Abstract
BACKGROUND Physical activity is associated with reduced risk of colorectal cancer, but most epidemiological studies have focused on occupational and recreational physical activity. The evidence for other domains of activity, and for sedentary behaviour, is limited. METHODS Medline, Embase and Web of Science were searched from inception to December 2015 for studies examining domain-specific physical activity or sedentary behaviour and the risk of colon and/or rectal cancer. We extracted maximally adjusted relative risks (RRs) except when RRs not adjusted for body mass index, were also presented. We used random-effects meta-analysis to compute pooled RRs comparing the highest versus the lowest level of exposure. We used meta-regression to assess sources of heterogeneity in estimates. RESULTS We identified 17 cohort and 21 case-control studies, of which 17 had occupational data, 23 had recreational data, three each had data on transport and household physical activity domains, and 6 studies had data on occupational sedentary behaviour. The pooled relative risks (RRs) for colon cancer were 0.74 (95% confidence interval (CI): 0.67, 0.82) for occupational activity, 0.80 (95% CI: 0.71, 0.89) for recreational activity, 0.66 (95% CI: 0.45, 0.98) for transport-related physical activity, 0.85 (95% CI: 0.71, 1.02) for household physical activity, and 1.44 (95% CI: 1.28, 1.62) for occupational sedentary behaviour. For rectal cancer, the pooled RRs were 0.88 (95% CI: 0.79, 0.98) for occupational activity, 0.87 (95% CI: 0.75, 1.01) for recreational activity, 0.88 (95% CI: 0.70, 1.12) for transport-related physical activity, 1.01 (95% CI: 0.80, 1.27) for household physical activity, and 1.02 (95% CI: 0.82, 1.28) for occupational sedentary behaviour. CONCLUSIONS In addition to increasing occupational and recreational physical activity, promoting physical activity during transport and reducing sedentary behaviour in the workplace may also be useful colorectal cancer prevention strategies.
