Abstract P322: Physical Activity Patterns Among Indian Adolescents: Independent and Joint Associations of Gender and Socio-Economic Status

Abstract

Introduction: The increasing global burden of CVD mortality is primarily attributable to the rising number of deaths in developing countries; India alone accounted for 15% of global CVD deaths in 2016. Little is known about patterns and predictors of physical activity (PA), a key modifiable risk factor for CVD, among Indian adolescents. Existing research has focused on play- and travel-related PA, which do not fully capture adolescent activity, and socio-demographic differences in PA are not well understood. We examined independent and joint associations of gender and socio-economic status (SES) with PA across multiple activity domains. Hypothesis: We assessed the hypotheses that 1) the types of PA in which adolescents engage will differ by gender and SES, and 2) gender differences in PA will vary by SES. Methods: We recruited 395 adolescents, ages 13-16, from six schools in Vijayapura, India. We classified public school students as low SES and private school students as high SES. We measured PA using adolescent-reported 24-hour time-use surveys. Activities were categorized into three domains: chores, errands, and jobs; play; and travel. We assigned each activity a metabolic equivalent (MET) value to assess moderate-to-vigorous PA (MVPA). Negative binomial regression modeled the adjusted count of PA minutes per day in each domain, and logistic regression modeled the probability of engaging in ≥60 minutes of MVPA in the prior 24 hours (WHO recommended level for ages 5-17). We assessed additive interaction between gender and SES with the relative excess risk due to interaction (RERI). Results: The proportion of adolescents engaging in ≥60 minutes of MVPA was higher among low SES (79.8% boys, 76.8% girls) than high SES (50.0% boys, 32.3% girls) adolescents. In adjusted models, there were no gender differences in the likelihood of engaging in ≥60 minutes of MVPA. However, girls had twice the PA minutes in chores, errands, and jobs (e(β)=1.98, 95% CI:1.32-2.98), while boys had twice the PA minutes in play (e(β)=2.11, 95% CI:1.23-3.62). Low versus high SES adolescents had higher PA across domains. Additive interaction between gender and SES was present in chores, errands, and jobs: while girls were more active in this domain across SES, gender differences were greater among low SES adolescents (RERI=2.53, 95% CI: 0.31-4.76). Conclusions: A far greater proportion of low SES adolescents engaged in recommended levels of MVPA. While there were no gender differences in the likelihood of engaging in ≥60 minutes of MVPA, girls and boys may accrue different advantages as a result of engaging in substantively different types of PA. Interaction in the chores, errands, and jobs domain suggests that gender hierarchies may emerge in the presence of socio-economic constraints, and lose salience as available resources increase. PA interventions must consider the role of gender- and SES-based facilitators and barriers.