Despite global concerns regarding physical inactivity, limited cross-national evidence exists to compare adolescents’ physical activity participation. We analysed data from 52 high- and low-middle income countries, with activity undertaken inside and outside of school in 2015. We investigated gender and socioeconomic disparities, and additionally examined correlations with country-level indices of physical education (PE) curriculum time allocation, wealth, and income inequality.

We compared adolescents’ reported activity levels inside and outside of school using nationally representative cross-sectional data from 52 high- and low-middle income countries (N = 347,935)—the Programme for International Student Assessment (PISA) in 2015. Students reported average attendance (days/week) in PE classes, and the days/week engaged in moderate activity (MPA) and vigorous activity (VPA) outside of school. We also compared gender and socioeconomic disparities, and additionally examined correlations with purported determinants—country-level estimates of PE curriculum time allocation, wealth, and income inequality.

Average activity levels differed substantially both between and within regions, with potentially important differences in distributions identified—such as a bimodal distribution in the U.S. and Canada in PE. Males were more active than females, as were those from households with higher rather than lower household wealth; these disparities were modest for PE, but higher for moderate and vigorous activity outside school—there was strong evidence for heterogeneity in the magnitude of these disparities (e.g., I² > 95% for gender differences across all countries). PE class attendance was positively correlated with PE curriculum time allocation (rho = 0.36); activity outcomes were inconsistently associated with country-level wealth and income inequality.

Our findings reveal extensive cross-country differences in adolescents’ physical activity; in turn, these highlight policy areas that could ultimately improve global adolescent health, such as the incorporation of minimum country-level PE classes, and the targeting of gender and socioeconomic disparities in activity conducted outside of school. Our findings also highlight the utility of educational databases such as PISA for use in global population health research.