To evaluate Food, Health, & Choices, two 10-month interventions.Cluster-randomized, controlled study with 4 groups: curriculum, wellness, curriculum plus wellness, and control.Twenty elementary schools (5/group) in New York City.Fifth-grade students (n = 1,159). At baseline, 44.6% were at the ≥85th body mass index (BMI) percentile for age and 86% qualified for free or reduced-price lunch.Curriculum was 23 science lessons based on social cognitive and self-determination theories, replacing 2 mandated units. Wellness was classroom food policy and physical activity bouts of Dance Breaks.For obesity, age- and sex-specific BMI percentiles were used (anthropometric measures). The researchers also employed 6 energy balance-related behaviors and 8 theory-based determinants of behavior change (by questionnaire).Pairwise adjusted odds in hierarchical logistic regression models were determined for >85th BMI percentile. Behaviors and theory-based determinants were examined in a 2-level hierarchical linear model with a 2 × 2 design for intervention effects and interactions.Obesity showed no change. For behaviors, there was a negative curriculum intervention change in physical activity (P = .04). The wellness intervention resulted in positive changes for sweetened beverages frequency (P = .05) and size (P = .006); processed packaged snacks size (P = .01); candy frequency (P = .04); baked good frequency (P = .05); and fast food frequency (P = .003), size (P = .01), and combo meals (P = .002). Theory-based determinants demonstrated no change.The findings of the lack of a decrease in obesity, behavior changes only for the wellness intervention, and no changes in theory-based determinants warrant further research.