Can a school gardening intervention improve children’s fruit and vegetable intake? An evaluation of two clustered randomised controlled trials.

Background This thesis describes the findings from the first clustered randomised controlled trials (RCT) designed to evaluate if a school gardening programme, the Royal Horticultural Society (RHS) Campaign for School Gardening, has an effect on children’s fruit and vegetable intake. Methodology Primary schools from 8 London boroughs were invited to take part in one of two related RCTs. Trial 1 consisted of 23 schools; randomised into either the RHS-led or the Teacher-led intervention. Trial 2 consisted of 31 schools; randomised into either the Teacher-led or a comparison group. A 24-hour food diary (CADET) collected baseline and follow-up dietary intake. Questionnaires were designed to measure children’s knowledge and attitudes towards fruit and vegetables and assess intervention implementation. Results In Trial 1, 1256 children were randomised to receive either the RHS-led (n=529) or Teacher-led (n=727) intervention. Of these, 356 children from the RHS-led and 329 from the Teacher-led arm provided data for the primary analysis. In Trial 2, 1475 children were randomised to receive either the Teacher-led (n=756) or comparison (n=719) intervention. Of these, 488 children from the Teacher-led and 428 from the comparison arm provided data for the primary analysis. Baseline analysis of children’s fruit and vegetable intake showed that eating a family meal together, cutting up fruit and vegetables and parental modelling of fruit and vegetable intakes were all associated with higher intakes of fruit and vegetables in children. Results from the RCTs found that in Trial 1, for combined fruit and vegetable intake, the Teacher-led group had a higher mean change of 8 g (95%CI: -19, 36) compared to the RHS-led group -32 g (95%CI: -60, -3). However, after adjusting for possible confounders this difference was not significant (intervention effect: -40 g, 95%CI: -88, 1; p=0.06). In Trial 2, the Teacher-led group consumed on average 15 g (95%CI: -36, 148) more fruit and vegetables than the comparison group; this difference was not significant. No change was found in children’s knowledge and attitudes. The process measures revealed that if schools improved their RHS gardening score by 3 levels, children had on average an increase of 81 g of fruit and vegetables (95%CI: 0, 163; p=0.05) compared to schools that had no change in gardening score, after adjusting for confounders. Conclusion The primary analysis of these two trials has found very little evidence to support claims that school gardening alone can improve children’s fruit and vegetable intake. Only when a gardening intervention is implemented at a high level within the schools can it improve children’s fruit and vegetable intake.