Physical activity, alcohol consumption, and digestive system cancer risk: a large prospective cohort study

Background: Whether physical activity (PA) could modify the association between alcohol consumption and digestive system cancer (DSC) is still unclear. The primary aim of this study was to evaluate the independent and joint effects of PAs in various intensities and alcohol consumption on DSC risk. Methods: A total of 305,699 participants free of cancer at baseline in the UK Biobank were included in this study. PAs, including total PA (TPA), vigorous PA (VPA), moderate-to-vigorous PA (MVPA), moderate PA (MPA), walking and leisure-time PA (LTPA), were self-reported at baseline. Multivariate Cox regression models were used to evaluate the independent and joint associations between PAs, alcohol consumption, and outcomes (DSC, digestive tract cancer, and digestive accessory organ cancer). Results: Except walking, higher levels of PAs were associated with a lower risk of DSC. The inverse associations were more evident for VPA (hazards ratio [HR] = 0.90 [95% confidence interval {CI} = 0.84–0.95]) and MVPA (HR = 0.86 [95% CI = 0.81–0.92]) than TPA, MPA, and LTPA. No additive or multiplicative interactions between PA and alcohol consumption on DSC risk were detected. When stratified by PA, harmful drinking was consistently associated with increased DSC risk. Similar trends were found for digestive tract cancer and digestive accessory organ cancer. Conclusions: PA, especially VPA and MVPA, was inversely associated with DSC risk. However, a high level of PA could not counteract the detrimental effect of alcohol consumption on DSC risk. Our results support current recommendations about alcohol cessation even in physically active individuals.