BACKGROUND: Evidence of an association between physical activity (PA) and mortality has mainly focused on leisure-time physical activity (LTPA) and moderate-to-vigorous-intensity physical activity (MVPA). We aimed to assess the associations of total, domain-specific, and intensity-specific PA with all-cause and cause-specific mortality. METHODS: We used baseline PA data from the China Kadoorie Biobank, including 482,067 participants aged 30-79 years from 10 areas in China. PA via self-report was quantified as a metabolic equivalent of task hours per day. Total PA was calculated by summing occupational, commuting, household, and leisure-time PA, and domain- and intensity-specific PAs were also calculated. Cox regression was used to estimate the associations of quintiles of different types of PA with all-cause and cause-specific mortality and adjust for potential confounders. Cause-specific mortalities were also examined in a competing risk analysis. RESULTS: During a median follow-up of 12.1 years, 47,281 deaths occurred. Total PA was inversely associated with the risk of all-cause mortality, with a hazard ratio (HR) (95% confidence interval [95% CI]) of 0.69 (0.67-0.71) in the highest quintile as compared with the lowest quintile. Similar associations were observed for disease-specific mortality risks from cardiovascular disease, cancer, respiratory disease, diabetes, and nervous system disease, with HR (95% CI) for top vs. bottom quintile of PA of 0.68 (0.64-0.71), 0.80 (0.76-0.83), 0.39 (0.35-0.44), 0.44 (0.35-0.55), and 0.52 (0.38-0.73), respectively. In addition, the risk of all-cause mortality was lowered by 34%, 13%, 17%, and 30% for occupational PA, non-occupational PA, low-intensity PA, and MVPA, respectively, when comparing the highest quintile with the lowest quintile. CONCLUSIONS: PA was inversely associated with the risk of all-cause and cause-specific mortality, regardless of domain and intensity. Any PA can bring long-term beneficial health effects.