Abstract
The number of cases of colorectal cancer in Japan has increased over the past few decades, and incidence rates are now among the highest in the world. The present investigation within the Fukuoka Colorectal Cancer Study, including 778 cases and 767 controls aged 20-74 years, examined the association between physical activity and colorectal cancer risk by subsite. Employment-associated and leisure time physical activity was assessed by a questionnaire and interview. Division of sites into the proximal and distal colon, as well as the rectum, revealed clear site-dependent protective effects, with adjustment for smoking, alcohol consumption, BMI and age. In males, greater job-related physical activity was associated with significant reduction of risk in the distal colon and rectum (P = 0.047 and 0.02, respectively), whereas total and moderate or hard non-job physical activity exerted effects limited to the rectum (P = 0.01 and 0.004, respectively). In females, job-related physical activity and moderate or hard non-job physical activity was also protective, but only in the distal colon. Separate assessment of the influence of BMI 10 years previous to the study showed increase in risk with obesity in males but not in females, limited to distal colon and rectum. The results of the present study indicate that physical activity associated with work and leisure-time exerts beneficial effects in Japanese, but not on the proximal colon.