Assessing eating context and fruit and vegetable consumption in children: new methods using food diaries in the UK National Diet and Nutrition Survey Rolling Programme

Background: Eating context is the immediate environment of each eating occasion (EO). There is limited knowledge on the effects of the eating context on food consumption in children, due to the difficulty in measuring the multiple eating contexts children experience throughout the day. This study applied ecological momentary assessment using food diaries to explore the relationships between eating context and fruit and vegetable consumption in UK children. Methods: Using 4 d unweighed food diaries, data were collected for 642 children aged 1.5-10y in two years of the UK National Diet and Nutrition Survey (2008-2010). Participants recorded all foods and drinks consumed at each EO, where and with whom the food was consumed, whether the TV was on and if eaten at a table. Mixed logistic regression and mixed multinomial logistic regression were used to calculate associations between eating contexts and fruit and vegetables (FV) consumed by quartiles. Results: Of 16,840 EOs, 73% took place at home and 31% with parents only. Frequency of eating alone and with friends increased with age. Compared to eating at home, children aged 1.5-3y were more likely to consume fruit at care outside home (>10-50g OR: 2.39; >50-100g OR: 2.12); children aged 4-6y were more likely to consume fruit (>50-100g OR: 3.53; >100g OR: 1.88) and vegetables at school (>30-60g OR: 3.56). Compared to eating with parents only, children aged 1.5-3y were more likely to consume fruit with friends (>10-50g OR: 2.69; >50-100g OR: 3.49), and with carer and other children/others (>10-50g OR: 2.25); children aged 4-6y were more likely to consume fruit (>50-100g OR: 1.96) and vegetables with friends (>30-60g OR: 3.56). Children of all ages were more likely to eat vegetables when the TV was off than on and at a table than not at table. Conclusions: The use of food diaries to capture multiple eating contexts and detailed fruit and vegetable consumption data was demonstrated at a population level. Higher odds of FV consumption were seen from structured settings such as school and care outside home than at home, as well as when eating at a table and the TV off. This study highlights eating contexts where provision of fruit and vegetables could be improved, especially at home. Future research should take eating context into consideration when planning interventions to target children's food consumption and eating behaviour.