Lifestyle factors and colorectal cancer risk (2): a systematic review and meta-analysis of associations with leisure-time physical activity

Objective Increased physical activity may decrease the risk of colorectal cancer. As a prerequisite to the determination of lifestyle attributable risks, we performed a systematic review and meta-analysis of prospective observational studies to quantify gender-specific risk associated with increased leisure-time physical activity (LT-PA). Method We searched MEDLINE and EMBASE (to December 2007), and other sources, selecting reports based on strict inclusion criteria. We used random-effects meta-analyses to estimate summary risk ratios (RR) and 95% confidence intervals (95% CI) for uppermost vs lowermost categories of physical activity. To investigate dose-response, we explored risks ratios as a function of cumulative percentiles of physical activity distribution. Results Fifteen datasets from 14 articles, including 7873 incident cases, were identified. For colon cancer, there were inverse associations with LT-PA for men (RR: 0.80; 95% CI: 0.67-0.96) and women (0.86; 0.76-0.98). LT-PA did not influence risk of rectal cancer. The dose-response analysis was consistent with linear pattern reductions in risk of colon cancer in both genders. There was evidence of moderate between-study heterogeneity but summary estimates were broadly consistent across potential confounding factors. Conclusion Increased LT-PA is associated with a modest reduction in colon but not rectal cancer risk; a risk reduction, which previously may have been overstated. LT-PA only interventions in public health cancer prevention strategies are unlikely to impact substantially on colorectal cancer incidences.