The effects of socio-economic status on physical activity participation in Hong Kong adolescents: asocial ecological approach

Abstract

Motivations: Socio-economic status (SES) is a major source of health disparities. Those

who lived in resource-deficient social and physical environments and with low individual

income have been found to be at increased risk of physical inactivity. In Hong Kong there

is virtually no research on the effect of SES on adolescent physical activity (PA). It is

therefore important to explore the SES differentials in PA among Hong Kong adolescents

and identify the mechanisms underlying such differentials. It is also important to examine

the extent to which SES acts as a moderator and mediator of associations between various

potential determinants of adolescent PA.

Methods: A hundred eighty-one adolescents aged 12-18 were recruited in the study. This

study applied an ecological framework to study the SES effects on adolescents’ PA

participation. Physical activity was measured both objectively using accelerometers and

subjectively using self-reports. Individual, social and school environment factors were

assessed by validated self-report measures. Neighborhood built environmental factors

were assessed using Geographic Information Systems (GIS) data. Individual-level SES

was based on parents’ report of monthly household income, while area-level SES was

based on census data on median household income of selected study neighborhoods.

Generalized linear models with robust standard errors were used to assess associations of

individual, social and environmental factors with adolescents’ PA. They were also used

to assess the moderating of SES and mediators of SES-PA relationships.

Results: The results supported the hypothesis that individual, social and environmental

factors would all contribute to the explanation of adolescents’ PA. Moderation analyses

showed that the PA level of adolescents living in low SES areas might have been

negatively affected by the presence of steep streets in the neighborhood environment and

the lack of sports facilities. These effects were not observed in adolescents living in high

SES areas. In contrast, only adolescents living in high-SES areas showed positive

associations of school-based social support and school PA-related environment with PA.

Mediation analyses showed that SES differences in adolescents PA participation could be

explained by social support from family and access to sports/recreational facilities

in/around the school.

Significance: The current findings would imply that environmental interventions at the

neighborhood level might benefit low-SES groups of adolescents as they seem to be more

influenced by aspects of the neighborhood environment. By contrast, high-SES

adolescents might be more reactive to school-based environmental interventions. To

narrow the SES gap in PA participation among Hong Kong adolescents, interventions

should encourage low-SES families to provide more support to their children by

widening the available range of opportunities for PA, encouraging their children to

participate in PA, and acting as role models. The findings from this study also suggest

that the provision of more PA facilities in/around school neighborhood in low-SES areas

could contribute to increasing PA participation among the adolescent population of such

areas and, thus, narrow the gap between high- and low-SES groups of adolescents.