Abstract
This randomized controlled trial examines the effects of a school garden intervention on children's fruit and vegetable (FV) consumption at school over two years. We randomly assigned schools to the intervention group that received gardens and related curriculum (n = 24) or to the waitlist control group that received gardens and curriculum at the conclusion of the study (n = 22). Children in second, fourth, and fifth grade at baseline (n = 2767) in low-income schools (n = 46) in four U.S. States (Arkansas, Iowa, New York, and Washington) participated. The intervention comprised gardens for each classroom; a curriculum focused on nutrition, plant science, and horticulture, including activities and FV tasting sessions; resources for the school that addressed topics such as soil contamination and food safety; an implementation guide focused on issues related to planning, planting, and maintaining the garden through the year, engaging volunteers, summer gardening, building community capacity, and sustaining the gardening program. FV consumption was measured by photographing lunches before and after children ate, for 2-3 days, at baseline and at each of 3 subsequent periods of data collection during the intervention. FV consumption was calculated using Digital Food Image Analysis. Among children in the intervention, fruit consumption and low-fat vegetable consumption increased from pre-garden baseline to post-garden more than among control group children. Garden intervention fidelity (GIF) also predicted changes in dietary intake, with more robust interventions showing a stronger effect than weaker interventions. GIF-lessons was a particularly potent predictor of change in dietary intake. School gardens modestly increase children's FV consumption at school.