Changes in the frequency of vigorous physical activity across four survey cycles: analyses of 12-year repeated cross-sectional surveys of 729,201 adolescents

Evidence on long-term population-level differences in vigorous physical activity (VPA) frequency among adolescents remains limited. We examined differences in VPA frequency across four survey cycles and whether these differences varied by sex and age group. We analyzed repeated cross-sectional Health Behaviour in School-aged Children (HBSC) data from 2006, 2010, 2014, and 2018. The balanced sample comprised 729,201 adolescents aged 11, 13, and 15 years from the same 34 countries or regions; the primary complete-case analysis included 700,039 participants. VPA frequency was modeled as seven categories from never to every day. Survey-weighted multinomial logistic regression incorporated harmonized survey weights and school/class-level primary sampling units, country-by-survey-cycle analysis strata, and categorical country/region fixed effects. We estimated relative risk ratios (RRRs), predicted probabilities, survey-year interactions with sex and age group, and performed 20-dataset multiple-imputation sensitivity analyses. Observed survey-weighted VPA frequency distributions differed across survey cycles (design-based Pearson Chi-square test, P < 0.001). Daily VPA decreased from 19.5% in 2006 to 16.5% in 2018, whereas VPA 4–6 times/week increased from 21.4% to 24.8%. For daily VPA versus never, RRRs were 0.87 (95% CI 0.83–0.90) in 2010, 0.99 (0.95–1.03) in 2014, and 0.83 (0.80–0.86) in 2018 versus 2006. Girls had a lower relative likelihood of daily VPA versus never than boys (RRR = 0.38, 95% CI 0.37–0.39), as did 13-year-olds (RRR = 0.70, 0.68–0.72) and 15-year-olds (RRR = 0.43, 0.41–0.44) versus 11-year-olds. Survey-year-by-sex and survey-year-by-age-group interactions were significant (both P < 0.001), and multiple-imputation results were materially unchanged. VPA did not show a uniform increase or decrease across survey cycles; instead, the frequency distribution shifted, with daily VPA becoming less common and several-times-per-week VPA more common by 2018. Girls and older adolescents had lower probabilities of daily VPA. Continued surveillance should preserve the full frequency distribution and consider sex- and age-specific patterns.