Best practices to evaluate the impact of school health policy on childhood obesity

INTRODUCTION: Evaluation of health outcomes linked to obesity policy is essential to advancing the science of policy evaluation and policy change. A methodological limitation in obesity research is that body mass index (BMI) is used as the sole criterion for obesity-mortality associations and health outcomes - without considering excess adipose tissue location. It is possible to be at increased risk for chronic disease without being overweight or obese as defined by BMI standards. METHODS: Anthropometric measures and BMI were assessed on 1,136 elementary school-age children in Mississippi. RESULTS: The prevalence of overweight and obesity (BMI > 85th percentile) was 47.1% (18.3% overweight and 28.8% obese). For this sample, 59.9% and 42.0% were “at risk” for weight-related chronic disease based on waist circumference (WC > 90th percentile) and waist-to-height ratio (WtHR > 0.5), respectively. There were differences among three diverse school districts in total body weight (p < 0.0005), BMI (p < 0.0005), WC (p < 0.0005), and WtHR (p < 0.0005). CONCLUSIONS: It is clear that the obesity epidemic continues to manifest in minority children of low SES and does so at an increasingly young age. In 2007, the State legislature in Mississippi passed the “Mississippi Healthy Students Act” that requires increased physical activity and healthy dietary environments for children at school. Measureable health outcomes should include WtHR because it accounts for growth in both WC and height over age, and may be a more accurate tracking indicator of fat distribution used to predict subsequent health risk.