Author(s): INGVARSDOTTIR, T., ARNARDOTTIR, N., ROGNVALDSDOTTIR, V., STEFANSDOTTIR, R., JOHANNSSON, E., Institution: UNIVERSITY OF ICELAND, Country: SWEDEN, Abstract-ID: 1417

INTRODUCTION:
Physical activity (PA) declines markedly during adolescence, however, longitudinal studies assessing the development of time- and domain-specific PA during this period are scarce. A better understanding of which time segments and domains of PA are prone to decline may support the planning of targeted PA interventions. This study aimed to examine changes in objectively measured PA during school hours and after-school hours, as well as changes in subjectively measured organised sports participation (OSP) and non-organised sports participation (non-OSP) between ages 15 and 17. In addition, the study aimed to assess the associations of sex and parental education with the development of these outcomes.
METHODS:
Time-specific PA (in school and after school) was measured objectively with ActiGraph GT3X+ accelerometers, while domain-specific PA (OSP and non-OSP) and parental education were assessed using a questionnaire. Low parental education was defined as having no parent with a university degree, whereas high parental education was defined as having at least one parent with a university degree. Measurements were conducted in 2015 (n = 272, mean age 15.8 ± 0.3 years) and again in 2017 (n = 208, mean age 17.7 ± 0.3 years). Mixed-effect linear and logistic regression models were used to examine the development of time- and domain-specific PA, respectively, by sex and parental education.
RESULTS:
Average decrease in in-school PA was 855 counts per minute (cpm) (p < 0.001), with no differences in the development by sex or parental education (p > 0.05). Average decrease in after-school PA was 544 cpm (p < 0.001). After-school PA decreased more among boys than girls (boys -725 cpm; girls -362 cpm; p = 0.012), however, there were no differences in the development of after-school PA by parental education (p = 0.192). The odds of participating in organised sports were significantly lower at age 17 compared to age 15 (OR = 0.18; p < 0.001 ) with no differences between sexes or levels of parental education (p > 0.05). The odds of participating in non-organised sports did not change between measurements for boys, girls nor participants with high parental education (p > 0.05), however, for those with low parental education the odds of participating in non-organised sports increased between measurements (OR = 2.75; p = 0.033).
CONCLUSION:
Future PA interventions should focus on maintaining in- and after-school PA, and OSP during late adolescence, specifically targeting boys for the maintenance of after-school PA. Targeting promotion of PA to these specific times and domain may help mitigate the decline in adolescents’ PA and improve their lifelong health.