As an integral part of the Five A Day campaign, the School Fruit and Vegetable Scheme (SFVS) is currently the largest national initiative to promote fruit and vegetables to children in England. Introduced into primary schools between 2002 and 2004, the scheme makes available one piece of fruit or a vegetable to children each school day for the first three years of school. The UK is not alone in introducing initiatives to promote children’s intake of fruit and vegetables(Reference Parmer, Salisbury-Glennon and Shannon1–Reference Tak, Velde and Brug3).
Several evaluations of the SFVS have shown an increased intake of fruit rather than vegetables while children participate in the scheme, but children’s intake falls when they are no longer eligible(Reference Ransley, Greenwood and Cade4–Reference Fogarty, Antoniak and Venn6). In schools without the SFVS children’s intake of fruit and vegetables falls as they progress from Reception (age 4–5 years) through to Year 2 (age 6–7 years)(Reference Schagen, Ransley and Blenkinsop7). To maintain and improve existing intakes of fruit and vegetables from Reception to Year 2 and beyond, it seems important for schools to extend initiatives to promote fruit and vegetables over and above the provision of free school fruit in Key Stage One (4–8 years).
Many English primary schools have embraced this idea and found opportunities for children to learn more about fruit and vegetables through lessons in the formal curriculum and extracurricular activities. For example, the National Curriculum enables children to learn about fruit and vegetables in Science, Design and Technology, and Personal, Social, Health Education and Citizenship. Geography, English and Art also provide some educational opportunities for children to learn about fruit and vegetables(8).
Outside the formal curriculum children can learn about fruit and vegetables through growing and cooking activities. The Royal Horticultural Society, for example, has spearheaded a national campaign called ‘Grow It, Cook It, Eat It’. This campaign encourages schools to set up growing activities in school which lead to cooking and eating opportunities for participating children(9). The School Food Trust is also running a £20 million ‘Lets Get Cooking’ campaign to help children learn relevant cooking and food preparation skills(10).
Research has shown that practical activities such as cooking and gardening facilitate behaviour change in children(Reference Thomas, Sutcliffe and Harden11, 12). Practical activities undertaken with peers and staff in school may help young children to overcome some of their natural fear of new food, known as food neophobia(Reference Birch13). This may occur through modelling of appropriate eating behaviour, repeated exposure to foods, providing encouraging and supportive environments for eating, and practical activities which help children become more familiar with foods(Reference Birch and Marlin14–Reference Wardle, Cook and Gibson16).
New school food standards have been introduced to improve the nutritional quality of food served at school. Provision has been made to increase the amount of fruit and vegetables in school lunches and place restrictions on the provision of foods with low nutritional value, such as chips, confectionery and soft drinks(10, 17). These standards are compulsory; however, children are still at liberty to bring a packed lunch which does not conform to the new standards. A recent intervention to improve the food and nutritional value of children’s lunch boxes found that only 19 % of children met the food-based guidelines for vegetables and 54 % for fruit(Reference Evans, Greenwood and Thomas18). The content and nutritional value of what children eat outside school is the responsibility of parents and other adult carers. There is some evidence that when children eat more fruit at school they eat less at home(Reference Schagen, Ransley and Blenkinsop7).
The National Healthy Schools Programme also addresses the promotion of fruit and vegetables as part of a healthy diet. This voluntary scheme sets targets for schools to achieve in four areas including Healthy Eating and leads to National Healthy School Status(19).
Schools are at liberty, in consultation with their governing bodies, to write and implement a policy on food in their school, which many have done. Some schools include parents in their initiatives to improve school food through correspondence with them and by involving them in activities such as cooking and growing. These arrangements for educating children about fruit and vegetables and their value in a healthy diet vary across English schools. Apart from the impact of the SFVS on the diet of young children, little is known about whether these initiatives have an effect on children’s intake of fruit and vegetables and the nutritional composition of their diets.
Therefore the aim of the present research was to explore whether children’s intake of fruit and vegetables is related to school initiatives to promote fruit and vegetables.
Methods
Sampling methods
The sample was drawn from maintained schools containing pupils in Years 2 to 4 with a minimum year group size of fifteen pupils. Independent schools, special schools, schools without all three years, and small schools with less than fifteen pupils per year group were excluded. The National Foundation for Educational Research (NFER) was responsible for recruiting schools and collecting data. Schools that had or were participating in other NFER projects were excluded.
A random national sample was stratified by ethnicity, deprivation, educational achievement, and region of England. Power calculations suggested that a sample of 2200 children would give approximately 90 % power to detect a difference of 0·33 portions of fruit per day based on a comparison of mean fruit intake in schools with a high proportion of children eligible for free school meals v. those with a low eligible proportion. Results from our initial evaluation of the SFVS found a 68 % response from pupils completing the CADET food diary(Reference Ransley, Greenwood and Cade4). To allow for this loss to follow-up, 130 schools would be recruited with an estimated total of 3250 children available.
One hundred and twenty-nine schools accepted to take part in the study. A letter was sent to parents or guardians of children in Year 2, two weeks in advance of the data collection, giving information about the study and providing the opportunity for children to be withdrawn from the study. Ethical approval was granted by the University of Leeds Research Ethics Committee.
Dietary assessment
The Child and Diet Evaluation Tool (CADET) was used to estimate the children’s mean intake of foods and nutrients. CADET was designed as a simple dietary assessment tool and records a child’s dietary intake over 24 h. The validation study compared CADET with a 24 h semi-weighed food diary obtained from the same children for the same day and showed close association with the usual diet(Reference Cade, Frear and Greenwood20). CADET was completed by NFER trained administrators during the school day, and sent home to be completed by parents and returned the following morning with the child.
Children with a total energy intake of less than 2092 kJ/d (500 kcal/d) or more than 14 644 kJ/d (3500 kcal/d) were excluded from the study, as were those for whom the parental part of the CADET was left blank. This resulted in a final sample size of 2530 children.
Initiatives to promote fruit and vegetables
A questionnaire was developed by researchers at the University of Leeds to measure the initiatives schools use to promote fruit and vegetables to pupils in Year 2 and across the school. The questionnaire investigated what is taught about fruit and vegetables in the formal curriculum; the amount of time spent learning about fruit and vegetables; school activities and resources for growing and cooking food; school catering; and the involvement of parents in promoting fruit and vegetables to children. The questionnaire was administered to all Year 2 teachers to complete.
A scoring system was developed to rate the extent to which schools engaged in the activities outlined above. A maximum score of 7 was awarded for each of five sections depending on the extent to which activities were undertaken. A maximum score of 35 could be awarded. The median of the scores was considered to be the cut-off point for schools falling into ‘high’ and ‘low’ scores.
Statistical analysis
A multivariable regression analysis using multilevel modelling techniques was conducted using the software MLwiN v2·10 (Centre for Multilevel Modelling, University of Bristol, Bristol, UK) to investigate the effect of initiatives to promote fruit and vegetable consumption on children’s intake of these foods(Reference Rasbash, Steele and Browne21). A two-level random intercepts model was used to allow for the dependency inherent in pupil observations nested within the same school to be taken into account.
Analyses were adjusted for ethnicity and deprivation. The interaction between ethnicity and deprivation was assessed by the likelihood ratio test and included in the model for foods where this was statistically significant (P < 0·05).
Results
Basic characteristics
We recruited 2709 children from 129 schools, a response rate of 72 % to CADET. After 179 exclusions for misreporting on CADET, a final sample size of 2530 children was achieved. The mean age of the children was 7 years (1290 girls and 1240 boys). English was spoken as an additional language by 10 % of the sample. Seventeen per cent of children received free school meals and 54 % ate a packed lunch. Thirty-five per cent of children had a member of the family educated to degree level or higher. Of the 130 participating schools, 100 returned the school questionnaire.
Table 1 shows the mean intake of foods and nutrients according to gender. The amount of vegetables eaten by boys exceeded that eaten by girls by 14 g/d; however, girls ate 38 g more fruit daily than boys. The combined daily intake of fruit and vegetables for all children was 309 g, equivalent to almost four 80 g portions daily but less than the Five A Day recommendation. Boys and girls consumed similar amounts of pulses, beans and seeds, about 20 g daily; while boys on average ate 8 g more dried fruit daily than girls. Children ate almost the same weight of chocolate, confectionery (sweets, toffees, mints, etc.) and savoury snacks each day (77 g) as they did vegetables (90 g).
Table 1 Mean intake of foods and nutrients in Year 2 children (aged 6–7 years, n 2530) from 129 English primary schools
MLM, multilevel model; RE, retinol equivalents.
*Log transformed and back transformed to estimate, treated as ratio instead.
Milk consumption was low for this age group, only 233 g/d (just over a quarter of a pint), as other drinks such as fruit juice, carbonated drinks and squashes featured highly in the diet. In both boys and girls the consumption of carbonated drinks and squash exceeded that of milk; however, these children were obtaining enough calcium from their diet to meet the Reference Nutrient Intake (RNI) of 550 mg/d.
Reported daily energy intake for boys and girls was 1255 kJ (300 kcal) and 753 kJ (180 kcal), respectively, below the Estimated Average Requirement for this age group. Vitamin A intake was about half the RNI of 500 μg/d. Vitamin C intake was more than twice the RNI for this age group and intake of folate was also well above the RNI for this age group of 150 μg/d. Iron intake was adequate and protein intake (55 g) was almost twice the RNI of 28 g/d. Percentage energy derived from fat was low and consequently the percentage of energy derived from carbohydrate was slightly higher than guidelines recommend. There are no dietary guidelines for fibre intake in children; however, an intake of 12 g/d appears low. Sodium intake was high at double the recommended intake for children of this age.
Table 2 explores differences in food intake and initiatives to promote cooking, gardening and improve catering at school. There were no significant differences in children’s intake of foods according to whether schools had a high or low score for cooking activities, although there were higher intakes of fruit (excluding dried fruit) in schools which had a high score.
Table 2 The effect of cooking, gardening and school catering on food intake in Year 2 children (aged 6–7 years, n 2530) from 129 English primary schools
MLM, multilevel model.
In schools that achieved a high score for gardening, children ate significantly more vegetables, but there were no other significant differences in children’s food intakes for a high score compared with a low score.
Where schools achieved a high score for improving catering, intake of pulses, beans and seeds was significantly higher, 24 (95 % CI 20, 29) g/d, compared with low-scoring schools, 17 (95 % CI 12, 22) g/d. A borderline non-significant but lower intake of sweets, toffees and mints was found in high-scoring schools compared with low-scoring schools.
Table 3 shows further effects associated with school initiatives to promote fruit and vegetables to children. Schools with a high score for lessons teaching children about fruit and vegetables were not associated with children’s food intake apart from a slightly reduced intake of savoury snacks. This was borderline non-significant.
Table 3 The effect of lessons, parents and combined initiatives to promote fruit and vegetables in school on food intake in Year 2 children (aged 6–7 years, n 2530) from 129 English primary schools
MLM, multilevel model.
In schools where parents had been informed about its guidance on food and involved in meetings to promote fruit and vegetables (high scoring schools), children ate significantly more vegetables (not pulses, beans or seeds) compared with schools with a low score.
A high total score for promoting fruit and vegetables was associated with a significant increase, with children eating 25 g more vegetables daily. Children in these high-scoring schools also ate more pulses, nuts and seeds and less chocolate products, although the differences were borderline non-significant.
Table 4 confirms the independent effect associated with total score on children’s intake of food by taking into account the effect of ethnicity and social deprivation. These results also show that children from schools with a high total score ate 25 g more vegetables daily than children from schools with a low score. These children also ate significantly more pulses, beans and seeds.
Table 4 IndependentFootnote * effect of initiatives to promote fruit and vegetables in schools, controlling for social class and ethnicity, in Year 2 children (aged 6–7 years, n 2530) from 129 English primary schools
* Linear regression of the total score on the foods listed was adjusted for ethnicity and deprivation. The interaction between ethnicity and deprivation was assessed by the likelihood ratio test and included in the model for foods where this was statistically significant (P < 0·05).
Discussion
The present results provide an overview of food and nutrient intake of a large sample of English children at the end of their third year of school (School Year 2).
Intake of fruit and vegetables in this group was almost four portions per day, which appears to be an improvement on the findings of earlier studies; however it is lower than the Five A Day recommended for current and future health. This finding is similar to the baseline intakes of children in an earlier evaluation of the SFVS(Reference Ransley, Greenwood and Cade4).
It is evident that foods other than fruit and vegetables have a prominent position in the diet of children. Sweets, confectionery and savoury snacks were eaten in almost the same amounts as vegetables, and more fizzy drinks and squash were consumed than milk.
From a nutritional point of view, however, calcium levels met recommended intakes and iron levels were adequate. Sodium levels were high as has been shown in many dietary surveys of children(Reference Ransley, Greenwood and Cade4, Reference Gregory, Lowe and Bates22). Large regular intakes of savoury snacks contribute to these high intakes of sodium. Vitamin A levels were low and may result from a poor intake of vegetables in some children. It is interesting to note that intake of folate was adequate and intake of vitamin C was high. Good sources of these vitamins in children’s diets are likely to be fortified breakfast cereals and fruit juice, respectively.
With regard to macronutrient intake, energy intake was low. This may be due to under-reporting foods consumed as a result of items being missed or assumed portion sizes which are too small for this age of child; however protein intake was more than adequate. Fibre intakes appeared low and the figures obtained are in line with a diet which is low in fruit and vegetables. The fibre intake of children in the current survey is on a par with that of the adult population. However, it should be noted there are currently no absolute recommendations for intake of fibre for this age group of children.
Schools across England vary in the number and type of initiatives they undertake to educate about and promote fruit and vegetables to children. The present study has provided some evidence to show that, in schools where gardening activities take place, children consume significantly more vegetables and pulses than in schools where gardening and growing activities are limited. This may provide some evidence to support the importance of practical activities in encouraging children to consume vegetables, as has been shown elsewhere(Reference Morris and Zidenberg-Cherr23); however, further work is required to confirm this.
Likewise, in schools where there was a high degree of parent involvement in promoting fruit and vegetables to children, more vegetables were eaten. Because of the nature of the current cross-sectional analysis it is not possible to deduce a causal relationship, but these results suggest there may be some association that needs to be tested further. Parents are vitally important to the acceptance by children of new fruit and vegetables in their diet. This is because of the importance of adults modelling appropriate eating behaviour and creating a positive environment to support and encourage children’s intake of these foods(Reference Birch13, Reference Johnson and Birch24–Reference Birch and Fisher26).
The number of lessons spent on promoting fruit and vegetables was not associated with consumption of fruit and vegetables. Perhaps lessons do not include the best behaviour change techniques such as modelling, repeated exposure and practical experience with fruit and vegetables(17, Reference Contendo, Contendo and Bronner27). Lessons may not include such an approach and may explain why efforts in this area are not associated with a higher intake of fruit and vegetables. It is therefore important to augment knowledge about fruit and vegetables with other approaches to encourage consumption of these foods.
Combining the five individual scores to produce a global score to reflect initiatives schools made to educate children about fruit and vegetables produced one notable finding regarding higher intakes of vegetables in schools with a high score. One might ask why this did not hold true for fruit. Perhaps because all schools now participate in the SFVS, which largely supplies fruit to children, intake of fruit cannot be improved upon. It has reached its upper threshold, leaving more scope for increasing vegetable intake. Certainly, the children were eating on average about 88 g more fruit than vegetables daily, equivalent to a portion of fruit.
Do the results differ in more deprived schools? The results reported in Table 4 provide some encouragement that efforts to promote fruit and vegetables to children have an effect regardless of the deprivation status of the area and the ethnic mix of the school.
The present study was a large, national, cross-sectional study of children’s diet; however there are limitations. Cross-sectional studies can only suggest associations between variables and do not provide robust evidence of causality. Little work has been undertaken to evaluate the impact of educational interventions on children’s intake of fruit and vegetables. Measuring exactly how and what is being taught in different parts of the formal curriculum relies on teacher recall and is therefore subject to error. Promotion of fruit and vegetables in the informal curriculum through activities such as cooking and gardening is also limited by reliance on teachers to record this involvement. However, we believe the questionnaire used to do this provided a reasonable record of the activities schools engaged in to promote fruit and vegetables to pupils.
CADET has been used in several large studies to estimate children’s intake of food and nutrients. It has the limitations of a 24 h record of food intake; however the sample size for the study is large and should compensate for this.
This is, we believe, the first time an attempt has been made to explore the relationship between initiatives schools themselves are taking to promote fruit and vegetables to children and their association with diet. Our findings show some encouraging results for schools that involve parents and promote fruit and vegetables through extracurricular activities such as gardening, but further works is needed to confirm these findings.
Acknowledgements
This work was supported by the National Prevention Research Initiative (NPRI), Medical Research Council Grant number G050129. The authors state that there are no conflicts of interest. The contribution of each author to the research was as follows: J.K.R. was the project manager and drafted the paper; E.F.T. conducted the statistical analysis; Y.R. designed the scoring system for school-based initiatives to promote fruit and vegetables, and undertook preliminary analysis; M.S.K. was the research assistant, liaised with schools and performed data collection; D.C.G. was the senior statistician and adviser; J.E.C. was the principal investigator for the project.
