Association between physical activity and colorectal cancer risk and prognosis: A meta-analysis

Reverse correlation has been frequently reported between physical activity (PA) and colorectal cancer (CRC) risk and mortality in numerous prospective studies. But the contradictive results make the conclusion elusive. Here we aim to cover the effects of pre-diagnosis in CRC incidence, pre- and post-diagnosis with CRC prognosis and overall mortality. With the eligible prospective studies up to 2015, the associations of pre-diagnosis or post-diagnosis PA with CRC risk, CRC prognosis, or all-cause mortality in the modes of exerciser Vs. non-exerciser and highest PA Vs. lowest PA were investigated by combing the hazard ratios (HRs) in random effects. We also tested the effect of single study on the summary estimates using influence-analysis. Publication bias was evaluated by Egger's and Begg's test. Compared with low pre-diagnosis PA group, the high PA group displayed a reduced risk (HR: 0.78 0.66-0.92) to develop CRC. Our findings also showed that both pre- and post-diagnosis PA were significantly associated with better CRC-specific prognosis and reduced risk of overall mortality in both exerciser Vs. non-exerciser and high PA Vs. low PA modes. With the most updated prospective studies, our systematic meta-analysis confirmed the beneficial effects of PA to CRC incidence and prognosis.