The effect of school district nutrition policies on dietary intake and overweight: a synthetic control approach.

School nutrition policies aim to eliminate ubiquitous unhealthy foods and beverages from schools to improve adolescent dietary behavior and reduce childhood obesity. This paper evaluates the impact of an early nutrition policy, Los Angeles Unified School District's food-and-beverage standards of 2004, using two large datasets on food intake and physical measures. I implement cohort and cross-section estimators using "synthetic" control groups, combinations of unaffected districts that are reweighted to closely resemble the treatment unit in the pre-intervention period. The results indicate that the policy was mostly ineffective at reducing the prevalence of overweight or obesity 8-15 months after the intervention but significantly decreased consumption of two key targets, soda and fried foods. The policy's impact on physical outcomes appears to be mitigated by substitution toward foods that are still (or newly) available in the schools.