Ecological correlates of sedentary time in adolescents: Individual, interpersonal, home & neighborhood

Purpose: Ecological models of physical activity (PA) have shown influences on behavior at each level of the model. Ecological models for sedentary time (ST) have not been tested. This study aimed to examine the independent association of personal, social and environmental factors with objectively-measured sedentary time in a large, heterogeneous sample of US adolescents.

Methods: Adolescents (N=880, age 14.1 ±1.4) recruited from neighborhoods varying in walkability and income in two regions of the US completed a survey and wore an Actigraph accelerometer for 7 days. Average daily minutes <100 counts per minute was the outcome. Survey items included child and parent demographics, self-efficacy for reducing ST, enjoyment of ST, pros and cons of reducing ST, parent ST, presence of others during ST, ST rules, having a TV in the bedroom, ST-related personal electronics, the Neighborhood Environment Walkability Scale (Youth; NEWSY) and GIS walkability index. Mixed effect linear regression analyses adjusted for census block group. Final models retained all significant variables, adjusting for accelerometer measured PA and wear time.

Results: The final model included child age and gender, ST enjoyment, parental ST, personal electronics, and TV in the bedroom. All these variables remained significant (p<.025) except for gender. Reported and GIS environment variables were not related to ST. All variables had a positive relationship with ST except TV in the bedroom.

Conclusions: Individual-level, psychosocial, interpersonal factors, and home, but not neighborhood, environments are related to adolescent ST. Intervention should therefore consider addressing each level of the ecological model.