Abstract
BACKGROUND: Schools are the most frequent target for intervention programs aimed at preventing child obesity; however, the overall effectiveness of these programs has been limited. It has therefore been recommended that interventions target multiple ecological levels (community, family, school and individual) to have greater success in changing risk behaviors for obesity. This study examined the immediate and short-term, sustained effects of the Switch program, which targeted three behaviors (decreasing children's screen time, increasing fruit and vegetable consumption, and increasing physical activity) at three ecological levels (the family, school, and community). METHODS: Participants were 1,323 children and their parents from 10 schools in two states. Schools were matched and randomly assigned to treatment and control. Measures of the key behaviors and body mass index were collected at baseline, immediately post-intervention, and 6 months post-intervention. RESULTS: The effect sizes of the differences between treatment and control groups ranged between small (Cohen's d = 0.15 for body mass index at 6 months post-intervention) to large (1.38; parent report of screen time at 6 months post-intervention), controlling for baseline levels. There was a significant difference in parent-reported screen time at post-intervention in the experimental group, and this effect was maintained at 6 months post-intervention (a difference of about 2 hours/week). The experimental group also showed a significant increase in parent-reported fruit and vegetable consumption while child-reported fruit and vegetable consumption was marginally significant. At the 6-month follow-up, parent-reported screen time was significantly lower, and parent and child-reported fruit and vegetable consumption was significantly increased. There were no significant effects on pedometer measures of physical activity or body mass index in the experimental group. The intervention effects were moderated by child sex (for fruit and vegetable consumption, physical activity, and weight status), family involvement (for fruit and vegetable consumption), and child body mass index (for screen time). The perception of change among the experimental group was generally positive with 23% to 62% indicating positive changes in behaviors. CONCLUSION: The results indicate that the Switch program yielded small-to-modest treatment effects for promoting children's fruit and vegetable consumption and minimizing screen time. The Switch program offers promise for use in youth obesity prevention.