Background The family has been acknowledged as central to developing physical activity (PA) beliefs and behaviours. However, increased diversity in family structures has developed over the last decades. This study examines the association between family structure and PA among adolescents and cross-national variations in the associations.

Methods The data are from the 2013/14 Health Behaviours in School-Aged Children study, involving nationally representative samples of 11-, 13- and 15-year-olds (n = 211,798) from 40 countries. Multilevel Poisson regression analysis was used to examine the associations between family structure and moderate to vigorous physical activity (MVPA) and vigorous physical activity (VPA) by age, gender, socioeconomic status (SES), and geographic region.

Results Living with one versus two parents was associated with a reduced likelihood of daily 60 min MVPA for boys (IRR = 0.96, 95% CI: 0.92, 0.99) and ≥ 4 times/week VPA (IRR 0.93, 95% CI: 0.91, 0.95). This impact on MVPA differed across individual-level SES (high SES; IRR = 0.92, (p <0.05), low SES; IRR = 1.04, (ns)), and was for VPA only significant for those with siblings (IRR = 0.93, 95% CI: 0.91, 0.96). Cross-country variations in the association between living with one versus two parents were observed, most pronounced for VPA. These differences varied by region, primarily explained by country-level SES differences between regions. The likelihood of daily 60 min MVPA also increased with siblings in the main house (IRR 1.11, 95% CI: 1.07, 1.14), and ≥ 4 times/week VPA decreased with grandparents in the main house (IRR 0.91, 95% CI: 0.89,0.94).

Conclusions Family structure correlated with PA, but cross-country differences exist. The findings are relevant for the development of policies and programs to facilitate PA, especially in countries where living with one versus two parents was unfavourable. Additional country-specific research is needed to identify challenges for engaging in PA related to family structure.