Association between competitive food and beverage policies in elementary schools and childhood overweight/obesity trends: differences by neighborhood socioeconomic resources.

IMPORTANCE

To our knowledge, few published studies have examined the influence of competitive food and beverage (CF&B) policies on student weight outcomes; none have investigated disparities in the influence of CF&B policies on children's body weight by school neighborhood socioeconomic resources.

OBJECTIVE

To investigate whether the association between CF&B policies and population-level trends in childhood overweight/obesity differed by school neighborhood income and education levels.

DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study, from July 2013 to October 2014, compared overweight/obesity prevalence trends before (2001-2005) and after (2006-2010) implementation of CF&B policies in public elementary schools in California. The study included 2 700 880 fifth-grade students in 5362 public schools from 2001 to 2010.

EXPOSURES

California CF&B policies (effective July 1, 2004, and July 1, 2007) and school neighborhood income and education levels.

MAIN OUTCOMES AND MEASURES

Overweight/obesity defined as a body mass index at or greater than the 85th percentile for age and sex.

RESULTS

Overall rates of overweight/obesity ranged from 43.5% in 2001 to 45.8% in 2010. Compared with the period before the introduction of CF&B policies, overweight/obesity trends changed in a favorable direction after the policies took effect (2005-2010); these changes occurred for all children across all school neighborhood socioeconomic levels. In the postpolicy period, these trends differed by school neighborhood socioeconomic advantage. From 2005-2010, trends in overweight/obesity prevalence leveled off among students at schools in socioeconomically disadvantaged neighborhoods but declined in socioeconomically advantaged neighborhoods. Students in the lowest-income neighborhoods experienced zero or near zero change in the odds of overweight/obesity over time: the annual percentage change in overweight/obesity odds was 0.1% for females (95% CI, -0.7 to 0.9) and -0.3% for males (95% CI, -1.1 to 0.5). In contrast, in the highest-income neighborhoods, the annual percentage decline in the odds of overweight was 1.2% for females (95% CI, 0.4 to 1.9) and 1.0% for males (95% CI, 0.3 to 1.8). Findings were similar for school neighborhood education.

CONCLUSIONS AND RELEVANCE

Our study found population-level improvements in the prevalence of childhood overweight/obesity that coincided with the period following implementation of statewide CF&B policies (2005-2010). However, these improvements were greatest at schools in the most advantaged neighborhoods. This suggests that CF&B policies may help prevent child obesity; however, the degree of their effectiveness is likely to depend on socioeconomic and other contextual factors in school neighborhoods. To reduce disparities and prevent obesity, school policies and environmental interventions must address relevant contextual factors in school neighborhoods.