Background: School procurement policies - e.g., free/reduced price provision of healthful foods/beverages, quality standards for competitive foods/beverages, or quality standards for school meals - are increasingly being used to promote healthy diets in kids. However, their effectiveness has not been systematically evaluated.
Methods: We used MOOSE and PRISMA guidelines to systematically search multiple online databases for original interventions (randomized, quasi-experimental) assessing influence of school procurement policies, alone or as part of multi-component strategies, on dietary intakes in children. Data were extracted independently and in duplicate. Inverse variance-weighted meta-analysis was used to pool estimates. Pre-specified sources of heterogeneity (study design, location; intervention duration, coverage, components; outcome type, ascertainment) were analyzed using meta-regression and subgroup analysis. Funnel plots, Begg’s, and Egger’s tests evaluated potential publication bias.
Results: From 6,193 abstracts, 76 interventional studies met inclusion criteria. Many were multicomponent. Thirty-two assessed free/reduced price provision of healthful foods/beverages, mostly in cafeterias or classrooms, with average duration 18 mo. In pooled analysis, free/reduced price provision of fruits and vegetables increased fruit intake by 0.22 servings/d (n=14 studies; 95% CI: 0.10, 0.34) and total fruit and vegetable intake by 0.28 servings/d (n=12; 0.07, 0.49), but not vegetable intake alone (n=8; 0.01 servings/d [-0.03, 0.05]). Twenty-seven interventions evaluated policies on competitive foods/beverages (most often sugar-sweetened beverages), with average duration 23 mo. Strategies included restrictions/bans, quality standards, or both. These interventions reduced sugar-sweetened beverage intake by 0.11 (12-oz) servings/d (n=7; -0.16, -0.05). Thirty-two interventions assessed quality standards for school meals (lunch and/or breakfast), with average follow-up 28 mo. Standards were typically based on types of foods, nutrient content, and/or portion size. Dietary targets varied appreciably, and results were generally inconsistent across studies, with no significant overall pooled effect. Findings for secondary outcomes of food content, food availability, and adiposity will be presented. Statistical heterogeneity in these analyses was variable; meta-regression did not identify significant sources. Little evidence for publication bias was seen.
Conclusions: These findings support efficacy of school procurement policies that provide free/reduced price healthful choices or target competitive foods/beverages. Efficacy of quality standards for school meals appears heterogeneous with less consistent benefits. These findings inform policy priorities for improving diets in children.