Harvest for Healthy Kids Pilot Study: Associations between Exposure to a Farm-to-Preschool Intervention and Willingness to Try and Liking of Target Fruits and Vegetables among Low-Income Children in Head Start

Background Most US children do not meet recommendations for daily fruit and vegetable intake. Objective Our aim was to evaluate the hypothesis that at post-intervention, children exposed to the Harvest for Healthy Kids pilot study will have greater willingness to try and liking of target foods vs children in the comparison group, controlling for baseline differences. Design We conducted a quasi-experimental pilot study with comparison, low-intervention, and high-intervention groups. Pre- and post-intervention survey data were collected. Participants/setting The intervention period was October 2012 to May 2013. The analysis sample was 226 children within the higher-level unit sample of five participating Head Start centers (Portland, OR); 231 children dropped out of or enrolled in Head Start mid-year, were absent during or refused to participate in the assessments, or were missing covariates. Intervention The comparison group received no intervention components; the low-intervention group received foodservice modifications; the high-intervention group received foodservice modifications and nutrition education. Main outcome measures Willingness to try and liking of target foods were tested and analyzed as binary variables. Statistical analyses performed McNemar's tests were used to assess differences between pre- and post-intervention scores by intervention group. Fixed slope, random intercept multilevel logistic models were used to assess associations between intervention group and post-intervention scores controlling for covariates, adjusting for baseline values, and accounting for center level clustering. Results The difference between pre- and post-intervention willingness to try and liking of target foods was statistically significant for a variety of foods; for example, 44.2% of children liked rutabaga pre-intervention compared with 78.1% post-intervention (P=0.004). Multilevel modeling indicated similar associations. Conclusions The Harvest for Healthy Kids pilot study suggests a positive association between the intervention and willingness to try and liking for target foods among study participants. Additional research is needed to assess the impact of the program on fruit and vegetable intake.