables (RR, 0.67; 95% CI, 0.47–0.97). Additional controlling for alcohol consumption, family history of colorectal cancer, smoking, physical activity, and use of aspirin and multivitamins did not alter the findings essentially (RR, 0.69; 95% CI, 0.48–0.99). We conducted further analyses after excluding cases that occurred during the first 2 years of follow-up (). The association of whole grain consumption with colon cancer risk became slightly stronger; compared with women in the bottom category of whole grain consumption, those in the top category had a significant 35% lower risk. High consumption of whole grains was associated with a lower risk of both proximal colon (RR, 0.69; 95% CI, 0.40–1.20) and distal colon cancer (RR, 0.54; 95% CI, 0.27–1.08).
RR and 95% CI of colorectal cancer according to consumption of whole grains
In analyses of individual whole grain foods, only hard whole grain rye bread had a statistically significant inverse association with colon cancer risk. After excluding cases that occurred within 2 years of follow-up, the multivariate RR of colon cancer for women who consumed two or more slices per day of hard whole grain bread compared with those who consumed less than four slices per week was 0.74 (95% CI, 0.55–0.98; P-value for trend=0.02). Each daily increment of one slice of hard whole grain bread was associated with a 12% reduction in colon cancer risk (multivariate RR, 0.88; 95% CI, 0.78–0.99).
Refined grain consumption was not appreciably associated with risk of colon or rectal cancer. The multivariate RRs for the highest (
) compared with the lowest category (<0.5
) of refined grain consumption were 1.29 (95% CI, 0.92–1.82) for colon cancer and 0.82 (95% CI, 0.48–1.40) for rectal cancer.
We also examined the association between cereal fibre intake and risk of colon cancer (). After controlling for age, education, body mass index, and intakes of total energy, saturated fat, calcium, red meat, fruits, and vegetables, women in the top quintile of cereal fibre intake had a 27% (RR, 0.77; 95% CI, 0.57–1.03) reduced risk of colon cancer compared with those in the bottom quintile. Further adjustment for intakes of vitamin C, vitamin E, β-carotene, folate, vitamin B6, and magnesium did not materially change the results, although the confidence interval widened (RR, 0.76; 95% CI, 0.53–1.08). There was no association of cereal fibre intake with rectal cancer risk (RR, 0.99; 95% CI, 0.64–1.53).
Figure 1 Multivariate RR and 95% CI of colon cancer according to cereal fibre intake. Multivariate RRs were adjusted for age, body mass index (quartiles), education (less than high school, high school, or university), total energy intake (continuous), (more ...)
To elucidate whether the inverse relation between whole grain consumption and colon cancer risk could be attributed to cereal fibre, we included intakes of whole grains and cereal fibre simultaneously in a multivariate model (adjusted for the same variables as in ). In this model, the RR of colon cancer comparing the extreme categories of whole grain consumption was attenuated from 0.65 to 0.75 (95% CI, 0.49–1.15).