Abstract
Alcohol consumption is common across Western culture, despite its associations with adverse health outcomes, including cancer and cardiovascular disease (CVD). Physical activity (PA) has beneficial effects on many alcohol related outcomes, with data suggesting PA may offset the association between alcohol consumption and mortality. This study examined the joint associations of PA and alcohol on all-cause, CVD and cancer mortality. Participants were recruited between 2006 and 2010 in the United Kingdom. Alcohol consumption was categorised based on current UK guidelines (14 units/week). PA was categorised based on the Metabolic Task Equivalent of PA as low, moderate and high. Data were analysed using Cox proportional-hazard models. The final analysis, conducted in 2019, included 297,988 adults aged ≥40. Over an average follow-up of 6.9 years, 6079 deaths were recorded, including 1219 CVD deaths and 3112 cancer deaths. We observed greater point estimates for risk of all-cause mortality among low PA individuals who consumed alcohol at the same level as active individuals. For example, low PA participants who reported alcohol consumption ≥double guidelines had a greater HR (1.55, 95% CI 1.25, 1.93) than active individuals (moderate PA HR 1.21, 95% CI 0.95, 1.54; high PA HR 1.21, 95% CI 1.00, 1.46). Considering CVD, we observed a similar trend with lower point estimates of risk of mortality among active individuals. We found some evidence that PA modified the associations of alcohol and all-cause and CVD mortality in this large population sample of British adults.