Abstract Background The WHO recommends adults get 150-300 minutes of moderate physical activity (MPA) and/or 75-150 minutes of vigorous physical activity (VPA) per week. Few studies have investigated the independent benefits of PA intensity. Those that have, have found VPA to be protective, while others found excessive VPA to be harmful. These studies have primarily been conducted in high-income countries using recreational PA as the exposure of interest. Purpose To evaluate the effects of activity intensity (walking, MPA and VPA) with all-cause mortality, and cardiovascular disease (CVD) in countries at different economic levels. Methods Participants (35-70 years) were recruited from 21 countries at various stages of economic development. Within each country, urban and rural areas in and around selected cities and towns were identified to reflect the geographical diversity. Physical activity (total, walking, MPA, VPA, as well as recreational vs. non-recreational PA) was assessed using the International Physical Activity Questionnaire. Primary outcomes were mortality plus major CVD (CVD mortality, myocardial infarction, stroke, or heart failure), either as a composite or separately. After excluding those with baseline CVD, cancer and HIV, a total of 136,766 participants who had complete data were analyzed. Models were adjusted for age, sex, urban or rural residency, country income level, education, household wealth index, smoking, baseline chronic diseases, physical impairments and center as a random effect. All models were mutually adjusted for PA intensity. Results During the median follow-up of 11.5 (8.6-12.4) years, there were 9846 deaths and 7900 major cardiovascular events. The mean min/wk for walking, MPA and VPA were 339±539, 640±825 and 103±390, respectively. Increasing walking time had a U-shaped relationship with the composite outcome, such that walking between 300-1200 min/wk had the lowest risk, whereas walking at this amount and ≥1200 min/wk was associated with the lowest risk for mortality (p<0.001 for trend). Both MPA and VPA were linearly associated with the composite outcome and mortality even at levels below the recommended guidelines and up to ≥1800 min/wk (p<0.001 for trend). At each PA category, the hazard ratio for VPA was lower than MPA, such that at ≥1800 min/wk it was 0.64 (0.50,0.81) and 0.70 (0.57,0.85), for VPA and MPA, respectively, for the composite outcome. When VPA was stratified into either recreational or non-recreational VPA, recreational VPA displayed a U-shape relationship, whereas increasing non-recreational VPA continued to confer reduced risk even ≥3000 min/wk (Figure). Conclusion For both MPA and VPA, the benefits began at very low levels of PA and continued to increase with increasing PA. At each amount of PA, the lowest risk was in those participating in VPA. We did not observe any harmful effects of VPA, even at extremely high levels. However, at higher recreational VPA, the benefits began to diminish.