Effects of the DAGIS randomized controlled trial on home environment and children's food consumption according to the degree of implementation.

ABSTRACT

BACKGROUND: Combining process evaluation data with effectiveness data and examining the possible mediators of intervention effects elicits valuable knowledge about how and for whom these interventions are effective. The aim of this study was to examine whether the parental degree of implementation (DOI) of a home-involving preschool intervention affected children's food consumption via home mediators. METHODS: The five-month Increased Health and Wellbeing in Preschools (DAGIS) intervention involved 476 participating children aged 3-6 years and was conducted in 2017-2018. Parents reported children's food consumption (g/day) outside childcare hours, the availability of foods at home, role modelling of food consumption, and the norms related to food consumption. In addition, parents reported the extent to which they had implemented the intervention program at home. Mediation analyses were conducted to examine the effect of low and high DOI compared to control group on the change in children's consumption of fruit and vegetables (FV), sugary everyday foods, sugary treats, and sugar-sweetened beverages (SSB) via food availability in the home, parental role modelling and parental norms. RESULTS: Compared to the control group, there was a direct effect of a high DOI on diminishing consumption of SSB (B -27.71, 95% CI -49.05, -4.80). No indirect effects were detected. In the high DOI group, a change in parental norm was associated with increased FV consumption showing an indirect effect (B 4.31, 95% CI 0.23, 10.59). In the low DOI group, there was an indirect effect via decreased food availability leading to decreased sugary everyday food consumption (B -2.17, 95% CI -5.09,  -0.09). CONCLUSIONS: Combining process evaluation and effectiveness data revealed a decrease in children's SSB consumption only in the high DOI group, as well as indirect effects on children's consumption of FV and sugary everyday foods. In order to gain more intervention effects, further studies are required in order to examine parental facilitators and barriers to the implementation of interventions and how to impact effectively the determinants of the targeted behavior. TRIAL REGISTRATION ISRCTN57165350 (8 January, 2015).