A cluster randomized factorial trial of school-lunch salad bars and marketing on elementary students’ objectively measured fruit and vegetable consumption

Background: Schools in the United States play a crucial role in promoting healthy eating habits. Despite numerous public health efforts, children’s consumption of fruits and vegetables (FVs) still fall short of recommended amounts. Advocates have promoted school lunch salad bars as an effective strategy to increase primary students’ FV consumption, but rigorous research has lagged behind their efforts. This study evaluated the effectiveness of introducing school lunch salad bars, FV marketing, and the combination of both on elementary students’ objectively measured fresh FV selection and consumption. Methods: A cluster-randomized factorial trial was conducted among 13 elementary schools from 12 public school districts participating in the U.S. National School Lunch Program with more than 50% of students eligible for free or reduced lunch. Schools were assigned randomly to one of four conditions: Salad Bar with FV Marketing (n = 4), Salad Bar-only (n = 3), FV Marketing-only (n = 3), and wait-listed control (n = 3). The conditions were assessed at three measurement waves. Students’ (N = 3,080) aggregated fresh FV selection and consumption were measured via digital scales (grams) using objective plate waste methodology. Zero-inflated negative binomial models were used to examine differences in consumption (accounting for excess zeros and overdispersion) by condition and wave. Results: No significant differences were observed for students selecting FVs across any condition. In contrast, after ten weeks of exposure, stand-alone school lunch salad bars significantly increased students’ consumption of FV (IRR = 1.84, 95% CI 1.12, 3.04) compared to the wait-listed control. Findings for FV marketing alone suggested increased consumption (IRR = 1.60, 95% CI 0.97, 2.64) relative to control, but were non-significant and inconclusive. Salad bars in combination with FV marketing showed the strongest effect on FV consumption relative to the wait-list control, with a significant increase observed at ten weeks (IRR = 2.07, 95% CI 1.29, 3.31). Conclusions: Stand-alone salad bars are effective at increasing elementary students’ FV consumption after a minimum of ten weeks. The combined intervention of salad bars and FV marketing demonstrated the greatest improvement in FV consumption. These findings support the promotion of salad bars in schools as a strategy to increase FV consumption among students. Trial registration: ClinicalTrials.gov Identifier: NCT03283033 (preregistered on: 9/14/2017).