Participation in Domains of Physical Activity Among Australian Youth During the Transition From Childhood to Adolescence: A Longitudinal Study

in Journal of Physical Activity and Health
Restricted access

Background: Information about the domains of physical activity (PA) that are most prone to decline between late childhood (11 y), early adolescence (13 y), and mid-adolescence (15 y) may support more targeted health promotion strategies. This study explored longitudinal trends in nonorganized PA, organized PA, active transport and active chores/work between childhood and adolescence, and potential sociodemographic moderators of changes. Methods: Data were sourced from the Longitudinal Study of Australian Children (n = 4108). Participation in PA domains was extracted from youth time-use diaries. Potential moderators were sex, Indigenous status, language spoken at home, socioeconomic position, and geographical remoteness. Results: A large quadratic decline in nonorganized PA (−48 min/d, P < .001) was moderated by sex (β = 5.55, P = .047) and home language (β = 8.55, P = .047), with girls (−39 min/d) and those from a non-English speaking background (−46 min/d) declining more between 11 and 13 years. Active chores/work increased between 11 and 13 years (+4 min/d, P < .001) and then stabilized. Active transport increased among boys between 11 and 13 years (+6 min/d, P < .001) and then declined between 13 and 15 years (−4 min/d, P < .001). Organized PA remained stable. Conclusions: The longitudinal decline in PA participation may be lessened by targeting nonorganized PA between childhood and adolescence. Future interventions may target girls or those from non-English speaking backgrounds during this transition.

Kemp, Parrish, and Cliff are with Early Start, University of Wollongong, Wollongong, NSW, Australia. Kemp and Parrish are also with the School of Health and Society, Faculty of Social Sciences, University of Wollongong, Wollongong, NSW, Australia. Parrish, Batterham, and Cliff are also with the Illawarra Health and Medical Research Institute, University of Wollongong, Wollongong, NSW, Australia. Batterham is with the School of Mathematics and Applied Statistics, Faculty of Engineering and Information Sciences, University of Wollongong, Wollongong, NSW, Australia; and the National Institute for Applied Statistical Research Australia, University of Wollongong, Wollongong, NSW, Australia. Cliff is also with the School of Education, Faculty of Social Sciences, University of Wollongong, Wollongong, NSW, Australia.

Kemp (bjk687@uowmail.edu.au) is corresponding author.

Supplementary Materials

    • Supplementary File 1 (PDF 297 KB)
    • Supplementary File 2 (PDF 222 KB)