... Introduction. Joia de Sa, Karen Lock. Public health drivers of EU agricultural policy for fruit and vegetables

Transcription

1 European Journal of Public Health, Vol. 18, No. 6, ß The Author Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. doi: /eurpub/ckn061 Advance Access published on 21 August Will European agricultural policy for school fruit and vegetables improve public health? A review of school fruit and vegetable programmes Joia de Sa, Karen Lock Introduction Background: For the first time, public health, particularly obesity, is being seen as a driver of EU agricultural policy. In 2007, European Ministers of Agriculture were asked to back new proposals for school fruit and vegetable programmes as part of agricultural reforms. In 2008, the European Commission conducted an impact assessment to assess the potential impact of this new proposal on health, agricultural markets, social equality and regional cohesion. Methods: A systematic review of the effectiveness of interventions to promote fruit and/or vegetable consumption in children in schools, to inform the EC policy development process. Results: School schemes are effective at increasing both intake and knowledge. Of the 30 studies included, 70% increased fruits and vegetables (FV) intake, with none decreasing intake. Twenty-three studies had follow-up periods >1 year and provide some evidence that FV schemes can have long-term impacts on consumption. Only one study led to both increased fruit and vegetable intake and reduction in weight. One study showed that school fruit and vegetable schemes can also help to reduce inequalities in diet. Effective school programmes have used a range of approaches and been organized in ways which vary nationally depending on differences in food supply chain and education systems. Conclusions: EU agriculture policy for school fruits and vegetables schemes should be an effective approach with both public health and agricultural benefits. Aiming to increase FV intake amongst a new generation of consumers, it will support a range of EU policies including obesity and health inequalities. Keywords: children, fruit, obesity, public health, schools, vegetables... Public health drivers of EU agricultural policy for fruit and vegetables Except for food safety issues, public health is not traditionally a consideration of the EU Common Agricultural Policy (CAP) despite its clear role in nutrition. On 17 January 2007, the European Commission (EC) presented proposals to reform agricultural policy for fruits and vegetables (FV) 1 in addition to a larger legislative process to modernize CAP in Uniquely for agricultural policy, health has been seen as a key driver of the FV proposals with stated goals which include increasing FV consumption in Europe particularly as an approach to tackle childhood obesity. The EC announced a 60% budget contribution towards promotion of FV consumption as long as this was targeted at children and adolescents. In addition, in April 2007, the EC announced new funding for school FV schemes, which had not been proposed in the initial reforms. Impetus was provided by the EC White Paper on Obesity (2007) stating that a school fruit scheme would be a step in the right direction. 3 In June 2007, European Ministers of Agriculture were asked by Agricultural Commissioner Fischer Boel to back a new E103 million annual FV school programme requiring co-funding by Member States. These plans were rejected but the Commission was asked to conduct an impact assessment of the school FV European Centre for Health of Societies in Transition, London School of Hygiene and Tropical Medicine, London, UK Correspondence: Karen Lock, London School of Hygiene and Tropical Medicine, Keppel Street, London WC1E 7HT, UK, tel: , fax: , Karen.Lock@lshtm.ac.uk proposals. 4 Between September 2007 and May 2008, DG AGRI conducted a mandatory impact assessment process 5 which informed new policy to be launched in autumn This aimed to assess the potential impact of a new school scheme on health, diet, agricultural markets, social equality and regional cohesion. It also assessed the value for money and added value of the proposals at an EU level. This literature review was conducted to directly inform this impact assessment. Its aim was to consider whether school FV schemes are an effective mechanism of improving FV consumption in children. Will improving FV intake in children improve public health? It is well accepted that low-fv intake in adulthood is associated with ill-health, particularly cardiovascular disease and cancer. 6,7 Inadequate FV consumption has been estimated to be responsible for over 1 million deaths annually in the EU. 8 There is limited evidence of an association between childhood FV consumption and risk of adult stroke and cancer from the Boyd-Orr Cohort, which some experts interpret as a direct effect. 9,10 There is also evidence of other benefits. Although FV consumption has been found to decrease in adulthood, 11 observational tracking studies show that eating recommended dietary guideline amounts in childhood positively influences healthy eating in adults. 12,13 The reasons for this constancy in the diet are not yet known. Diets high in FV have also been proposed as reducing obesity. 14 Despite health benefits, children worldwide are not meeting minimum FV intake goals of 400 g person 1 day 1 advocated internationally. 15 In a dietary survey of schoolchildren in nine European countries, none met FV intake guidelines. 16 The 2007 World Cancer Research Fund report recently proposed the average population consumption should be over 600 g day 1. 6

2 EU school fruit and vegetable review 559 Even with a range of initiatives targeting FV intake and obesity throughout Europe, there has been little improvement. Schools appear to be an ideal setting to improve diet and tackle obesity in children, but what factors make school FV interventions effective? METHODS The aim of this review was to systematically synthesize worldwide evidence from published and unpublished literature on interventions to promote fruit and/or vegetable consumption in children in school settings. The objective was to consider all interventions based in school or pre-school, applied to individuals or populations, encouraging fruit and/or vegetable consumption. The primary outcome was either change in intake of fruits and/or vegetables, or a change in knowledge, attitude or preference to fruits or vegetables. Study inclusion was conditional on the presence of a control group for comparison and a follow-up period of at least 3 months. The intervention had to promote a diet high in fruit and/or vegetables in school children under 18 years. Secondary outcomes including impact on other food intake and body mass index (BMI) were considered. In August 2007, the following databases were searched from the earliest record Pubmed, CABDirect, Cochrane Library, Web of Knowledge, IBSS, Psycinfo(BIDS), Embase and Biomed Central. The search strategy was developed for Pubmed and adapted for other databases. This was complemented by a search for unpublished literature, through the internet and contact with organizations in countries known to have 5 a day type FV programmes. References cited in articles were searched, and 34 experts worldwide were contacted to provide information on published and unpublished studies. Full details of the search strategy, including key words used, for each database are available in the complete report. 17 Papers with an English abstract published in any language were considered. Abstracts were screened by one reviewer and rejected if it was clear that the article did not report a school FV intervention or if inclusion criteria were not met. If there was uncertainty, the full text was obtained and studies read by two reviewers who agreed on final inclusion. All studies meeting inclusion criteria were independently read by both reviewers. Study quality was assessed using a quality assessment tool used in a previous published systematic review of FV interventions. 18 Meta-analysis was not attempted due to marked heterogeneity of populations, interventions and outcome measures among the included studies. We followed the Cochrane handbook guidance supporting the use of a systematic, narrative approach when meta-analysis is inappropriate. We thus synthesized the study results according to age group of the target population and intervention type. RESULTS From the search strategy 1021 unduplicated papers were identified of which 128 reported on FV interventions in school-based settings. Of these 87 studies were excluded because they did not meet inclusion criteria and 7 whose quality was rated as weak. The final pool consisted of 34 articles reporting on 30 studies. The general study characteristics are summarized in Table 1. The studies were divided by age group into two groups: 5- to 11-year-olds and 11- to 18-year-olds to approximate with primary and secondary school age groups worldwide. Three studies did not fit these categories and were assigned based on the age of the majority of children Details on the studies Table 1 Summary of all the studies included in the review Age group of children Young (5 11) Older (11 18) Total number of studies 23 7 Countries Ireland (2) Belgium (1) Netherlands (3) Norway (2) Norway (1) USA (4) New Zealand (1) Scotland (1) Spain (1) The United Kingdom (6) United States of America (10) Study design 12 6 Non-randomized controlled trial 11 1 Number of participants < > Intervention components FV provision (free or subsidized) 7 3 Classroom based (e.g. curriculum) 17 7 School, wide (e.g. inc FV 13 5 exposure) and policy Teacher involvement 7 3 (e.g. training) Peer leader involvement 2 1 School food service involvement 8 4 Parent involvement 11 3 School nutrition policy 2 3 Community involvement 5 0 Other (e.g. gardening) 0 1 Length of follow-up (months) > FV intake measurements Food frequency 10 6 Food record/diary h recall 11 3 Plate waste 3 0 Observation 7 0 Survey 1 3 Interview 1 0 Parent 4 2 Other outcomes measured (apart from FV intake) Knowledge/attitudes/preferences 6 1 Psychosocial variables 2 0 Determinants of FV intake 3 0 Micronutrient intake 0 2 Cholesterol 1 0 Anthropometry/BMI 5 2 Physical activity 5 2 Reduced TV viewing hours 2 1 Numbers of countries in which studies were in younger children >27 as one study was multi sited (Pro children 13 ). including intervention type and length of follow-up can be found in Tables 2 and 3. Evidence of effectiveness Twenty-two studies (over 70%) reported a significant positive intervention effect on FV intake at follow-up. Differences in intervention effect ranged from servings per day to +0.99

3 Table 2 Details of primary school studies Study Design Participants Data collection Intervention Results Bash Street kids intervention 35 Integrated Nutrition Project (INP), USA 52 Cluster randomized controlled trial Follow-up: 10 months Follow-up: 4 years 2 Intervention schools (511 children) 2 Control schools (464 children) Two school year groups children 6 7 and children in 3 Denver schools only reports on year 3 and 4 Age appropriate assessments Food diaries Interviews (1) Plate waste assessment (2) Food recall/record (3) Classroom survey on knowledge and attitudes to FV (4) 5 min interview with kindergarten kids about knowledge of FV Increased provision of FV in schools (tuck shops and school lunches) Tasting opportunities Pont-of-purchase marketing Newsletters for parents Curiculum materials (1) 24 weekly classes that included food preparation and eating. Taught by special resource teacher (2) Teacher training (3) Parent education (4) Community nutrition/food resource development control: no exposure Intervention children tasted more FVover time (P < 0.001)22.4/32 to 27 no of foods tasted Also tasted several FV that had not been tasted at baseline. Weight of fruit intake increased in both groups. Intervention (+50 g) P = Control (+7 g) Treatment students consumed significantly more FV than comparison students: 0.19 more F serving, 0.25 more V servings and 0.4 FV servings in total. Treatment children demonstrated higher levels of knowledge 560 European Journal of Public Health Gimme 5, USA 24 Follow-up: 3 years 1253 children in 4th and 5th grade from 16 elementary schools 7 day food record Process evaluation 12 sessions over 6 weeks including handouts, posters, worksheets, newsletters, videos, point of purchase education at shops. Lower decrease in intervention vs control group: net effect of +0.3 servings per day 5 a day power play! Campaign, USA 53 National school fruit scheme (NSFS), England 32 Follow-up: 1 school year Non-randomized controlled study of National school fruit scheme (NSFS) implemented in different regions of country over 2 years Follow-up: 3 years 49 schools 151 classrooms (4th and 5th grade) 2684 cases established 15 schools control T1 19 schools T2 15 schools Random sample of 113 schools in East Midlands (intervention) and 122 schools in Eastern region (control) Students: California Children s Food Survey 24 h recall self-reported food diary Fruit intake completed by parents for 3 consecutive years, before and after participation T1 power play! Activities conducted only in school. School Idea and Resource Kit T2 power play! Activities in schools, community youth organisations, farmers markets, supermarkets, mass media Control: no nutrition activities Intervention region: Free piece of school fruit every day for 4- to 6- year-old children ( ). In Western region NSFS implemented June Control (eastern region): no fruit as NSFS implemented later (September 2004) and study controls then too old to qualify for participation in NSFS Both intervention sites reported significant increases in self-reported FV intake compared with control site but not with each other. Increases highest for T2 (0.4 serving, from 2.9 to 3.3) compared with 0.2 serving (from 2.7 to 2.9 in T1). May 2004 proportion eating F every day in intervention was markedly higher +11% (95% CI +7.4 to 14.6) But in May 2005 proportion fell to less than the control region ( 2.8%) Nutrition education at primary school (NEAPS), Ireland 38 Follow-up: 3 months 821 children aged 8 10 years from 8 intervention and 3 control schools in urban and rural areas 453 intervention 368 control 5 day food diary also assessed knowledge and preferences 20 sessions over 10 weeks including worksheets, homework and exercise regime; parent involvement More intervention children consumed 4 or more FV per day intervention group demonstrated significant changes in reported behaviour and food preferences overall (P < 0.01) Downloaded from at Pennsylvania State University on February 27, 2014

4 Eat Well and Keep Moving, USA 54 Kids Choice school lunch program, USA 43 Non randomized controlled trial Follow-up: 7 months 6 intervention schools, 8 matched schools for control 470 students initially Student food and activity survey and 24 h recall and youth food frequency 346 children 1st, 2nd and 4th grades Observed FV intake Interviews with children Classroom based. Food school service and family involved All children given same FV at lunch (2 choices F and V) Intervention: half classrooms randomly assigned to receive token reinforcement for fruit or vegetable consumption if they ate at least 1/8 cup of assigned food group Control: no reward Increase in the consumption of FV (0.36 servings 4184 KJ 95% CI P = 0.01) = 0.73 servings per day Intake increased during Ix but not measured after Preferences increased for range FV 2 weeks after but returned to baseline at 7 months (greater fruits than veg) Food Dudes, UK 36 Follow-up: 4 months 2 inner city London Primary Schools 794 Children 5- to 11-year old. Observation Home using parental 24 h recall, plus subset of parents interviewed (paid 35) 16 day Ix: 6 6min episodes of video homepacks, rewards for eating FV at snack and lunch some maintenance ix Control: received free FV Significant higher increases in FV intake at snacktime, lunchtime and at home in intervention group Food Dudes, Ireland 40 Follow-up: 1 year 2 experimental schools, 1 control 435 children Observation, weighed measures 16 day intervention featuring video, rewards, letters from FD homepacks and help with maintenance period Control: free FV At 12 month follow-up children in experimental school were provided with and consumed significantly more lunchbox FV Netherlands 19 Follow-up: 3 months 30 7th grade classes 16 intervention 14 control Total of 675 children Internet-administered School based intervention Combination of internet tailored advice for children followed by internet-supported brief dietary counselling by the nurse in the presence of at least one parent Control: no internet advice FV intake did not differ significantly between intervention and control However knowledge was significantly different in treatment group Paradis et al., USA 22 Non- randomized controlled trial Follow-up: 8 years 5 a day power plus 55 Follow-up:10 months N = 458 in 1994 N = 420 in community elementary schools Children in 4th grade from 20 ethnically, culturally and economically diverse schools (10 matched pairs) N = 1750 initially 7 day food FFQ Anthropometric measurements Physical activity Health behaviour for all; self-completed 24 h food record for random sample; lunchroom observation Health education curriculum involving diet and physical activity (designed for diabetes prevention) delivered in grades 1 6 in community s 2 elementary schools. Community activities School nutrition policy (1) Behavioural curricula (2) Parental involvement (3) School food service changes (4) Industry support Some early positive effects on skinfold thickness but not BMI, physical activity, fitness or diet. Key high-fat and high-sugar foods consumption decreased Intervention students had a higher mean intake of FV than control. Difference was 0.4 servings per day at follow-up (continued) EU school fruit and vegetable review 561 Downloaded from at Pennsylvania State University on February 27, 2014

5 Table 2 Continued Study Design Participants Data collection Intervention Results 5 a day cafeteria power plus, USA 34 CATCH study, USA 23 Follow-up: 3 years 1668 students in 1st and 3rd grades form 26 elementary schools 5106 students initially of which subset of 1186 students were followed Observations by trained staff 24 h recalls at baseline and follow-up; 30 min face to face interviews also School food service involvement, daily activities and special FV events Modifications in school food service, physical education, classroom curricula and parental involvement Significant increase of FV intake (P = 0.02) verbal encouragement by lunch staff significantly associated with higher intakes. Difference +0.3 servings per day No difference at follow-up 562 European Journal of Public Health UK School Fruit and Vegetable Scheme 28 Infant and primary schools in N England 3703 children aged 4 6 years. CADET (child and diet evaluation tool) 1 portion of F or V provided per child on each school day between February and December 2004 Control: no fruit Increased FV intake across reception and year 1 of 0.5 portions (95% CI ) and 0.7 portions (CI ) at 3 months which fell to 0.2 at 7 months in reception and 0.2 in year 1 Impact on year 2 inc FV intake of 0.5 portions ( ) 3 months fell to 0.2 at 7 months. (no longer eligible for free FV) No long term impact on V intake High 5, USA 56 (matched pair design) 28 elementary schools pair-matched 1698 children (1) 24 h recall (2) Cafeteria observations (3) Parents food frequency 14 lesson curriculum delivered on 3 consecutive days each week. Components: classroom, parent, food service. Control: no intervention Intervention group had higher intakes of FV at 2 years servings per day (P < ) Differences in psychosocial variables APPLES: Active programme promoting lifestyles in schools, UK 57 Follow-up: 1 year 10 primary schools in Leeds 634 children aged 7 11 years 24 h recall 3 day food diary growth measures physical activity Teacher training, school meal changes, curriculum development, physical education, tuck shops Control: no intervention Intervention children had increased intake of vegetables by +0.3 servings per day but no change in F intake Schoolgruiten, Netherlands 30 Follow-up: 1 year 565 children of Dutch ethnicity 388 children of non-western ethnicity mean age 9.9 years at baseline Validated pro-children s Questions on intake and determinants Children and parents completed s (i) Availability and accessibility of FV at school Free FV twice a week at morning break (ii) Inc exposure to FV (iii) School curriculum changes Children of non-western ethnicity in intervention group reported significantly higher V intake (+20.7 g day 1 CI 7/6 33.7). Dutch children 0.23 F pieces per day (CI ) No significant effects based on parent reports Significant positive effects also found for perceived accessibility among children of non-western ethnicity. Downloaded from at Pennsylvania State University on February 27, 2014

6 EU school fruit and vegetable review 563 BMI significantly lower in intervention children (due to differences in relative weight) Fruit intake increased by 0.8 servings in intervention children (P < 0.01) No effect on V intake (i) Community activity co-ordinators (ii) Teacher resources, cooled water filters (iii) Science lessons, healthy eating resource, interactive card game during 2nd year Measurements of height, weight, waist circumference, blood pressure, physical activity. Diet by validated short food 730 children aged 5 12 years 4 intervention schools 3 control schools APPLE program, New Zealand 21 (FV only 1 year) Significant improvements in nutrition knowledge were seen in all children (P < 0.01) 1 control group, 3 intervention groups nutrition groups, physical activity group, combined nutrition and physical activity group Anthropometry Nutrition knowledge Physical activity Dietary assessment by parents 24 h recall, food frequency Children recruited from 3 primary schools in oxford, aged 5 7 years n = 213 Be Smart, UK 58 Follow-up: 14 months Overall FV intake increased significantly P < 0.01 and P < 0.05 Infants receiving free fruit statistically significant 50 g 1 day higher consumption (117g 1 dvs 67g 1 d excluding juices) Free piece of school fruit for CHILDREN aged 4 6 in NSFS pilot schools every day Control: no fruit 24 h ticklist And food frequency s 17 schools in low-income areas 8 NSFS 9 control Pilot National School Fruit Non randomized controlled trial scheme (NSFS) 59 Follow-up: 8 months Short-term and long-term increases in FV consumption and preferences Pro-children s Classroom curriculum n = 4192 students n = 2106 students Parental involvement Free FV during intervention Control: normal curriculum, FV dependent on country 62 schools in three European counties Cluster randomized controlled trial Pro children study Norway, Netherlands, Spain 39 servings per day. 17 Studies did not measure or report changes in intake in similar ways which makes comparisons of effect size or meta-analysis difficult. In addition, three studies reported significant effects on intake at some point during the study though this was not maintained at follow-up. Statistically significant results for changes in nutrition knowledge or preferences were reported in five out of seven studies. No studies found overall decreases in FV intake following interventions. Results by age group Twenty-three studies were aimed at younger children, while seven targeted older children. Ten studies on the younger children were carried out in the United States, 12 in Europe, and 1 in New Zealand. In the older age group, four were conducted in the United States and three in Europe. All studies involved both boys and girls. Participant numbers varied from 99 students to over 4000 students. Follow-up times varied from 3 months to 8 years. Nineteen studies in younger children reported a statistically significant increase in fruit and/or vegetable consumption at some stage during the intervention, which was maintained at follow-up in 16 studies (70%). Of the four studies that did not report increased intake, one reported significant increase in knowledge of the health benefits of FV, 19 two were primarily targeted at obesity prevention and reported a decrease in consumption of high-fat foods, 22,23 and one study prevented further decline in FV intake. 24 In the older age group, five studies (70%) reported statistically significant increases in fruit and/or vegetable consumption. One study reported increases in intake during the intervention which was not sustained at follow-up 25 and another, primarily targeted at obesity, only found decreased fat intake. 26,27 Results by intervention type The studies reported on a wide range of different interventions, although many had elements in common (Table 1). Provision of free or subsidized fruit and/or vegetables was part of 10 studies, including two national programmes in England 28 and Norway, 29 and a national pilot scheme in the Netherlands. 30 The English and Norwegian programmes have been evaluated after 3 years, showing statistically significant increases in consumption during the scheme, 28,31,32 with the Norwegian study showing sustained increases after the scheme finished. 31 This is unsurprising as accessibility and availability have been found to be important determinants of childhood FV intake. 33 Though FV were not provided directly as part of other studies, accessibility was increased through other interventions such as school food service modification, 34 tuckshops, 35 tasting or cooking 24,36 and school gardens. 20 A programme of supporting activities may improve the longer term effects of increasing availability on children s intake. Multicomponent interventions have been shown to be effective at both increasing FV intake and reducing obesity in children. 18,37 Twenty-three studies had a multicomponent design, which we defined as having two (or more) programme elements, for example, changes to school food services, point of purchase information (Table 1). An education component was delivered in the majority of studies. Six (of 7) studies in older children, and 17 (of 23) studies in younger children had some educational component. Many studies delivered a specifically tailored education resource or curriculum, although some schemes ran concurrently with other national curriculum initiatives on healthy eating. 32 Parental involvement was a feature of 11 studies in younger children and 3 in older children. Parents were involved in a

7 Table 3 Details of secondary school studies Study Design Participants Data collection Intervention Results Norwegian School Fruit Programme Fruit and Vegetables Make the Marks 29 Norwegian School Fruit Programme 31,60 TEENS study, USA 41 Planet Health USA 61 Belgium 27 School Garden project, USA 20 Gimme 5, USA 25 Cluster randomized controlled trial Follow up: 1 year Cluster randomized controlled trial Follow-up: 3 years Cluster randomized controlled trial Follow-up: 12 weeks (schools) Follow-up: 3 years 9 intervention schools 10 control schools 369 pupils age 11.3 at baseline 9 schools free fruit 9 schools paid 20 schools no fruit Total: 1950 students 16 schools with at least 20% of students approved for free and reduced price lunch and at least 30 students in each of 7th and 8th grades students 5 intervention and 5 control schools 1295 ethnically diverse grade 6 and 7 students 5 schools intervention with support 5 schools intervention no support 5 schools control 2840 pupils 6th grade students at 3 elementary schools. 99 students 9th grade students in 12 schools (6 matched pairs) 2213 students Survey 24 h FV recall parental Food frequency Survey Questionnaire Behavioural risk factor surveillance 24 h recall Food frequency s (also measured obesity, TV viewing hours) Food frequency s 1 subset completed assessments of physical activity Pupils receive free piece of F/carrot each day. Free fruit and educational programme Control: no intervention Initially free subscription to scheme then paid (E0.30) Control: no subscription scheme 4 groups Group 1: control Group 2: school environment interventions only Group 3: as 2 but with classroom lessons Group 4: as 3 but with peer leaders School based interdisciplinary intervention. Teacher training, classroom lessons, physical activity, wellness sessions Control: usual curriculum Increasing fruit to 2 pieces per day decreasing soft drinks, decreasing fat intakeenvironmental change focus with tailored computer feedback. Parental involvement. Control: no intervention 3 24 h recalls 1 group control 1 group nutrition education 1 group nutrition education plus gardening activities Knowledge, Attitudes and Practice Gimme 5 measurement + intervention - school wide media campaign, classroom activities, school meal modification, parental involvement Control: measurements without intervention FV all day and at school 0.6 portions higher in intervention Sustained in 2nd year (no longer had free fruit or education) Free fruit sustained effects on FV intake 3 years after intervention. Increased by g 1 day Significant increase in intervention group 4 with peer leaders (+0.9 servings per day, P = 0.012) at interim evaluation but no significant effect at 2 year follow-up. Higher increase in intervention group servings per day (P = 0.003) but only in girls No statistically significant difference in fruit intake. Statistically significant decrease fat intake in girlsincrease in physical activity at year 2 for both sexes Gardening students increased FV servings more than others. Combined FV intake inc to 4.5 servings per day from 1.93 No difference at follow-up. Initially reported consumption of FV servings was significantly higher in intervention schools but not sustained. 564 European Journal of Public Health Downloaded from at Pennsylvania State University on February 27, 2014

8 EU school fruit and vegetable review 565 variety of ways from helping with homework, 38 to accessing tools to judge their own FV intake. 39 A motivational component, for example, in the form of peer or fictional role models, or rewarding children for increasing intake, led to increased intake in three studies 34,36,40 42 in both age groups. This approach has been used successfully in a pilot of the Irish national Food Dude programme which has seen sustained results at 1 year. The use of incentives was effective in some studies 34,36,40 however, in others the positive effect was not maintained at follow-up. 24,43 It is difficult to unravel the effects of multiple intervention components. It is obviously preferable to determine which components contribute to effectiveness in order to improve potency and/or cost effectiveness. Only one randomised controlled trial explicitly studied the impact of different levels of exposure to a multicomponent intervention. 44 Neither school environment interventions on their own, nor environmental change plus curriculum sessions had a significant impact on FV intake. The addition of trained peer leaders led to a small, significant increase, but this was not sustained at 2 year follow-up. Seven studies targeted obesity reduction, with FV intake as a secondary aim. Only one study managed to produce positive impact on BMI and FV intake. 21 Other studies produced significant positive results in other aspects of diet e.g. fat intake. 22,23,27 DISCUSSION The finding that 70% of school schemes increase FV intake is very encouraging, and supports previous reviews of childhood FV interventions. 18,44 Only two previous reviews have concentrated on school interventions; one on US studies only, 45 and an older review of primary schools. 46 This review is more comprehensive in geographical scope, age and intervention types. Due to the heterogeneity of FV interventions, it is unsurprising that only one review has attempted meta-analysis, which was of seven US school-based FV studies showing a net increase of 0.45 servings per day (95% confidence interval (CI) ). 45 It is important to be aware of this review s limitations. Although increasing FV intake in children is likely to have health benefits worldwide, the included studies are limited to developed countries. Several studies from developing countries did not meet the inclusion criteria. This included novel interventions such as school gardening and school food policies in South America, Africa and Asia. It also included some national programmes, e.g. the New Zealand initiative whose one year evaluation showed both increases in FV consumption and decreased amount of TV children watched. 47 There was heterogeneity of both FV intake measurement and reporting across studies. Reliable data collection in children is complicated by difficulty with recall, social desirability and observer bias. 30 Not all studies used the same international standard definition of what to include in FV measurement. 15 Studies varied in reporting using both grams per day and servings per day (with varying definitions of serving sizes). Several lessons may be learnt from the review findings for increasing FV consumption in children. Much of the current focus for obesity policy is on younger children, with the perception that diets of younger children are easier to change. 12 This review shows that increasing FV intake is possible across a wide age range (70% of studies in both age groups). This is particularly important in teenagers when FV intake decreases dramatically. School FV schemes can also have the added benefit of reducing health and social inequalities. Children from low socio-economic status (SES) backgrounds traditionally have lower FV intake. One Norwegian study directly compared the effect of providing fruit through a subsidized subscription or free scheme. The free scheme was used by all groups whereas the subsidized scheme was mainly used by traditional high consumers (high SES). Compared to the subscription scheme, the free scheme reduced differences in FV intake between socioeconomic groups, with increases in FV intake sustained 3 years after the free programme. 31,48 Will school FV schemes result in long-term dietary changes? A total of 23 studies had follow-ups >1 year. Evaluation of the Norwegian programme after 3 years provides some evidence that large-scale schemes that increase FV availability can increase consumption in the long term. The evidence to date suggests that <1 year free FV is not sufficient for long-term dietary change. This has implications that any EU-funded programme should not only provide FV to children free of charge, but this should run over several years and allow further evaluation of long-term effectiveness. Over 75% studies involved a multicomponent approach. Although it is not possible to determine the most effective components, most programmes included some education (either integral or through simultaneous healthy eating initiatives). This review also shows that increasing children s access to FV can be achieved in a number of ways; changing school meals, snack provision, gardening, cooking or tasting programmes. Policy-makers, however, need to understand that multiple changes in social, economic and physical aspects of children s environments are also likely to be required to sustain increased FV intake, and that schools are only one aspect of this. It is important to recognize that school FV schemes can be organized in various ways (figure 1) and factors for successful implementation vary nationally depending on differences in the food supply chain and education system. For example, the review showed that schemes differed in whether the school or government was the purchaser, and whether FV was supplied direct from producers, wholesalers or retailers. The education systems of countries vary widely. Some countries provide cooked school lunches e.g. the United Kingdom, while others provide no meals with all children bringing food from home e.g. Ireland. National school food policies vary in terms of dietary guidelines and what food can be provided or consumed in schools. Educational curricula also differ in whether they include nutrition education and cooking skills. Clearly, any EU proposals for school FV schemes must have the potential to be adaptable to the varying national contexts. This is supported by the Pro Children intervention study which was conducted in three countries with different school systems. 39 The study combined free FV provision with educational initiatives and parental support. However, the timing of the FV provision differed between countries, as Norway and the Netherlands have no school meal provision. A strong partnership approach to implementation ensured that the intervention was adapted to the different country environments, resulting in an increase in FV consumption in intervention schools. A final lesson is that sustainable funding is essential as some national FV schemes (such as Denmark) initially failed due to lack of recurrent government resources and reliance on private industry and parental contribution. Clearly, EU funding to support national school FV provision would be a welcome initiative especially in central and eastern Europe where FV intake is lower. 8 To do this the EU have to be persuaded that they offer value for money. To date there has only been one Norwegian modeling study which attempted to measure the potential cost-effectiveness of school FV provision. This shows that even very small increases in lifelong FV intake make a school programme cost-effective, 49 and suggests that

9 566 European Journal of Public Health Provision to schools In school Out of school provision producers etai retail SCHOOL cafeteria/ lunch provision free FV at lunch i.e. child purchase e.g. local shops CHILD ani catering companies FV snack inbreak FV tasting in class Government procurement tuckshop / vending machines Lunchbox from home investment in school FV is justified even in the face of some remaining uncertainty over long-term effectiveness. The public consultation on the EU school FV policy proposed four options: Option 1: no change ; Option 2: creating networks of experts to advise best practice ; Option 3: Member States would submit proposals on a call for tender basis with funds allocated based on fulfilment of specific criteria; Option 4: EU-funded school FV provision for all member states with any additional programme elements funded nationally. Clearly Options 1 and 2 would do little to change the current situation, and Option 3 funding may increase inequalities as monies may only be awarded to countries with the resources to navigate complicated EC tendering processes. Option 4 was supported most strongly in the consultation and impact assessment processes by all sectors, and forms the basis of the Euro 90 million per year school fruit scheme proposed by the European Commission in July An EU supported school FV scheme appears to be logical public health and agricultural policy. Aiming at increasing intake amongst a new generation of consumers, it could also support other EU concerns including health inequalities, stimulating agricultural markets, and reducing health care costs 51 through prevention. EU proposals should be adaptable to context-specific factors, including differences in education systems, school meal programmes, supply chains and food cultures, and should be made sustainable through long-term funding. This policy process is significant as it would be one of the first public health policies organized and funded through the EU agricultural budget. This funding of health promoting policies from other policy sectors is an important example of new approaches for tackling diet-related disease. Acknowledgements JDS and KL presented this work as part of the EC impact assessment process for school fruit and vegetable schemes. They were not paid to do this work by any external source. The views expressed are the authors own. Conflicts of interest: None declared. Curriculum implementation Training/education of staff School nutrition policy Figure 1 Stages in the organization and delivery of school FV schemes Key points The EU is proposing to introduce policy and funding for school fruit and vegetable schemes as part of EU agricultural policy in autumn This policy is important as it would be one of the first public health policies organized and funded through the EU agricultural budget. Although some European countries have school FV programmes, previous evidence reviews of FV interventions in children have been limited in scope. School FV schemes are effective at increasing FV intake worldwide, and the results can be sustained long term in large national studies. Free school schemes can contribute to reducing inequalities in diet. There is currently little research on the impact of FV interventions on reducing overweight and obesity. References parental activities 1 European Commission. Regulation COM (2007) 17 final. Proposal for a COUNCIL REGULATION laying down specific rules as regards the fruit and vegetable sector and amending certain Regulations. Brussels, European Commission. Communication from the Commission to the Council and the Parliament. Preparing for the Health Check of the CAP reform. COM(2007) 722, 20 November 2007, Brussels. Available at: ec.europa.eu/agriculture/healthcheck/index_en.htm (Accessed 30 July 2008). 3 Commission of the European Communities. COM (207) 279 Final. White paper on a strategy for Europe on nutrition, overweight and obesity-related health issues. DG Sanco, SEC (2007) 706, 707, Brussels, FT Observer. Fruitless discussion. Financial Times. 07 June DG AGRI. School fruit scheme: increasing the consumption of fruit and vegetables by school children. Which role for the European Union? Brussels, Available at: index_en.htm (Accessed 30 July 2008). 6 World Cancer Research Fund/American Institute for Cancer Research. Food, nutrition, physical activity, and the prevention of cancer: a global perspective. Washington DC: AICR, Lock K, Pomerleau J, Causer L, et al. The global burden of disease attributable to low consumption of fruit and vegetables: implications for the global strategy on diet. Bull World Health Organ 2005;83:100 8.

10 EU school fruit and vegetable review Pomerleau J, Lock K, Mckee M. The burden of cardiovascular disease and cancer attributable to low fruit and vegetable intake in the European Union: differences between old and new member states. Public Health Nutr 2006;9: Ness AR, Maynard M, Frankel S, et al. Diet in childhood and adult cardiovascular and all cause mortality: the Boyd Orr cohort. Heart 2005;91: Maynard M, Gunnell D, Emmett PM, et al. Fruit, vegetables, and antioxidants in childhood and risk of adult cancer: the Boyd Orr cohort. J Epidemiol Community Health 2003;57: Lien N, Lytle L, Klepp KI. Stability in consumption of fruit, vegetables, and sugary foods in a cohort from age 14 to age 21. Prev Med 2001;33: Kelder SH, Perry CL, Klepp KI, Lytle LL. Longitudinal tracking of adolescent smoking, physical activity, and food choice behaviors. Am J Public Health 1994;84: te Velde SJ, Twisk JW, Brug J. Tracking of fruit and vegetable consumption from adolescence into adulthood and its longitudinal association with overweight. Br J Nutr 2007;98: Carlton Tohill B. Dietary intake of fruit and vegetables and management of body weight. Background paper for the Joint FAO/WHO Workshop on Fruit and Vegetables for Health, 1-3 September 2004, Kobe, Japan. Atlanta: Centers for Disease Control and Prevention, World Health Organization. Diet, Nutrition and the prevention of Chronic Diseases. Report of a Joint WHO/FAO Expert consultation. Geneva: WHO Technical Report Series 916, Yngve A, Wolf A, Poortvliet E, et al. Fruit and vegetable intake in a sample of 11-year-old children in 9 European countries: The Pro Children crosssectional survey. Ann Nutr Metab 2005;49: de Sa J, Lock K. School-based fruit and vegetable schemes: a review of the literature, London: LSHTM, December Available at: ecohost/projects/schoolfv.htm (Accessed 30 July 2008). 18 Knai C, Pomerleau J, Lock K, McKee M. Getting children to eat more fruit and vegetables: a systematic review. Prev Med 2006;42: Mangunkusumo RT, Brug J, de Koning HJ, et al. School-based Internettailored fruit and vegetable education combined with brief counselling increases children s awareness of intake levels. Public Health Nutr 2007;10: McAleese JD, Rankin LL. Garden-based nutrition education affects fruit and vegetable consumption in sixth-grade adolescents. J Am Diet Assoc 2007;107: Taylor RW, McAuley KA, Barbezat W, et al. APPLE Project: 2-y findings of a community-based obesity prevention program in primary school age children. Am J Clin Nutr 2007;86: Paradis G, Lévesque L, Macaulay AC, et al. Impact of a diabetes prevention program on body size, physical activity, and diet among Kanien kehá:ka (Mohawk) children 6 to 11 years old: 8-year results from the Kahnawake Schools Diabetes Prevention Project. Pediatrics 2005;115: Perry CL, Lytle LA, Feldman H, et al. Effects of the child and adolescent trial for cardiovascular health (CATCH) on fruit and vegetable intake. J Nutr Educ 1998;30: Baranowski T, Davis M, Resnicow K, et al. Gimme 5 fruit, juice, and vegetables for fun and health: Outcome evaluation. Health Educ Behav 2000;27: O Neil CE, Nicklas TA. Gimme 5: an innovative, school-based nutrition intervention for high school students. J Am Diet Assoc 2002;102(Suppl 3):S93 S6. 26 Haerens L, Deforche B, Maes L, et al. Evaluation of a 2-year physical activity and healthy eating intervention in middle school children. Health Educ Res 2006;21: Haerens L, De Bourdeaudhuij I, Maes L, et al. The effects of a middle-school healthy eating intervention or adolescents fat and fruit intake and soft drinks consumption. Public Health Nutr 2007;10: Ransley JK, Greenwood DC, Cade JE, et al. Does the school fruit and vegetable scheme improve children s diet? A non-randomised controlled trial. J Epidemiol Community Health 2007;61: Bere E, Klepp KI. Free vs. paid school fruit programme big difference with respect to social inequality. In: Kaiser M, M L, editors. Ethics and the politics of food Preprints of the 6th Congress of the European Society for Agricultural and Food Ethics. Wageningen, Netherlands: Wageningen Academic Publishers, 2006, Tak NI, Te Velde SJ, Brug J. Ethnic differences in 1-year follow-up effect of the Dutch Schoolgruiten Project promoting fruit and vegetable consumption among primary-school children. Public Health Nutr 2007;10: Bere E, Veierød M, Skare Ø, Klepp K-I. Free school fruit sustained effect three years later. Int J Behav Nutr Phys Activ 2007;4:5. 32 Fogarty A, Antoniak M, Venn A, et al. Does participation in a populationbased dietary intervention scheme have a lasting impact on fruit intake in young children? Int J Epidemiol 2007;36: Blanchette L, Brug J. Determinants of fruit and vegetable consumption among 6-12-year-old children and effective interventions to increase consumption. J Hum Nutr Diet 2005;18: Perry CL, Bishop DB, Taylor GL, et al. A randomized school trial of environmental strategies to encourage fruit and vegetable consumption among children. Health Educ Behav 2004;31: Anderson AS, Porteous LEG, Foster E, et al. The impact of a school-based nutrition education intervention on dietary intake and cognitive and attitudinal variables relating to fruits and vegetables. Public Health Nutr 2005;8: Horne PJ, Tapper K, Lowe CF, et al. Increasing children s fruit and vegetable consumption: a peer-modelling and rewards-based intervention. Eur J Clin Nutr 2004;58: Summerbell CD, Ashton AV, Campbell KJ, Edmunds L, Kelly S, Waters E. Interventions for treating obesity in children. Cochrane Database Syst Rev 2003(CD001872). 38 Friel S, Kelleher C, Campbell P, Nolan G. Evaluation of the Nutrition Education at Primary School (NEAPS) programme. Public Health Nutr 1999;2: Te Velde SJ, Brug J, Wind M, et al. Effects of a comprehensive fruit- and vegetable-promoting school-based intervention in three European countries: the Pro Children study. Br J Nutr 2008;99: Horne PJ, Hardman CA, Lowe CF, et al. Increasing parental provision and children s consumption of lunchbox fruit and vegetables in Ireland: the Food Dudes intervention. Eur J Clin Nutr 2008; May 21. [Epub ahead of print] PMID: Lytle LA, Murray DM, Perry CL, et al. School-based approaches to affect adolescents diets: results from the TEENS study. Health Educ Behav 2004;31: Lytle LA, Kubik MY, Perry C, et al. Influencing healthful food choices in school and home environments: results from the TEENS study. Prev Med 2006;43: Hendy HM, Williams KE, Camise TS. Kids Choice school lunch program increases children s fruit and vegetable acceptance. Appetite 2005;45: Birnbaum AS, Lytle LA, Story M, et al. Are differences in exposure to a multicomponent school-based intervention associated with varying dietary outcomes in adolescents? Health Educ Behav 2002;29: Sutcliffe Thomas J, Harden A, Oakley A, et al. Children and healthy eating: a systematic reveiw of barriers and facilitators. London: EPPI-Centre, Social Science Research Unit, Institute of Education, Howerton MW, Bell BS, Dodd KW, et al. School-based nutrition programs produced a moderate increase in fruit and vegetable consumption: meta and pooling analyses from 7 studies. J Nutr Educ Behav 2007;39: Burchett H. Increasing fruit and vegetable consumption among British primary schoolchildren; a review. Health Educ 2003;103: Boyd S, Dingle R, Campbell R, et al. Taking a bite of the apple: the implementation of fruit in Schools. (Healthy Futures Evaluation Report to the Ministry of Health). Wellington: New Zealand Council for Educational Research, Bere E, Veierod MB, Klepp K-I. The Norwegian school fruit programme: evaluating paid vs. no-cost subscriptions. Prev Med 2005;41: