Determinants of objectively measured physical activity in rural East Timorese children

The human juvenile period evolved as a period of learning and physical development in a family environment that subsidizes the costs of these processes. Children allocate energy to physical activity, maintenance, and growth. How energy is allocated has consequences for adult body size and other life-history traits. In subsistence agriculture populations, where child contributions to the household economy are common and energy availability is low, trade-offs in energy expenditure between activity and growth may help explain poor growth.

Using accelerometry, we measured physical activity over 2 years in 88 free-living children aged 5-19 years in two ecologically varying communities in rural Timor-Leste. We model characteristics related to variation in activity, and subsequently, activity is modeled against growth, illness, and aspects of household and local ecology using linear mixed models.

Physical activity in Timorese children is characterized by high levels of moderate ( $$ = 8.8 h/day), no sustained vigorous, and little sedentary activity ( $$ = 4.6 h/day). Children in the mountainous community show a slight trade-off between activity and growth (P = .077). Males down-regulate both growth and activity relative to females. Variation in household characteristics does not predict child activity. Both activity and growth are lower in the mountainous community than in the flat, coastal community.

Household demands on child behavior may constrain children's ability to moderate activity relative to nutritional status. Activity in this population is high relative to other subsistence populations, possibly because children face the dual pressures of contributing to household subsistence and attending school.