Nutrition action in schools: a review of evidence related to the nutrition-friendly schools initiative

NUTRITION ACTION IN SCHOOLS: A REVIEW OF THE EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE

NUTRITION ACTION IN SCHOOLS: A REVIEW OF THE EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Nutrition action in schools: a review of evidence related to the Nutrition-Friendly Schools Initiative ISBN 978-92-4-151696-9 (electronic version) ISBN 978-92-4-001632-3 (print version) © World Health Organization 2020 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization. Suggested citation. Nutrition action in schools: a review of evidence related to the Nutrition-Friendly Schools Initiative. Geneva: World Health Organization; 2020. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see http://www.who.int/about/licensing. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. Photo credit (front cover): @ Ingimage and playactioninternational.org. Cover design and layout: Alberto March (Barcelona, Spain) CONTENTS ACKNOWLEDGEMENTS v ABBREVIATIONS AND ACRONYMS vi EXECUTIVE SUMMARY vii BACKGROUND 1 METHODS 6 Study design and evidence considered 6 Review of reviews 6 Search strategies and inclusion criteria 6 Assessing the quality of review articles (Modified AMSTAR) 6 Data synthesis 7 RESULTS 9 Summary characteristics of the evidence identified 9 Search results 9 Review designs, study settings and population 10 Relevance of interventions and programmes to the NFSI essential criteria 11 Outcomes reviewed 13 Narrative synthesis and summary of the synthesized evidence 13 NFSI Component 1: Having a nutrition-friendly schools policy 13 NFSI Component 2: Enhancing awareness and capacity building of the school community 28 NFSI Component 3: Developing a nutrition and health promoting curriculum 37 NFSI Component 4: Creating a supportive environment 52 NFSI Component 5: Providing supportive school nutrition and health services 71 DISCUSSION AND CONCLUSIONS 74 Summary of findings 74 Strengths and limitations 75 Future research 76 Conclusion 78 REFERENCES 80 ANNEX 1 SUMMARY OF EXPERIENCE IN PILOTING, IMPLEMENTING AND EVALUATING THE NFSI IN COUNTRIES 93 ANNEX 2 DETAILS OF REVIEWS INCLUDED 95 iii @ In gi m ag e ACKNOWLEDGEMENTS T he review of evidence related to the Nutrition-Friendly Schools Initiative was led by Ms Kaia Engesveen under the guidance of Dr Chizuru Nishida of the World Health Organization (WHO) Nutrition Policy and Scientific Advice Unit, Department of Nutrition and Food Safety. Ms Krista Lang provided technical support throughout different stages of the process, and contributed to evidence collection, analysis and report preparation. An earlier draft was developed by Dr Julianne Williams under the guidance of Dr Kremlin Wickramasinghe and Dr Nick Townsend, all three were then at the University of Oxford, as well as by Ms Savanna Henderson and Ms Nabeeha Kazi, both then at Humanitas Global. Special thanks are due to the experts and focal points for school health and nutrition in our partner agencies, as well as WHO staff from different programmes, who helped shape the review by providing peer-review comments: Dr Hélène Delisle (University of Montreal), Ms Melissa Vargas (Food and Agriculture Organization of the United Nations), Ms Deepika Sharma (United Nations Children’s Fund), Dr Mutinta Hambayi, Ms Domitille Kauffmann, Ms Jutta Neitzel and Ms Emilie Sidaner (World Food Programme), and Dr Valentina Baltag, Dr Katrin Engelhardt, Ms Kid Kohl, Mr Lendert Nederveen, Ms Scarlett Storr and Dr Julianne Williams (WHO). Special appreciation is also expressed to the interns who helped with searches, evidence compilation and reviews: Dr Sarah Krull Abe, Ms Phani Jakkilinki, Ms Rewena Mahesh, Dr Ana Elisa Pineda, Dr Bruno Sunguya, Ms Marisa Tsai, Dr Elisa Vargas Garcia and Ms Camilla Haugstveit Warren. Ms Karen McColl supported the technical editing and finalization of the report. v ABBREVIATIONS AND ACRONYMS AMSTAR Assessment of Multiple Systematic Reviews BMI body mass index CBA controlled before and after CI confidence interval DBP diastolic blood pressure eLENA WHO e-Library of Evidence for Nutrition Actions F&V fruit and vegetables FAO Food and Agriculture Organization of the United Nations FFE food for education GNPR2 Second WHO Global Nutrition Policy Review 2016-2017 HAZ height-for-age z-score HDL-C high-density lipoprotein-cholesterol HIC high-income countries HPS WHO Health Promoting Schools ICT information and communication technology IDD iodine deficiency disorders LDL-C low-density lipoprotein-cholesterol LIC low-income countries LMIC lower-middle-income countries MD mean difference MMN multiple micronutrient MNP micronutrient powders MUAC mid-upper arm circumference NCD noncommunicable disease NFSI Nutrition-Friendly Schools Initiative NUGAG WHO Nutrition Guidance Expert Advisory Group OR odds ratio PA physical activity PE physical education PR prevalence ratio RCT randomized controlled trial RR relative risk SBP systolic blood pressure SDG Sustainable Development Goal SDS standard deviation score SMD standard mean difference SSB sugar-sweetened beverage UMIC upper-middle-income countries UNESCO United Nations Educational, Scientific and Cultural Organization UNICEF United Nations Children’s Fund UNSCN United Nations System Standing Committee on Nutrition USDA United States Department of Agriculture WASH water, sanitation and hygiene WAZ weight-for-age z-score WFP World Food Programme WHO World Health Organization WMD weighted mean difference vi NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE EXECUTIVE SUMMARY Good nutrition during childhood and adolescence is key to ensuring optimal growth, health and well-being. Healthy dietary practices in early life have an immediate impact on healthy growth and help prevent noncommunicable diseases (NCDs) later in life and across generations. Good nutrition in childhood is also good for school performance and educational outcomes, impacting on nations’ economic and social development. Since children spend so much time in school, the school environment is an important setting for children to acquire habits, skills and knowledge related to healthy diets and physical activity. In 2016-17, the vast majority of countries (89%) reported implementation of school health and nutrition programmes although inclusion of a comprehensive set of interventions was rare and implementation of such programmes appears to have weakened since the start of the decade. The Nutrition-Friendly Schools Initiative In 2006, building on existing programmes and initiatives including the Health Promoting School approach and the FRESH (Focusing Resources on Effective School Health) Initiative, WHO and partners launched the Nutrition- Friendly Schools Initiative (NFSI) to provide a framework for ensuring integrated school-based programmes which address the burden of nutrition-related ill health, whether that be undernutrition and/or overweight/ obesity. The NFSI has since been used around the world, including as a self-appraisal tool for schools in 18 countries, in national NFSI programmes and as part of academic research and evaluation projects. The NFSI Framework outlines 26 essential criteria within five broad components: 1. school nutrition policies 2. awareness and capacity building of the school community 3. nutrition and health promoting curricula 4. supportive school environment for good nutrition 5. supportive school nutrition and health services. Review methodology and evidence identified This review aims to identify and summarize the synthesized evidence from reviews related to the five components and the 26 essential criteria of the NFSI, and to highlight areas where more research is needed. In total, 117 reviews were included: 43 meta-analyses, 63 systematic reviews (including eight Cochrane reviews) and 11 other reviews. The studies included in the reviews were published between 2006 and 2019, and most reviews required studies to follow an experimental design. The vast majority of studies included were based in high-income countries, and only nine reviews exclusively or largely included studies exclusively from low- and middle-income countries while another four reviews focused on disadvantaged population groups in high-income countries. Most reviews considered studies at the primary and/or secondary school level, while fewer included studies in preschool establishments. A 10-item modified version of the Assessment of Multiple Systematic Reviews (AMSTAR) method was used to assess the quality of the reviews collected. vii @ In gi m ag e viii NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE @ Ingimage The included reviews provided an analysis of a wide range of school-based interventions. There was a high level of heterogeneity in relation to the school-level exposures, the impact on children and the outcome measures. Due to the differences in study aims, methods and measures, meta- analysis was not appropriate and the findings are summarized in a narrative way. Summary of findings Evidence from the included reviews is summarized for each of the relevant NFSI components, with many reviews relevant to multiple components and criteria. NFSI Component 1: Having a written nutrition- friendly schools policy Essential criteria 1.1 Development of a written nutrition-friendly schools policy that addresses all five points described in this framework, and includes elements described in 1.2 - 1.5 1.2 Rationale 1.3 Objectives 1.4 Action plan 1.5 Monitoring and evaluation plan Twenty-three meta-analyses and 28 reviews were identified that examined various aspects of comprehensive school nutrition policies and programmes relevant to essential criterion 1.1. Findings suggest that comprehensive school nutrition policies using multiple components and approaches (e.g. diet, physical activity, educational interventions, environmental changes) are associated with positive weight- related, dietary and other outcomes among school children. Improvements in BMI standard mean difference (SMD) ranged from <-0.1kg/m2 to -1.8kg/m2 through school-based programmes. The majority “ The majority of the reviews considering multicomponent programmes combining diet and physical activity interventions found this combination effective in addressing weight-related outcomes” of the reviews considering multicomponent programmes combining diet and physical activity interventions found this combination effective in addressing weight-related outcomes. A smaller number of reviews examined combinations of approaches, e.g. nutrition improvement through curriculum, environment and parental involvement, and also found positive effects. The four reviews which focused specifically on low- and middle-income countries or vulnerable populations in high-income countries reported positive effects on weight-, diet- or health behaviour-related outcomes from multicomponent ixEXECUTIVE SUMMARY school nutrition programmes. Only one scoping review which examined facilitators and barriers to implementing school nutrition policies was identified that could shed some light on essential criteria 1.2 to 1.5. NFSI Component 2: Enhancing awareness and capacity building of the school community Essential criteria 2.1 Dissemination of the nutrition-friendly schools policy 2.2 Activities for families and community, community involvement and outreach in the area of nutrition and health related issues 2.3 School staff training in nutrition and health related issues Fourteen meta-analyses and 18 reviews were identified relevant to essential criterion 2.2, examining family and community involvement in school nutrition programmes. A large number of reviews found positive effects from enhancing awareness and capacity building of the wider school community. The effect of engaging families and communities through community involvement and outreach in the area of nutrition and health-related issues was the most frequently studied intervention. Parental involvement was identified as a moderating factor leading to better results (e.g. on body mass index (BMI) or other weight-related, biochemical or dietary outcomes), with a greater impact for a higher level of parental involvement. Direct involvement methods (e.g. face-to-face), where parents are expected to take an active role, were more effective than indirect methods (e.g. newsletters), where parents may remain passive recipients of information. Less evidence was identified for the effect of staff training in delivering nutrition and healthy diet interventions relevant to essential criterion 2.3, but from the five reviews identified beneficial effects were noted from training teachers and involving experts in school-based nutrition activities. No reviews were identified relevant to essential criterion 2.1. NFSI Component 3: Developing a nutrition and health-promoting school curriculum Essential criteria 3.1 Culturally-appropriate and effective nutrition education 3.2 Age, sex and culturally-appropriate physical education curriculum 3.3 Healthy living and life-skills education curriculum 3.4 Regular monitoring of school curriculum relevant to NFSI, and evaluation of impact of how well the education meets the objectives “Parental involvement was identified as a moderating factor leading to better results (e.g. on body mass index (BMI) or other weight- related, biochemical or dietaryoutcomes) with a greater impact for a higher level of parental involvement” x NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Ten meta-analyses and 28 reviews were identified relevant to essential criterion 3.1 that examined various aspects of nutrition education in schools. Eight of the 10 meta-analyses found significant improvements in weight-related or dietary outcomes. Effective nutrition education programmes were behaviourally-focused, context and age-specific, of longer duration and intensity, facilitated by trained staff, integrated into the curricula and/or implemented in combination with other approaches (e.g. changes to the environment). Hands-on learning strategies and practical skills were considered promising, as were those that built children’s self-efficacy in making healthy changes in their diet and physical activity levels. Thirteen meta-analyses and four systematic reviews were identified relevant to essential criterion 3.2 comparing physical activity education alone or in combination with nutrition activities. The majority of these found that physical activity education was more effective when combined with nutrition interventions. Eight reviews were identified relevant to essential criterion 3.3 examining life-skills education in the context of nutrition. Effective intervention components were suggested to be experiential learning, such as cooking or gardening skills. Self-efficacy and children’s perceived competence in making healthy changes to their diet and physical activity levels were found to be associated with better outcomes. No reviews were identified relevant to essential criterion 3.4. NFSI Component 4: Creating a supportive school environment Essential criteria 4.1 School meals, food vendors and snack bars (if present) promote healthy eating 4.2 Positive messaging towards nutrition and active living 4.3 Absence of marketing of foods and beverages at school 4.4 Access to an adequate eating place if the school provides food and/or beverages to the children 4.5 Adequate school cooking facilities 4.6 Access to safe drinking water 4.7 Promotion of safe hygiene and sanitary behaviour 4.8 Availability of clean and separate toilets, for boys and girls 4.9 Opportunity for all age groups to access space and school sporting facilities for physical activity within and outside of the curriculum 4.10 Affirmative action against bullying, stigmatization and discrimination 4.11 School staff as role-models in encouraging healthy eating and healthy lifestyle Nine meta-analyses and 24 reviews were identified that considered either the availability of foods and beverages or the provision of foods, beverages or snacks, relevant to essential criterion 4.1. Many of these reviewed current policies and standards implemented in countries. There was a considerable body of evidence to support the creation of healthy environments for nutrition, particularly through increasing availability of healthy foods and beverages and simultaneously reducing availability of unhealthy foods and beverages. Positive effects were reported on dietary and food availability outcomes from school lunch standards, restrictive policies on unhealthy foods and beverages and school food procurement policies. Provision of school food had positive effect in low-income or disadvantaged settings, including school feeding programmes, provision of fortified “ Effective nutrition education programmes were behaviourally-focused, context and age-specific, of longer duration and intensity, facilitated by trained staff, integrated into the curricula and/ or implemented in combination with other approaches” xiEXECUTIVE SUMMARY foods and school breakfast programmes. Provision of fruit and vegetables in schools increased children’s consumption. Less evidence was identified relevant to other essential criteria in this NFSI component. Relevant to essential criterion 4.2, two meta-analyses and seven systematic reviews considered messaging, incentivising desired behaviour or use of labelling to encourage healthy eating in the school setting. These reported positive effects on diet-related outcomes but mixed results for weight-related outcomes. Relevant to essential criterion 4.3, one review found that children are commonly exposed to marketing of unhealthy foods and beverages in and around schools and that restrictive policies — where they existed — were often breached. Relevant to essential criteria 4.4 and 4.5, one systematic review identified overcrowding in the cafeteria or lack of appropriate kitchen facilities and equipment as barriers to healthy school food environments. Relevant to essential criterion 4.6, three systematic reviews analysed the effect of access to safe drinking water on consumption of water or sugar-sweetened beverages (SSBs) and weight-related outcomes. Relevant to essential criterion 4.7, one meta-analysis and one systematic review found that adequate hand washing and sanitation in schools reduced risk of diarrhoea, amongst other outcomes. Relevant to essential criterion 4.11, one meta-analysis and two systematic reviews found that modelling healthy dietary behaviours by parents, peers or teachers had positive effects. No reviews were identified relevant to essential criteria 4.8, 4.9 and 4.10. NFSI Component 5: Providing supportive school nutrition and health services Essential criteria 5.1 Regular monitoring of children’s growth development 5.2 Effective feedback system for parents and children on findings of the regular monitoring 5.3 Supportive school health service, including a referral system Some evidence was identified regarding supportive school health services for nutrition. One meta-analysis and one review examined regular monitoring and feedback for children’s growth, relevant to essential criteria 5.1 and 5.2. School-based weight monitoring and screening programmes demonstrated positive effects in both children and their families. Three meta- analyses and one review were identified relevant to essential criterion 5.3, that examined the impact of delivering nutrition interventions through school health services. School health service delivery of interventions such as micronutrient supplementation, deworming and nutrition counselling also had positive results and these findings support the use of schools as a delivery platform for essential nutrition actions to school- age children and adolescents. “ There was a considerable body of evidence to support the creation of healthy environments for nutrition, particularly through increasing availability of healthy foods and beverages and simultaneously reducing availability of unhealthy foods and beverages” xii NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Strengths and limitations The bulk of evidence considered in this review was from high-income countries focusing on obesity prevention and this imbalance constitutes a major bias. Other limitations of the review include the high level of heterogeneity amongst studies in the reviews, the likelihood of small study bias and the difficulty of drawing conclusions about the effects of programme components individually when they are implemented as part of a multicomponent package. There were also great differences in categorization (e.g. of “diet” and “physical activity” interventions or content and intensity of “nutrition education” interventions) between reviews. In addition, many of the studies included relied on cross-sectional data which are unable to measure long-term effects. Nonetheless, this review has comprehensively and systematically assessed and summarized the evidence underpinning the NFSI criteria, on the basis of 117 reviews synthesizing findings from primary studies. Conclusions This review of reviews provides an extensive summary of the evidence for the effectiveness of many components of the NFSI framework. The findings may be used — in conjunction with existing UN and WHO guidance and tools — to inform the work of governments, policy- makers and researchers concerned with school- based health and nutrition promotion programmes and initiatives. A strong evidence base to support many specific criteria of the NFSI was identified. This highlighted, for example, the importance of having comprehensive nutrition-related school policies, involving parents in school-based nutrition interventions, implementing effective nutrition and physical activity education, creating healthy school food environments and integrating nutrition into school health services through, for example, school-based monitoring of children’s growth. Synthesized evidence on the nutrition impact of other criteria, however, such as the dissemination of school nutrition policies, monitoring and evaluation of school nutrition-related curricula, separate toilets for boys and girls, access to physical activity spaces and the need to prevent bullying, do not yet exist. The review also points to the need for more research on school nutrition policies and programmes in low- and lower-middle income countries and on educational outcomes, as well as health and nutrition outcomes. Member States are encouraged to draw on this evidence base and to use the NFSI framework to build school nutrition-related policies and programmes appropriate to their particular country schooling system and nutrition situation. In doing so, they can fully realize the potential of schools as a key setting for improving nutrition throughout childhood and adolescence, while laying good foundations for health and well- being later in life. @ Ingimage “School health service delivery of interventions such as micronutrient supplementation, deworming and nutrition counselling also had positive results and these findings support the use of schools” xiiiEXECUTIVE SUMMARY playactioninternational.org xiv NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE BACKGROUND Nutrition during childhood and adolescence is key to ensuring optimal growth, health and well-being during childhood and beyond (1, 2). Healthy dietary practices and the foundation for good nutrition are initiated in early life stages, with an immediate impact on healthy growth during this period characterised by rapid growth spurts. They also have long-term health impact, such as the prevention of noncommunicable diseases (NCDs) later in life, as well as inter-generational impact through ensuring optimal nutrition status of mothers, particularly those who are adolescent girls (1, 3). Moreover, a child who has good health and nutrition performs better in school (4, 5), and a good education provides a child with a foundation for the future, which in turn impacts on nations’ economic and social development (6, 7). While the focus of the critical period to ensure good nutrition typically has been on the first 1000 days — from conception to a child’s second birthday — there is now growing recognition of the importance of the subsequent 7 000 days, which includes critical childhood and adolescent periods up to the age of 21, to reach their full development potential (8, 9). From a human rights perspective, school- based nutrition policies and programmes may be an effective mechanism for contributing to the realization of children’s rights to the highest attainable standard of health, to adequate food and to education. Schools play an important role in promoting nutrition by providing access to children from a range of socioeconomic backgrounds. Young people spend a considerable amount of time in school, so school-based programmes will reach many children in a relatively short period of time and, therefore, the school environment is an ideal setting to acquire habits, skills, and knowledge related to healthy diets and physical activity (10). Additionally, school programmes provide an infrastructure (such as a network of teachers and staff, along with facilities) which may contribute to the sustainability of an intervention over time (10, 11). The second Global nutrition policy review 2016-2017: country progress in creating enabling policy environments for promoting healthy diets and nutrition (GNPR2) (12) undertaken in 2016- 2017 reported that 89% of 160 countries reporting on school health and nutrition programmes implemented related actions. The most frequently reported component of school health and nutrition programmes was nutrition education being included in the school curriculum (61%), followed by training of school staff on health and nutrition (56%). More than half of the countries had standards or rules for foods and beverages in schools (54%), provided school meals (54%) or made safe drinking water available free of charge (53%). Provision of school meals was most common in the WHO regions of Africa and South- East Asia; however, not all countries providing meals had standards or guidance for school meals. Other actions to support healthy diets or improve the school food and beverage environment were less common. Only 18% of countries had a ban on vending machines, 24% had standards for regulating marketing of foods and beverages in schools and 30% had school fruit and vegetable schemes. Just over half of the countries provided any fruit and vegetables at school through school meals or fruit and vegetable schemes. But there were misconceptions about healthy options, with several countries providing 100% fruit juices, which often contain as much (or even more) free sugars as some other SSBs including soft drinks. The most common school health and nutrition services were growth monitoring (43%), deworming (36%) and micronutrient supplementation (19%). Furthermore, in contrast to other nutrition- related programme areas surveyed in the GNPR2, there had been a notable weakening in most specific school health and nutrition programme components since the first such review in 2009-2010 (13). All 1 the regions saw large overall decreases in most programme components, although there were some exceptions; for example, increased inclusion of deworming and safe drinking water in the WHO African Region and of growth monitoring and deworming in the WHO Eastern Mediterranean Region. This may well reflect an increased focus by national nutrition authorities on the 1000 days window, which is a critical period for healthy growth and development, as well as prevention of stunting. Nonetheless, it is during school years that children can and should learn the importance of a healthy diet and acquire healthy eating habits, to help prevent all forms of malnutrition. There are, however, still countries where school food programmes are not linked to nutrition and health considerations. In a recent survey of school meal programmes in 33 low- and middle- income countries by the Food and Agriculture Organization of the United Nations (FAO) (14), some country programmes did not include as main objectives improving nutrition, learning healthy dietary habits or improving the quality of school meals. The importance of schools in promoting population health is widely acknowledged. WHO launched the Global School Health Initiative in 1995 with the intention of strengthening health promotion and educational activities aimed at improving the health of students and the surrounding school community. Since then, numerous countries have implemented the WHO Health Promoting Schools approach based on the Global School Health Initiative. A health promoting school constantly seeks to strengthen its capacity to promote healthy living, learning and working conditions. It aims to provide a multifaceted response to the health needs of students. Its key features are to engage health education and community leaders, provide a safe healthy environment, provide skills-based health education, provide access to health services, improve health promoting policy and practice and improve the health of the community . Healthy nutrition has been considered an “essential element of a health promoting school” (15). WHO, in partnership with FAO, the United Nations Children Fund (UNICEF), the United Nations Educational, Scientific and Cultural Organization @ In gi m ag e 2 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE (UNESCO), the World Food Programme (WFP), the World Bank Group, Education Development Center, Save the Children (SC), Partnership for Child Development, University of Montreal, Durham University and the United Nations System Standing Committee on Nutrition (UNSCN), as well as government agencies from Brazil, Finland and Ireland, developed the Nutrition-Friendly Schools Initiative (NFSI). The initiative was launched in 2006 to provide a framework for ensuring integrated school-based programmes which address the double burden of nutrition-related ill health and to become the nutrition module of the Health Promoting Schools. The NFSI builds on and connects the work from other programmes such as the Focusing Resources on Effective School Health (FRESH) Initiative (16), UNICEF and WFP’s Essential Package (17), UNICEF’s Child-Friendly Schools (18), WHO’s Health Promoting Schools (19) and FAO’s curriculum planning guide for nutrition education in primary schools (20), to mention a few. The rationale behind the development of the NFSI was to provide common policy options that enable schools to implement programmes and improve the health and nutrition status of young people. The NFSI targets the school setting, including preschool, primary and secondary schools. Building on the FRESH framework and the Health Promoting School approach, the NFSI Framework outlines 26 essential criteria within five broad components: 1. school nutrition policies 2. awareness and capacity building of the school community 3. nutrition and health promoting curricula 4. supportive school environment for good nutrition 5. supportive school nutrition and health services (Box 1). The components and essential criteria were developed by WHO and partners based on a number of sources. These include the outcomes of the WHO Expert Meeting on Child Obesity, held in Kobe, Japan in 2005 (21), a conceptual meeting held in Montreux, Switzerland in 2006 (22), the content of ongoing school-based programmes to address all forms of malnutrition, a 2005 Cochrane systematic review (23), a 2006 review of studies by the National Institute for Health and Clinical Excellence in the United Kingdom (24) and a 2006 qualitative review by Flynn et al. (25). In the years following its development, the NFSI has been used around the world. As a self- appraisal tool for schools, NFSI has been piloted in 18 countries, mainly in the European region and a handful of other countries including Brazil, India, Benin and Burkina Faso (26). Some countries still have national NFSI policies and programmes, such as Turkey where the General Directorate of Public Health reports that more than 16 000 schools in the country participate in the Turkish Nutrition-Friendly Schools Program (27). The NFSI framework was recently used by University of Oxford researchers to evaluate school nutrition policies before and after multisectoral health promotion programmes and initiatives. A brief summary of experience in piloting, implementing and evaluating the NFSI is provided in Annex I. To date, there has not been a published review of the scientific evidence behind all five components of the NFSI in one single document, including the relationship between the recommendations and the associated nutrition- related outcomes among school children. The purpose of this review is to identify and summarize the synthesized evidence related to the NFSI components and essential criteria, and to highlight areas where more research is needed. Since the development of the NFSI Framework in 2006, WHO has taken important steps to further advance the agenda of promoting healthy diets and good nutrition in the school setting. In 2016, the implementation plan for the recommendations of the Commission on Ending Childhood Obesity (28) identified settings such as schools and child care 3BACKGROUND Box 1. The Nutrition-Friendly Schools Initiative framework 1. Having a written nutrition-friendly schools policy Essential criteria 1.1 Development of a written nutrition-friendly schools policy that addresses all five points described in this framework, and includes elements described in 1.2 - 1.5. 1.2 Rationale: e.g. promotion of healthy diet and eating practices, increased physical activity; prevention of undernutrition, micronutrient deficiencies, overweight and/or obesity and other nutrition-related chronic diseases; also, contributory conditions such as unhygienic school environments, untrained school staff, lack of information in the community. 1.3 Objectives: with timelines and clear milestones. 1.4 Action plan: Outline of the whole-school approach contributing to healthy living and lifestyles to include description of processes, organizational structure, roles and responsibilities, principles of rights/ equity/ non- discrimination and strong commitment by all concerned stakeholders, including families and communities. 1.5 Monitoring and evaluation plan for the nutrition-friendly schools policy. 2. Enhancing awareness and capacity building of the school community Essential criteria 2.1 Dissemination of the nutrition-friendly schools policy. 2.2 Activities for families and community, community involvement and outreach in the area of nutrition and health- related issues. 2.3 School staff training in nutrition and health-related issues. 3. Developing a nutrition and health-promoting school curriculum Essential criteria 3.1 Culturally-appropriate and effective nutrition education. 3.2 Age, sex and culturally-appropriate physical education curriculum. 3.3 Healthy living and life-skills education curriculum. 3.4 Regular monitoring of school curriculum relevant to NFSI, and evaluation of impact of how well the education meets the objectives. 4. Creating a supportive school environment Essential criteria 4.1 School meals, food vendors and snack bars (if present) promote healthy eating. 4.2 Positive messaging towards nutrition and active living. 4.3 Absence of marketing of foods and beverages at school. 4.4 Access to an adequate eating place if the school provides food and/or beverages to the children. 4.5 Adequate school cooking facilities. 4.6 Access to safe drinking water. 4.7 Promotion of safe hygiene and sanitary behaviour. 4.8 Availability of clean and separate toilets, for boys and girls. 4.9 Opportunity for all age groups to access space and school sporting facilities for physical activity within and outside of the curriculum. 4.10 Affirmative action against bullying, stigmatization and discrimination. 4.11 School staff as role-models in encouraging healthy eating and healthy lifestyle. 5. Providing supportive school nutrition and health services Essential criteria 5.1 Regular monitoring of children’s growth development. 5.2 Effective feedback system for parents and children on findings of the regular monitoring. 5.3 Supportive school health service, including a referral system. 4 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE facilities as key settings for delivering nutrition and health education, increasing opportunities for physical activity and promoting a healthy school environment. In 2017, WHO and partners launched the Global Accelerated Action for the Health of Adolescents (AA-HA!): Guidance to support country implementation document, which provides guidance for implementing comprehensive adolescent health programmes and services specific to the health needs of adolescents, including nutrition interventions (10, 11). Many actions suggested by the AA-HA! Guidance could be offered and implemented in the school environment. In 2018, WHO and UNESCO launched the Making Every School a Health Promoting School initiative through the development and promotion of Global Standards for Health Promoting Schools (29). The initiative will serve over 2.3 billion school-age children, and will contribute to WHO’s Thirteenth General Programme of Work 2019 - 20231 target of achieving 1 billion more people enjoying better health and well-being by 2023 (30). Furthermore, WHO is currently developing guidelines on school food and nutrition policies, in the context of protecting and promoting healthy diets for children, as well as on school health services including nutrition. 1 WHA 71.1 @ Ingimage 5BACKGROUND METHODS STUDY DESIGN AND EVIDENCE CONSIDERED T his review draws upon research findings from reviews that pertain to the essential criteria of the NFSI. These criteria cover a broad range of research and the relevant scientific literature is very heterogeneous in terms of research questions, study design, interventions and outcome measures. Therefore, meta-analysis was not appropriate for this review and findings from the relevant literature are summarized using a narrative approach. REVIEW OF SYSTEMATIC REVIEWS Search strategies and inclusion criteria To be included in this exercise, reviews were required to present results of interventions or exposures to school components that relate to those recommended within the NFSI essential criteria (Box 1). Since school-based programmes seldom focus on single interventions, reviews were included if they evaluated the effect on nutrition- related outcomes1 of school-based interventions related to the NFSI essential criteria as primary intervention(s) or in combination with other interventions, in a way that made it possible to link findings to specific essential criteria. For reviews considering multiple settings to be included, either a majority of the primary studies should have taken place in the school setting or the review presented separate subgroup analyses for the school setting. There were no geographical restrictions. Reviews were identified through a keyword search for synthesized literature conducted in school settings assessing the impact of interventions related to the 26 NFSI essential criteria (e.g. school policies, community and parental involvement, training of staff, curricula and education, school foods and beverages and other school nutrition-related environment, nutrition services delivered in schools) and measuring nutrition-related outcomes (e.g. weight- or diet-related, food literacy). 1 Reviews analysing the effect of interventions on outcomes relating to nutrition status, diet and nutrition-related disease were included, whereas reviews analysing the effect of interventions solely on outcomes not directly relating to nutrition status – such as school performance or physical activity levels – were excluded. The PubMed, Campbell, 13E UKAID, Cochrane library and PDQ-Evidence databases were searched on multiple occasions between 2012 and 2019 for relevant reviews published in English language between January 2006 and August 2019. Additional reviews were added as result of a peer review in 2019 by school health and nutrition programme focal points in the United Nations agencies and experts who have supported the implementation of the NFSI. One researcher reviewed titles and abstracts from the search results. Cases of doubt were resolved via consensus with another reviewer. Two research team members reviewed most of the remaining full text articles. An extraction sheet was used to identify each study’s aims, methods, inclusion criteria, study population and setting, overall results (e.g. description of studies included, study effect of interventions) and relevance to the NFSI essential criteria. Assessing the quality of review articles (Modified AMSTAR) A 10-item modified version of the Assessment of Multiple Systematic Reviews (AMSTAR) method was used to assess the quality of the reviews and systematic reviews collected. AMSTAR was developed using factor analysis to reduce a 37-item assessment tool, the Overview Quality Assessment Questionnaire, to a shorter one (31). AMSTAR was tested for validity (32-34) and reliability (34) and was found to have good agreement, reliability, construct validity and feasibility. In this review of reviews, a modified version of this tool was used, which retained 6 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE most of the items, including research design, data extraction methods, literature search methods, publication status of included reviews, description of inclusion/exclusion process and results, assessment study quality, testing for homogeneity/heterogeneity and disclosure of potential conflict of interest. Weighting was applied to 10 items, each with between one and six sub items, with each sub item equally summed up to give one mark per item. The total of the 10 items (marks) was then divided into terciles, designating low-, medium-, and high-quality grades, respectively. Data synthesis As noted previously, due to differences in study aims, methods, and measures, meta-analysis was not appropriate. Therefore, findings are summarized in a narrative way, grouped according to the relevant NFSI criteria. Many studies were relevant to more than one component of the NFSI. WHO/Yoshi Shimizu 7METHODS WHO/Yoshi Shimizu 8 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE @ UNICEF RESULTS SUMMARY CHARACTERISTICS OF THE EVIDENCE IDENTIFIED Search results T he search as described was executed on various occasions over the period during which this review has been conducted. This resulted in initial search results of more than 2 000 results. After title and abstract screening, 228 articles were evaluated based on the full text, of which 117 reviews met the inclusion criteria and 111 reviews were excluded.1 The included reviews were published between 2006 and 2019, with 60% of the reviews published since 2014 (Fig. 1). Details of the reviews are included in Annex II. Figure 1. Number of reviews included by year of publication 0 2 4 6 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 3 2 5 18 5 10 15 10 18 2018 2019 8 6 3 3 8 11 10 12 14 16 18 20 1 The 111 reviews were excluded because they were not conducted in a school setting or did not provide subanalysis of school-based studies (13 reviews), did not include interventions relevant to the NFSI framework (6), lumped interventions together in the analyses so they could not be allocated to specific NFSI essential criteria (39), did not evaluate nutrition interventions or measure impact on nutrition outcomes (19), did not evaluate the effectiveness of interventions (22) or focused on physical activity interventions only (12). 9 Review designs, study settings and population Of the 117 reviews included, there were 43 meta-analyses, 63 systematic reviews (including eight Cochrane reviews) and 11 other reviews (e.g. literature reviews, scoping reviews). Most of these reviews required studies to follow an experimental design, including randomized controlled trials (RCTs) or other studies with pre- and post-intervention tests and a control group. All of the reviews included children – ranging from 2 years up to 19 years of age – in the pre-, primary or secondary school setting. Eighty-six of the reviews (75%) only included studies conducted in these settings and another 22 reviews (19%) had a majority of studies conducted in the school setting (ranging from 55% to 95% of studies included). Six reviews (5%) had smaller subsets of studies conducted in the school setting (ranging from 6% to 38% of studies included) but reported subgroup analyses results specifically for the school setting. The number of school-based studies included in each review ranged from one to 124 studies, with an overall average of 23 school-based studies. Typically, reviews included between 10 and 19 school- based studies each (Fig. 2). Figure 2. Number of school-based studies included in individual reviewsa ª One review did not report the exact number of school-based studies. 0 5 10 15 20 25 30 35 40 19 39 22 11 11 14 <10 20-29 >4930-39 40-4910-19 45 Most reviews did not restrict studies included based upon the country of research.1 Yet, the vast majority of studies included in the reviews were based in high-income countries, particularly the United States and in Europe. Half of the reviews (58) only included studies from high-income countries. Not more than nine reviews only or largely included studies from low- and middle-income countries, while another four reviews focused on disadvantaged subpopulations in high-income countries. 1 For example, 10 reviews focused on school interventions in a single country (Canada, China or the United States). In the assessment of the modified AMSTAR score for review quality, 77 review papers were assessed as high quality, 34 as medium quality and six as low quality. Meta-analyses scored higher than systematic reviews, which in turn scored higher than other review types included (Fig. 3). No particular score distribution pattern emerged based on the geographic or thematic focus of the review papers. Thus, in the subsequent narrative review of evidence by essential criteria, findings of meta-analyses are presented separately from the narrative reviews. 10 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Figure 3. Assessed modified AMSTAR scores by review type Relevance of interventions and programmes to the NFSI essential criteria Most reviews considered studies at the primary (86% of reviews) and/or secondary (75% of reviews) school level, while fewer included studies in preschool establishments (24% of reviews). One hundred and fourteen reviews (97%) considered interventions and programmes related to diet and nutrition, the remaining three reviews included nutrition-relevant outcomes in their analyses of water, sanitation and hygiene (WASH) programmes. Forty-seven reviews (47%) considered multicomponent programmes combining diet and physical activity interventions. There was wide variation in the number of available published reviews for each component (Fig. 4). In terms of the number of relevant reviews, the majority of the evidence was available for the following criteria: having a written nutrition- friendly schools policy that addresses all five points described in the NFSI framework (essential criterion 1.1); activities for families and community, community involvement and outreach in the area of nutrition and health-related issues (2.2); culturally appropriate and effective nutrition education (3.1); and, school meals, food vendors and snack bars (if present) promoting healthy 33 41 3 10 20 2 0% 20% 40% 60% 80% 100% Meta analyses (n=43) High Medium Low 4 4 Systematic reviews (n=63) other reviews (n=11) eating (4.1). Limited evidence was available for criteria on rationale of the nutrition-friendly schools policy (1.2), objectives of the nutrition- friendly schools policy (1.3), school staff training in nutrition and health-related issues (2.3), age, sex and culturally appropriate physical education curriculum (3.2), healthy living and life-skills education curriculum (3.3), positive messaging towards nutrition and active living (4.2), absence of marketing of foods and beverages at school (4.3), access to safe drinking water (4.6), promotion of hygiene and safe sanitary behaviour (4.7), school staff as role-models in encouraging healthy eating and healthy lifestyle (4.11), regular monitoring of children’s growth and development (5.1), effective feedback systems for parents and children to report findings of regular monitoring (5.2) and supportive school health services, including a referral system (5.3). No evidence was available for criteria on having an action plan for the nutrition- friendly schools policy (1.4), having a monitoring and evaluation plan for the nutrition-friendly schools policy (1.5), dissemination of the NFSI policy (2.1), regular monitoring of school curriculum relevant to NFSI and evaluation of impact of the education against the objectives (3.4), access 11RESULTS to an adequate eating place (4.4), adequate school cooking facilities (4.5), availability of clean and separate toilets for boys and girls (4.8), opportunity for all age groups to access space and school sporting facilities for physical activity in and outside of curriculum (4.9) and affirmative action against bullying, stigmatization and discrimination (4.10). Figure 4. Number of included reviews by relevance to essential criteria 10 20 30 40 50 60 1.1 Written, comprehensive nutrition-friendly schools policy 1.2 Rationale for the policy 1.3 Objectives of the policy 1.4 Action plan for the policy 1.5 Monitoring and evaluation plan for the policy 2.1 Dissemination of the nutrition-friendly schools 2.2 Family and community involvement 2.3 School sta training in nutrition and health-related issues 3.1 Nutrition education 3.2 Physical education curriculum 3.3 Healthy living and life-skills education curriculum 3.4 Monitoring and evaluation of school curriculum relevant to NFSI 4.1 Healthy eating promoted in school meals, food vendors and snack bars 4.2 Positive messaging towards nutrition and active living 4.3 Absence of marketing of foods and beverages at school 4.4 Adequate eating place 4.5 Adequate school cooking facilities 4.6 Access to safe drinking water 4.7 Promotion of hygiene and safe sanitary behaviour 4.8 Clean and separate toilets, for boys and girls 4.9 Access to space and school sporting facilities for physical activity 4.10 Armative action against bullying, stigmatization, and discrimination 4.11 School sta as role-models for healthy eating and healthy lifestyle 5.1 Monitoring of children´s growth and development 5.2 Feedback systems for ndings of growth monitoring 5.3 School healthy services, including referral system 0 1 51 1 1 1 1 32 5 38 17 8 0 33 9 1 1 1 3 2 0 0 0 3 2 2 4 12 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Outcomes reviewed The included reviews provided an analysis of a wide range of school-based interventions. Results had a high level of heterogeneity, both in terms of the school-level exposures and the impact on school children. There was also wide heterogeneity between studies in terms of outcome measures. Sixty-eight reviews (58%) considered changes in body mass index (BMI) or other weight- related outcomes in relation to overweight (e.g. body fat percentage, skinfold thickness, waist circumference or prevalence of overweight or obesity), but only five reviews (4%) considered height as a distinct outcome of interest. Sixty reviews (51%) examined the effect of interventions in relation to dietary outcomes (e.g. fat intake, fruit and vegetable consumption), diet behaviour, knowledge or related determinants (e.g. breakfast skipping, purchase, attitudes, self-efficacy, food waste) or the food environment (e.g. availability, access). Eight reviews (7%) reported effectiveness in relation to biochemical measures (e.g. anaemia, blood lipids). A range of other outcomes were considered in the reviews. These related to other health measures (e.g. blood pressure, diarrhoea, infection, mortality), education (e.g. attendance, performance), educational resources (e.g. presentation of foods and diets in text books), family impact (e.g. follow-up action taken by parents) and implementation issues (e.g. compliance with regulations and standards). As explained above, only the nutrition-related outcomes were considered in this review. NARRATIVE SYNTHESIS AND SUMMARY OF THE SYNTHE- SIZED EVIDENCE Evidence from the included reviews was summarized according to its relevance to each of the NFSI components. NFSI Component 1: Having a nutrition-friendly schools policy The first component of the NFSI is having a nutrition-friendly schools policy. This includes five essential criteria related to the policy’s content, in terms of being comprehensive (addressing all other components), having a rationale (addressing the relevant context), having objectives (with timelines and clear milestones), being operationalised in an action plan and having a plan for monitoring and evaluation. Two types of evidence were identified and reviewed in relation to NFSI essential criterion 1.1: • the effectiveness of having nutrition-related school policies, and/or • comprehensive or multicomponent programmes, with or without an explicit underpinning policy. Some of the evidence touched upon process factors related to the development of school nutrition polices, but no review was identified that examined the effects of having school policy rationale, objectives, action or monitoring and evaluation plans. 13RESULTS NFSI Component 1: Having a nutrition-friendly schools policy • Twenty-three meta-analyses and 28 reviews were identified that examined various aspects of comprehensive school nutrition policies and programmes, i.e. NFSI essential criterion 1.1. • Among this evidence, one systematic review examined the effect of “having a policy” and “it being comprehensive” and found that school-based policies focusing on both the food and beverage and physical activity environments (versus targeting only one of these areas) consistently showed improvements in BMI. • Furthermore, one meta-analysis and nine systematic reviews examined school food and nutrition-related policies. These typically addressed the school food environment and achieved improvements in outcome measures related to availability, purchasing and/or consumption but less often for weight-related outcomes. • Another 22 meta-analyses and 18 systematic reviews examined the effect of multicomponent school nutrition interventions, whether underpinned by an explicit policy or not. Most of the reviews analysed the effects of different programme areas (usually diet and physical activity), while fewer analysed the effect of multiple approaches (i.e. curricula, environment, parental involvement). • Positive effects on weight-related outcomes, biochemical measures, such as blood pressure and blood lipids, and dietary outcomes such as intakes of SSBs or fruits and vegetables, have been shown in studies on comprehensive school nutrition policies and programmes, like the Health Promoting Schools initiative, using multiple approaches (i.e. curricula, environment, parental involvement) or having multiple components (e.g. diet, physical activity). • Seventeen of the 22 meta-analyses found that comprehensive school nutrition policies and programmes improved various outcomes — such as BMI and other weight-related outcomes, dietary outcomes, blood pressure and blood lipids — and usually more so than single component trials. Improvements in BMI standard mean difference (SMD) ranged from < 0.1kg/m2 to 1.8kg/m2 through school-based programmes. • Of all the evidence considered under essential criterion 1.1, four systematic reviews focused specifically on low- and middle-income countries or vulnerable populations in high-income countries. These reviews reported positive effects on weight-, diet- or health behaviour-related outcomes from multicomponent school nutrition policies and programmes in these populations. • One scoping review was identified that could shed some light on essential criteria 1.2 to 1.4. This review examined facilitators and barriers to implementing school nutrition policies. 14 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 15RESULTS @ Graham Crouch Essential Criterion 1.1: Development of a written nutrition-friendly schools policy that addresses all five components described in this framework, and includes elements described in 1.2 – 1.5 The first NFSI essential criterion concerns both having a written school nutrition policy and it being comprehensive in addressing all other essential criteria. One 2018 systematic review was identified that examined these aspects together (35). The remaining evidence focused on either having school food and nutrition-related policies (one meta-analysis (36) and nine systematic reviews (37-45)), or implementing comprehensive school food and nutrition-related policies, programmes or interventions (22 meta-analyses (46-67) and 18 systematic reviews (68-85)). Examining both aspects of “policies” and “comprehensiveness”, Bramante et al. (35) reviewed 33 natural experiments of national and sub-national policies for childhood obesity prevention and control, 24 of which were school-based. They found that school-based policies focusing on both the food/beverage and physical activity environments (versus targeting only one of these areas) consistently showed improvements in BMI, with four out of four such studies achieving favourable outcomes (versus eight out of 14 focusing on the food and beverage environment only). Additionally, the impact on BMI was assessed by the level of government implementing the policy. Among the 26 studies in the United States evaluating government policies, 15 (58%) reported favourable effects on BMI—six of seven (85%) federal policies, three of nine (33%) state/regional policies and six of 10 (60%) local policies. Regarding the effect of school-based nutrition interventions implemented through policies, one meta-analysis (36) and nine systematic reviews (37-45) were identified. Typically, these addressed the school food environment and achieved improvements in availability, purchasing and/ or consumption, but less often in weight-related outcomes. Most of these reviews did not examine comprehensive policies, but focused on single components that are addressed under other NFSI essential criteria and are therefore discussed in greater detail under those respective sections. For example, a school nutrition policy altering the food and beverage environment would be discussed under NFSI essential criterion 4.1. In their 2018 systematic review and meta- analysis of 91 studies on different types of school food environment policies, Micha et al. (36) found improved targeted dietary behaviours in children but inconclusive effect on adiposity and metabolic measures. Direct or indirect provision policies, which largely targeted fruit and vegetables, increased consumption of fruits (+0.27 servings/day; 95% CI: +0.17, +0.36), fruit and vegetables (+0.28 servings/day; 95% CI: +0.17, +0.40) and vegetables (+0.04 servings/day; 95% CI: +0.01, +0.08). Standards on foods and drinks sold outside school meal programmes (referred to as competitive foods and beverages) reduced total consumption of SSBs ( 0.18 servings/day; 95% CI: -0.31, -0.05) and unhealthy snacks (-0.17 servings/day; 95% CI: 0.22, 0.13), as well as in- school consumption of unhealthy snacks (-0.05 servings/day; 95% CI: 0.08, -0.02). School meal standards (mainly lunch) increased fruit intake (+0.76 servings/day; 95% CI: +0.37, +1.16), reduced intake of total fat (-1.49%E; 95% CI: -2.42, -0.57) and saturated fat ( 0.93%E; 95% CI: 1.15, -0.70), reduced sodium intake (-170 mg/day; 95% CI: -242, -98) and reduced BMI percentile (-1.01; 95% CI: -1.62, -0.39). Positive effects were also seen on increased milk consumption and carbohydrate intake in studies of policies targeting these elements. In a 2009 systematic review of 27 studies of school food and nutrition policies worldwide targeting the food environment, for example, 16 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE through legislating nutrition guidelines, regulating food and beverage availability or intervening on prices, Jaime and Lock (37) found that some policies may improve the food environment and dietary intake but found no evidence for impact on BMI. Similarly, in 2014 Driessen et al. (38) reviewed 18 studies of changes to the school food environment, 11 of which were implemented through policy, and found that all but one study reported positive outcomes. Among the policy studies, positive outcomes were reported by eight of nine studies measuring purchasing and/or consumption and by two of four studies measuring weight-related outcomes. Mayne, Auchincloss and Michael in 2015 (39) reviewed 37 studies evaluating the efficacy of policy and built environment changes on obesity-related outcomes, five of which were school nutrition- related policies at the district, municipal or federal level to restrict sugary foods and beverages or higher fat foods, and/or increase the availability of milk or fruit and vegetables. Four of these five policy studies reported results in the expected directions (e.g. reducing consumption of energy- dense nutrient-poor food, SSBs, calories or fat, or increasing fruit and vegetable intake) while one study had mixed results. One of the five studies measuring BMI reported mixed results. In 2015, Afshin et al. (40) reviewed the evidence for specific policies to improve dietary habits and reduce cardiovascular and metabolic risk factors, including 76 studies of school procurement policies. Among 26 studies of the effect of restricting unhealthy foods and beverages in schools (e.g. restrictions or bans, nutrient standards for competitive foods in cafeterias, vending machines, school stores, snack bars or snack trucks), interventions appeared generally effective in reducing intake of unhealthy 17RESULTS “Positive effects have been shown in studies on comprehensive school nutrition policies and programmes, like the Health Promoting Schools initiative, using multiple approaches (i.e. curricula, environment, parental)” Th e H un ge r P ro je ct competitive foods and beverages and decreasing incidence and prevalence of overweight/obesity, with governmental and school policies being more effective than voluntary programmes. Chriqui, Pickel and Story in 2014 (41) investigated the effects of state and district competitive food and beverage laws and policies in the school setting in the United States. Overall, 15 of the 24 studies reviewed reported results in the expected direction, especially for outcomes related to availability and access, but less often for outcomes related to purchasing, consumption, dietary intake and/or weight-related outcomes. In another 2018 systematic review of 26 studies on competitive foods and beverage in schools in the United States — 14 of which specifically analysed policies implemented in schools — Sildén (42) concluded that there was a positive association between competitive food and beverages policies and weight-related and/or dietary outcomes, and provided examples where both caloric intake and overweight or obesity had been improved. Also in the United States, Mansfield and Savaiano (43), in 2017, reviewed 31 studies of the U.S. National School Lunch Program nutrition standards resulting from three sequential federal policies. They note that improved food consumption behaviours (increased selection, intake and sales of healthy foods, and decreased plate waste) were reported in 14 of the 19 intervention and longitudinal observation studies. On the other hand, only two of 12 one-time observation studies reported food consumption behaviours meeting target nutrition standards. In 2019 Von Philipsborn et al. (44) published a Cochrane review of 58 studies aiming to reduce consumption of SSBs, 20 of which were conducted in elementary, middle and high school. In a subanalysis of seven studies of school nutrition standards, five reported reduced SSB intake in children as result of lower availability. In three of these studies, the effect was observed after implementing a policy-level measure. Likewise, in their 2017 systematic review of 36 school-based interventions aimed at reducing SSB consumption among adolescents, Vézina-Im et al. (45) noted that the 10 studies involving legislative measures or structural changes to the environment (e.g. banning SSBs from school cafeterias or replacing them with healthier alternatives) were the most effective. Regarding the effect of comprehensive school- based nutrition interventions implemented — whether underpinned by an explicit policy or not — a larger evidence base was identified, notably 22 meta-analyses (46-67) and 18 systematic reviews (68-85). These reviews examined multicomponent interventions combining multiple approaches as described in the FRESH framework and the Health Promoting School approach (e.g. curricula, environment, parental involvement) and/or addressed multiple programme areas (e.g. diet, physical activity, water). Most of the reviews analysed the effects of different programme areas (usually diet and physical activity) in multicomponent interventions where the majority of studies focused on educational approaches. Of the 22 meta-analyses of multicomponent school nutrition policies and programmes, 17 found that these improved @ In gi m ag e 18 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE BMI and other weight-related outcomes, dietary outcomes, blood pressure and blood lipids in at least one age group of children within schools, and often more so than single component interventions (46-62). Thirteen of these meta- analyses measured weight-related outcomes and demonstrated improvements in BMI SMD ranging from < -0.1kg/m2 to -1.8kg/m2 through school-based programmes. Four meta-analyses did not find any effect of multicomponent school-based nutrition policies and programmes (63-66), whereas one meta-analysis was unable to pool results, but concluded positively regarding multicomponent programmes (67). In a 2014 Cochrane review of the effectiveness of the WHO Health Promoting Schools approach, Langford et al. (46) concluded that holistic school-based programmes aligning with (but not necessarily making explicit reference to) all three Health Promoting Schools framework strategies (i.e. curriculum, environment and families and/or communities) can be effective at improving a number of health outcomes in students. The review included 67 RCT studies, of which 34 concerned the effect of diet and/or physical activity interventions on nutrition-related outcomes. Meta-analysis on these 34 multi- approached studies demonstrated significant improvements in BMI from three physical activity only studies (MD: -0.38 kg/m²; 95% CI: -0.73, -0.03), in BMI z-score from one physical activity only study (MD: -0.47; 95% CI: -0.69, -0.25), and in fruit and vegetable intake from nine nutrition only studies (SMD: +0.15; 95% CI: +0.02, +0.29). However, no significant improvements were found for combined nutrition and physical activity interventions or on other nutrition-related outcomes (e.g. fat intake). In another 2011 Cochrane review of 55 long- and short-term childhood obesity prevention programme studies by Waters et al. (47), 43 included an educational setting, and this was also identified as the most effective setting. Interventions were mostly delivered through educational approaches, but some also addressed the environment and/or employed school health nurses. It included a meta-analysis of 37 studies covering 27 946 children, in which the greatest overall effect on BMI or BMI z-score was found for combined physical activity and dietary interventions (SMD: -0.18; 95% CI: -0.27, -0.09) versus physical activity alone (SMD: – 0.11; 95% CI: -0.19, -0.02), while there was no significant effect of diet alone. Updating the paper by Waters et al (47), Peirson et al. in 2015 (48) reviewed 90 studies, 71 of which were school-based and largely focused on, but not restricted to, behavioural interventions. Like Waters et al. in 2011 (47), they found that schools were the most effective setting, showing significant reductions in BMI or BMI z-score (SMD: -0.09; 95% CI: -0.17, -0.03). In the overall meta- analysis of 76 studies involving 56 342 children, small but significant reductions were observed in BMI or BMI z-score (SMD: -0.07; 95% CI: -0.10, -0.03), in BMI (MD: -0.09 kg/m2; 95% CI: -0.16, -0.03), and in the prevalence of overweight and obesity (RR: 0.94; 95% CI: 0.89, 0.99). Significant reductions in BMI or BMI z-score were found for diet and exercise combined (SMD: -0.10; 95% CI: -0.17, -0.03), but not for either component alone. In a subsequent update of the original analysis by Waters et al. (47), Gori et al. (49) in 2017 did a review and meta-analysis of 72 studies in overweight or obese primary school age children, 47 of which were school-based. Interventions typically took an educational approach, though some studies also changed the environment, engaged school health nurses and/or involved parents. In this update significant results were found for combined diet and physical activity interventions in children 6-12 years (mean BMI 19RESULTS standard deviation score (BMI-SDS): -0.13; 95% CI: -0.19, -0.06) as well as physical activity alone (mean BMI-SDS: -0.11; 95% CI: 0.16, -0.06, 14 studies). However, the largest effect on BMI-SDS was from combined school-based diet and physical activity interventions with a parental involvement component (mean BMI-SDS: -0.15; 95% CI: -0.21, -0.03). No other significant results were found for any intervention or setting combination or other age groups. In a meta-analysis of 52 school-based and after-school diet, physical activity and/or lifestyle interventions delivered through educational or other approaches involving 28 236 children 5–18 years of age, Vasques et al. in 2014 (50) found that the combined interventions were more effective than single interventions in reducing children’s BMI. Those combining diet and physical activity were the most effective with a medium effect size (r = 0.148) compared to small effect sizes for any other intervention type or combination. In a 2012 meta-analysis of 68 single- and multicomponent obesity prevention studies among children and adults, Luckner, Moss and Gericke (51) found significant reduction in BMI among children and adolescents in six school-based studies combining educational activities and non-educational activities to increase physical activity and improve nutrition (MD: -0.1 kg/m2; 95% CI: -0.17, -0.04). However, larger, significant BMI effect sizes were seen for reduced TV viewing and for physical activity. Other intervention combinations had too high heterogeneity or showed no statistically significant difference in outcomes. In a 2013 meta-analysis of 32 school-based obesity prevention programme RCT studies involving 52 109 children 5-18 years of age, Sobol- Goldberg, Rabinowitz and Gross (52) found an overall significant reduction in BMI (SMD: -0.076; 95% CI: -0.123, -0.028). Meta-regression analyses showed a significant linear hierarchy of studies based on their design. The largest effects were from six comprehensive multicomponent and multiapproached programmes of a duration of 1 year or longer and including education on nutrition and physical activity, modification to the environment and parental involvement (SMD: -0.151; 95% CI: -0.334, -0.031). In 2012 Friedrich, Schuch and Wagner (53) reported a meta-analysis of 23 school-based studies. All studies included an educational approach and seven studies also had an environmental component. The authors found that combined physical activity and nutrition education significantly improved BMI outcomes (SMD: -0.37; 95% CI: -0.63, -0.12), whereas neither component alone gave any significant result. Katz et al. in 2008 (54) reviewed 19 school-based obesity prevention studies delivered through multiple approaches (e.g. information sessions, practical hands-on skill building, teacher training, school cafeteria modifications) and with a minimum duration of 6 months. Results from the meta- analysis of eight studies of combined nutrition and physical activity interventions involving 10 752 children 3-18 years of age showed significant reduction of body weight in children (SMD: -0.29; 95% CI: -0.45, -0.14). This was similar to the effect of a single nutrition intervention (SMD: -0.39; 95% CI: -0.56, -0.23) and a single TV reduction intervention (SMD: -0.35; 95% CI: -0.63, -0.06), whereas other single component interventions were not found to be effective. In a 2012 meta-analysis of 43 studies on school-based interventions covering 36 579 children 18 years or younger, Lavelle, Mackay and Pell (55) found that interventions combining physical activity with nutrition were associated with significant improvements in BMI (MD: -0.17kg/m2; 95% CI: -0.29, -0.06). The majority of studies applied educational approaches and only a few modified the environment. Overall, interventions targeted at overweight or obese children were more effective in reducing BMI than those delivered to all children. Similarly, in China in 2017, Feng et al. (56) reviewed 76 school-based obesity intervention studies 20 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE covering 72 620 children 6-19 years of age. The majority of the interventions were multicomponent applying multiple approaches such as education, weight monitoring and counselling and/or school policies. They found that the comprehensive interventions involving physical activity and health education had the largest effect on BMI in both treatment studies (SMD: -1.80; 95% CI: -2.15, -1.44) and prevention studies (SMD: 0.19; 95% CI: -0.27, -0.11). Among the prevention studies targeting all children, 19 of 27 multicomponent interventions (70.4%) were effective versus two out of nine single component interventions (22.2%). In 2019, Brown et al. (57) conducted a Cochrane systematic review and meta-analysis of 153 obesity prevention studies applying various approaches (e.g. educational, environmental, policy, parental involvement), of which 91 were school-based and 22 took place in kindergartens or preschools. In children aged 6-12 years in the school setting, significant effect was found on BMI z-score from combined diet and physical activity interventions (-0.04; 95% CI: -0.08, -0.01). In children of 13-18 years of age, significant effect was found on BMI z-score and BMI from physical activity alone. No effect was found from any combination of interventions in children under five years of age. They also pooled results from studies that compared different interventions against each other (e.g. diet alone vs combined diet/ physical activity) and all such analyses did not find differences in impact. In the United States, Wang et al. (58) in 2013 reviewed 124 obesity prevention interventions in multiple settings, 104 of which were conducted in schools. Over half of the school-based interventions reported statistically significant beneficial intervention effects in at least one weight related outcome (BMI, BMI z-score, prevalence of overweight and obesity, waist circumference, skinfold thickness and percent body fat). High strength of evidence was found for combined diet and physical activity interventions with a home and community component and moderate strength of evidence for combined interventions with either a home or a community component. Comprehensive interventions addressing the environment (e.g. modified food and beverage items offered in school cafeteria, or structural changes in school physical activity) and individuals’ knowledge and attitude 21RESULTS Th e H un ge r P ro je ct 22 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE David Snyder for The Hunger Project (e.g. nutrition education) were more likely to be successful than those addressing either one alone. Meta-analyses of school-based studies combining diet and physical activity found improved BMI z-score (-0.08; 95% CI: -0.14, -0.02, P = 0.009) from four studies and improved BMI (-0.32 kg/m2; 95% CI: -0.49, -0.16, P < 0.001) from seven studies. Beneficial effects of school-based interventions were also observed for outcomes other than BMI or weight. Evans et al. (59) reported a 2012 systematic review and meta-analysis of 27 studies to increase fruit and vegetable intake using multiple approaches such as nutrition education, parental involvement, fruit and vegetable provision, food preparation and/ or tasting, where the majority of studies were multicomponent programmes. In a meta-analysis of 21 of the 27 studies they found that total fruit and vegetable intake increased (+0.25 portions/ day; 95% CI: +0.06, +0.43), mainly due to increases in fruit consumption. The authors noted that multicomponent programmes tended to result in larger improvements while single component interventions, including free and subsidized fruit and vegetable distribution schemes, tended to be less effective although there were too few studies included to make firm conclusions. In another meta-analysis of school-based interventions to promote fruit and vegetable consumption, Howerton et al. (60), in 2007, reviewed seven studies on classroom-based activities coupled with other strategies, such as family involvement or food service changes, and involving 8 156 children aged 7–12 years of age. They found increased fruit and vegetable consumption in the intervention group compared to the control group (+0.45 servings; 95% CI: +0.33, +0.59). In a 2014 meta-analysis of 23 studies of child obesity prevention programmes, Cai et al. (61) found that combined diet and physical activity interventions mainly delivered through educational approaches, although in some instances additionally by modifying the environment, most of which were implemented in the school setting, led to a significant reduction in both systolic blood pressure (SBP) (WMD -2.11 mmHg, 95% CI: -3.19, -1.03) and diastolic blood pressure (DBP) (WMD: −1.51 mmHg; 95% CI: 2.55, -0.47), but not for either component alone. In a similar meta-analysis of 17 child obesity prevention studies focusing on blood lipids, Cai et al. (62), found in 2014 overall significant improvements in low-density lipoprotein-cholesterol (LDL-C) and high-density lipoprotein-cholesterol (HDL-C), but not total cholesterol and triglyceride. The programmes that combined diet and physical activity led to larger but non-significant improvements in HDL-C and total cholesterol than the single component programmes, and to significant improvements in LDL-C (WMD: 8.49 mg/dL; 95% CI: -12.77, -4.21) which was larger than the programmes focusing on physical activity alone (WMD: -2.67 mg/dL; 95% CI: -4.30, -1.03) but lower than the programmes focusing on diet alone (WMD: −17.59 mg/dL; 95% CI: −25.58, −9.60). In contrast, four meta-analyses did not find positive effects of school-based multicomponent nutrition policies or programmes (63-66). Two of these were 2017 reviews of obesity treatment studies using behavioural change interventions addressing diet and physical activity (63, 64). Overall positive effects in analyses of all settings were found in both reviews, but not in subgroup analyses of the school setting where studies were low in numbers. Mead et al. (63) found overall positive effect on BMI, BMI z-score and weight change when pooling the results from 70 studies in any setting, but in subgroup analyses based on setting, the two school-based studies included in the meta analyses did not show significant effect on weight, BMI or BMI z-score. Likewise, Al-Khudairy et al. (64) found significant positive effect on BMI, BMI z-score and weight change among overweight and obese adolescents aged 23RESULTS Finally, one 2013 meta-analysis of 21 studies of school-based diet and physical activity interventions by Williams et al. was unable to pool results from the six programmes that combined diet and physical activity programmes due to high heterogeneity (67). However, the authors noted that these programmes demonstrated promising results, particularly those which involved stakeholder involvement and engagement with families. In addition to these meta-analyses, 18 systematic reviews were identified as relevant to essential criterion 1.1. Of these reviews, 15 identified benefits of multicomponent programmes on weight-related and/or dietary outcomes (68-82), while three reviews did not find improved outcomes as a result of multicomponent nutrition policies or programmes (83-85). Four of the reviews considered disadvantaged populations (69-71, 76, 82). Four of the 15 reviews identifying benefits of multicomponent programmes, considered both multiple approaches and programme 12-17 years when pooling results from 44 studies in any setting. However, the effects on BMI in seven school-based studies and on BMI z-score in two school-based studies were positive but not significant. Two other meta-analyses found larger effects of single component programmes compared to multicomponent programmes. Hung et al. (65) in 2015 did not find strong evidence of school- based childhood obesity prevention programmes in a meta-analysis of 27 studies. Moreover, single component programmes resulted in larger — although small — effect than multicomponent programmes. Likewise, in a meta-analysis of 38 studies on school-based combined physical activity and nutritional education interventions, Guerra et al. (66) in 2014 did not find significant reductions in BMI, although nutrition education alone was found to be effective. @ In gi m ag e 24 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE components. De Bourdeaudhuij et al. (68) in 2011 conducted a systematic review of 11 school-based obesity prevention interventions and identified five studies on promotion of physical activity and nutrition to adolescents through schools. They found moderate evidence that multicomponent interventions focusing on healthy diets and physical activity habits, and combining an educational with an environmental component, had a positive impact upon weight-related measures in adolescent girls. These programmes were more effective than focusing on education alone. In primary-school-age children, only one of four multicomponent interventions improved anthropometric outcomes, but they still showed more favourable results than those only using education. In low- and middle-income countries, Verstraeten et al. (69) conducted a review in 2012 of 15 studies targeting physical activity and/or dietary behaviour for the primary prevention of obesity in children and adolescents aged 6–18 years. While neither of the two dietary behaviour interventions had a significant effect on BMI, four out of the five physical activity intervention studies did. However, the largest effect size on BMI came from two high-quality studies of combined diet and physical activity interventions using integrated curriculum delivered by teachers and accompanied by nutrition education of parents. In a 2019 scoping review of 15 preschool interventions in low socioeconomic settings in high-income countries, Luybli, Schmillen and Sotos-Prieto (70) found that both successful and inconclusive studies included multiple components or approaches, such as nutrition, physical activity and parental involvement. The authors concluded that obesity outcomes can be targeted in low socioeconomic settings through school interventions with a multicomponent approach, but that further research is needed regarding the appropriate interventions. In elementary schools in the United States with large minority populations, Johnson, Weed and Touger-Decker (71) found in 2012 that all seven multicomponent programme studies included reported some benefits in health behaviours and/or anthropometric measures, five of which reported a positive benefit on overweight and/or adiposity. The authors noted that effectiveness was greater when programme objectives were specific, implemented across the school environments, extended into the community and were culturally relevant. Another eight of the 15 reviews identifying benefits of multicomponent programmes, considered studies employing multiple approaches. In a 2013 systematic review of 19 studies of nutrition promotion programmes using a full or partial WHO Health Promoting Schools approach, Wang and Stewart (72) found that such approaches may improve children’s diet and nutrition knowledge, attitude and skills, and behaviour. Improvements observed included: increased consumption of high-fibre foods, healthier snacks, water, milk, fruit and vegetables; reduced breakfast skipping; and reduced consumption of nutrient-poor foods including high-fat foods and sweet drinks. In another systematic review of 17 school-based childhood obesity prevention studies, Brown et al. (73) found in 2016 that multifaceted programmes lasting between 6 and 12 months and involving multiple environmental, educational and physical strategies were most likely to improve BMI and BMI z-score. Similarly, in 2016, Meiklejohn, Ryan and Palermo (74) reviewed 11 studies on school-based nutrition education coupled with complementary strategies such as parental involvement or school food policies and found significant improvements in BMI or dietary intake in 10 of these studies. In a 2012 series of systematic reviews on various aspects of nutrition education in countries with a high or very high human development index, the United States Department of Agriculture (USDA) (75) conducted a systematic review of 25RESULTS five school-based studies combining nutrition education with changes to the school food environment. They found that combining changes to the school food environment with nutrition education was effective for improving children and adolescents’ dietary intake (e.g. fruit and vegetable intake, fish consumption). In two studies the combined approach was more effective, in one study it was more effective in boys only and in two studies both approaches were effective. In 2018 Colley et al. (76) reviewed nine multicomponent school food programme studies in Canada, many of which targeted disadvantaged children. All studies included school food provision, and eight of them also included other interventions such as policy, education, family and community involvement. They found that these programmes were positively associated with children’s development of nutrition knowledge in four out of six studies which measured it, in dietary behaviour and food preferences in all seven studies measuring this and in intake of healthy foods in six out of eight studies which assessed intakes. De Sa and Lock (77) reported a 2008 systematic review of 30 studies of school-based interventions to promote fruit and vegetable intake using multiple approaches — such as free or subsidized fruit and vegetables, classroom activities, school policy, parental involvement and/or modification to the school food service — and where the majority of studies were multicomponent. Overall, 22 of the 30 studies (> 70%) increased fruit and vegetable intake. In another systematic review of 14 studies to increase fruit and vegetable consumption, Aloia et al. (78) in 2016 found that 8 out of the 13 multicomponent studies that included parental involvement, teacher involvement, nutrition education, food service and school gardens showed significant increases in consumption of fruit and vegetables among school children. In 2006, Knai et al. (79) also reviewed interventions to promote fruit and vegetable consumption in children, using multiple approaches such as classroom- and non-classroom-based education, changes to the environment, parental involvement and/or training for staff. Of the 15 studies reviewed, 14 were conducted in the school setting and 10 had a significant positive effect on fruit and vegetable consumption. The authors noted that the strongest evidence was found for interventions combining multiple approaches. Finally, three of the 15 reviews identifying benefits of multicomponent programmes, considered studies addressing multiple programme areas. In a 2018 systematic review of 56 studies in various settings — 47 of which were conducted in schools, preschools or after-school — Bleich et al. (80) found that all 17 of the 24 school-based RCTs that reported positive results in weight-related outcomes (BMI, BMI z-score, BMI percentile, body fat percentage, skinfold thickness, waist circumference or prevalence of overweight or obesity) used combined diet and physical activity interventions. Brown and Summerbell in 2009 (81) investigated the effectiveness of 38 school-based obesity prevention interventions and found that nine of 20 (45%) studies on combined diet and physical activity interventions significantly improved BMI, which was better than for either component alone. In ethnic minority school children aged 10-14 years, in 2010 Stevens et al. (82) reviewed eight studies of diet and physical activity obesity prevention programmes and found improvements in various dietary outcomes including decreased consumption of high-fat and other high-calorie foods and soft drinks, as well as increased consumption of fruits and vegetables. On the other hand, three systematic reviews did not find multicomponent 26 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE school-based programmes addressing multiple programme areas to be more effective than single component programmes. In a 2006 systematic review of 25 largely school-based obesity prevention studies focusing on diet and/ or physical activity, Doak et al. (83) compared features of 17 studies that were effective in improving outcomes related to height/weight and/or skinfolds against features of the eight non- effective interventions. They found that a slightly smaller proportion of the effective interventions vs. the non-effective interventions intervened on both diet and activity (82% of studies vs. 88% of studies), included activities outside school as part of the intervention (67% vs. 88%), targeted the physical environment (24% vs. 25%), required active participation of children (82% vs. 100%) and actively involved parents (47% vs. 57%) or the broader community (6% vs. 14%). The effective interventions were, however, more often considered sustainable in terms of being accepted by the school community or financially viable (94% vs. 88%). Likewise, in 2007 Lissau (84) reviewed 14 school-based obesity prevention studies in the school arena and found that half of the studies had an effect on either overweight or obesity. However, only two of the six multicomponent studies that included nutrition/diet were effective. In a 2008 systematic review of 14 school-based obesity prevention programmes of at least 6 months duration, Kropski, Keckley and Jensen (85) did not find that combined diet and physical activity interventions were more effective than interventions with either component alone. Significant reduction in overweight was reported in at least one subgroup in five of 11 combined studies and one of two physical activity studies, whereas the sole nutrition only study reported reduction in overweight but without indicating significance. Essential Criteria 1.2 Rationale, 1.3 Objectives, 1.4 Action plan, and 1.5 Monitoring and evaluation plan No systematic reviews were identified that specifically examined the effects of having school policy rationale, objectives, action or monitoring and evaluation plans. However, McIsaac et al. (86) conducted a 2019 scoping review of 59 studies and identified facilitators and barriers to the implementation of school nutrition policies and derived a set of opportunities for action to make such policies more successful. Some of these factors are relevant to the elements described in essential criteria 1.2-1.5. According to the review, the opportunities for action were, first, to provide macro-level support to encourage policy implementation. This was analysed in 41 studies in terms of clarity of policy execution and policy language, training support and resources for school staff, monitoring and enforcement of the policies and communication between schools, policy-makers and other stakeholders. The second opportunity identified was to address the financial implications of healthy food access, and this was included in 39 studies in terms of access to and procurement of fruit and vegetables and other nutritious options that often have higher costs, but generate more revenue and higher profit. This can lead to trade-offs between keeping costs low for students and families and generating sufficient income from cafeterias and canteens, while also eliminating historic practices in schools of using unhealthy foods to raise funds. Third, aligning nutrition and core school priorities was identified as an opportunity for action; this was examined in 20 studies in terms of competing priorities and demands on, for example, academic assessment that make it difficult for schools to implement nutrition-related changes and the time-consuming nature of policy implementation and partner involvement. Fourth, developing a common purpose and responsibility among stakeholders, which was included in 45 studies in terms of 27RESULTS school climate, values, opinions and cultural norms related to food within schools, at home or in the communities. These cultural norms included the perceived student preference for less healthy foods, the school capacity for leadership and coordination and engagement of parents. A fifth area of opportunity was recognition of school and community characteristics, which was examined in 33 studies in terms of contextualising the policies to school location and demographics, addressing issues of poverty and food insecurity that may lead to inequity, school infrastructure (such as cafeterias or gardens), wider infrastructure and the access to competitive foods from the school surroundings or at home. NFSI component 2: Enhancing awareness and capacity building of the school community The NFSI component 2 entails awareness-raising and capacity-building. The essential criteria cover the dissemination of the nutrition-friendly schools policy, ensuring the active involvement of families and communities in promoting healthy diets and training school health staff on nutrition and health-related issues. Numerous reviews were identified examining the involvement of parents in school nutrition programmes and activities, there were also some that considered staff training related to school nutrition, while no evidence was identified for dissemination of the school nutrition policy. @ Ingimage 28 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE NFSI component 2: Enhancing awareness and capacity building of the school community • Fourteen meta-analyses and 18 reviews were identified relevant to essential criterion 2.2, examining family and community involvement in school nutrition programmes. • All five meta-analyses that compared studies with or without parental involvement found better results in BMI and blood pressure among the studies with parental involvement. Parental involvement was identified as a moderating factor leading to greater success in three of these meta-analyses. • Five of the eight meta-analyses that examined studies or sub-sets of studies with a parental involvement component, but without comparison to no parental involvement, found significant results in BMI and other weight-related outcomes, as well as fruit and vegetable consumption. • Eight out of 11 reviews analysing the effect of parental involvement pointed to family or community involvement as a factor leading to positive outcomes in terms of weight-related outcomes, fruit and vegetable consumption or NCD risk factors. • Direct parental involvement methods where parents are actively participating (e.g. attendance of educational or counselling sessions) more often led to changes in dietary outcomes than indirect methods where parents are passive recipients of information (e.g. newsletters, tip sheets, invitations to attend events). • Parental involvement was identified as particularly important for the younger age groups of children. • Only one scoping review considered under criterion 2.2 focused on disadvantaged populations in higher-income countries, which noted that parental involvement is important in preschool multicomponent interventions in low socioeconomic settings. • Five systematic reviews were identified relevant to essential criterion 2.3, which examined aspects of school staff training on nutrition or recognized it as a success factor. For example, studies that trained existing teachers or engaged nutrition and physical activity experts in the implementation were more likely to succeed. • Two Cochrane reviews found insufficient evidence to draw any firm conclusions regarding effective strategies to improve implementation or staff knowledge, attitudes and skill, although significant improvements frequently were reported in primary studies. • Working with teachers to develop programmes and providing ongoing training, support and communication can have moderate positive results in terms of reducing BMI, enhancing physical activity and improving fruit and vegetable consumption. • A body of research on school-based nutrition interventions has included staff training as part of its work, but additional analyses are needed to unravel the effects of this component from other components of the intervention. 29RESULTS Essential criterion 2.1: Dissemination of the Nutrition- Friendly Schools Policy No specific evidence was identified for essential criterion 2.1. The dissemination of school policies was not disentangled from other programme elements in the synthesized evidence on school policies reviewed under NFSI component 1. Essential criterion 2.2: Activities for families and community, community involvement and outreach in the area of nutrition and health-related issues Both the FRESH Framework and the WHO Health Promoting Schools approach recommend that health and well-being are integrated into every aspect of school life, which is why families and communities should be included in school- based health promotion. They may play an important role in prioritizing and implementing various components of the intervention and ensure that health promotion activities extend beyond the school gates and into other areas of a child’s life. Parents and caregivers are critical in ensuring that improved behaviours and practices resulting from school-based interventions are sustained and not hindered at household level. Fourteen meta-analyses (46, 49, 50, 52, 54, 58, 67, 87-93) and 18 reviews (70, 74, 78-80, 83, 94-105). were identified that examined the involvement of parents in school-based nutrition programmes. Among the 14 meta-analyses, five compared studies with or without such involvement (50, 52, 87-89), eight investigated studies with a parent or community involvement component included in all or a subset of studies (46, 49, 54, 58, 90-93), while one was unable to pool results for this aspect specifically (67). Two of these meta-analyses had parental or community involvement as a main objective (87, 90) and three compared the effect of different levels of involvement (50, 87, 90). All five meta-analyses that compared studies with or without parental involvement found better results in BMI and blood pressure among the studies with parental involvement. Niemeier, Hektner and Enger (87) conducted a 2012 meta- analysis of 36 RCTs in multiple settings to compare the effectiveness of nutrition and/or physical activity interventions with and without parent participation (e.g. special lessons, complementary sessions, handbooks, counselling or meetings for parents). Among the 17 preventive studies, which were largely school-based, parent participation significantly contributed to prevention of BMI increase compared to those with no such component (weighted average differences in effect sizes 0.30, P = 0.027). Moreover, significant differences existed among levels of parent participation (P < 0.05) and intervention duration (P = 0.006), and the linear combination of parent participation and intervention duration significantly predicted intervention effectiveness (P = 0.001). In another meta-analysis of 11 school- based obesity interventions involving school health nurses, Schroeder, Travers and Smaldone (88) in 2016 found in a subgroup analysis of six studies that, while both were effective, the three studies with parental involvement (e.g. participating in counselling, consultations or education) resulted in larger decreases in mean BMI (-0.72 kg/m2; 95% CI: -1.42, -0.01) than the three studies without this component (-0.06 kg/m2; 95% CI: -0.11,-0.01). Parental involvement was identified as a moderating factor leading to greater success in three of these five meta-analyses. Oosterhoff, Joore and Ferreira (89) conducted a 2016 meta- analysis of 85 RCTs on interventions to improve healthy eating and/or physical activity among children aged 4–11 years, and 53 of these had a parental involvement component (newsletters, information sessions and/or participation in homework tasks for children). The meta-analysis 30 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE found significant beneficial pooled effect sizes in BMI (-0.072; 95% CI: -0.106, -0.038), SBP (-0.183; 95% CI: -0.288; -0.078) and DBP (-0.071; 95% CI: -0.185, 0.044). Using a magnitude of SDs around the BMI mean values (on average 3.1 kg/m2), the authors noted that the reduction in BMI translated to approximately -0.22 kg/m2. Multivariable meta- regression analyses of moderators showed that parental involvement significantly improved the beneficial effects on BMI and SBP, but for BMI these only explained a third of the variation and the authors call for further comprehensive analyses of moderating factors. In the 2014 meta-analysis of 52 school-based and after-school studies involving 28 236 children and adolescents described under essential criterion 1.1, Vasques et al. (50) found that any level of parental involvement had significant and positive effects, with an increasing effect size for increasing level of parental involvement: no involvement (r = 0.047), minimal (r = 0.057), moderate (r = 0.082) and high (r = 0.094), with significant differences between groups (P = 0.001). In the 2013 meta-analysis of 32 school-based obesity prevention interventions covering 52 109 children aged 5–18 years also described under essential criterion 1.1, Sobol-Goldberg, Rabinowitz and Gross (52) found significant reduction of BMI of all age groups (SMD: -0.076; 95% CI: -0.123, -0.028) and in children alone (SMD: -0.104; 95% CI: -0.195, -0.01) albeit not in adolescents. A meta- regression analysis based on key characteristics of intervention programmes showed the largest effects for comprehensive programmes with parental support. Five of the eight meta-analyses that examined effects in studies or sub-sets of studies with a parental involvement component without @ In gi m ag e 31RESULTS comparison to studies that did not involve parents, found significant results in BMI and other weight-related outcomes, as well as fruit and vegetable consumption. In a meta-analysis of 16 obesity prevention studies with different levels of community engagement among school-aged children in the United States, Krishnaswami et al. (90) found in 2012 that overall community engagement performance scores were positively correlated with diet, physical activity and weight- related outcomes (r: 0.66; 95% CI: 0.25, 0.88). The two studies with the highest community engagement performance scores achieved the highest outcome performance. In another meta- analysis Nathan et al. (91), in 2019, examined 10 school- or centre-based studies to improve the healthiness of foods provided from home and found a moderate increase in provision of vegetables (SMD: 0.40; 95% CI: 0.16, 0.64, P = 0.001), but not fruit, from four of the 10 studies. Of the studies not included in the meta-analysis, mixed results were observed for foods and beverages brought from home, though two out of three studies reported significant effect on consumption of fruits and vegetables. While all included studies in the review incorporated a parent component, many simply relied on passive information dissemination strategies (e.g. pamphlets, newsletters). The other three meta-analyses were described in essential criterion 1.1 (46, 49, 54). The 2014 meta-analysis by Langford et al. (46) of studies of interventions following the WHO Health Promoting Schools framework (i.e. including a parent/family involvement component) found positive intervention effects in terms of reducing BMI, enhancing physical activity and improving fruit and vegetable consumption. In the 2017 meta-analysis of 72 studies among overweight or obese primary school age children, Gori et al. (49) found that best results were achieved for combined diet and physical activity interventions in children, 6-12 years, in a combined school and family setting (mean BMI-SDS: -0.15; 95% CI: -0.21, -0.03). In the 2008 review of 19 school-based obesity prevention studies in children aged 3–18 years focusing on diet and/or physical activity, Katz et al. (54) did a meta- analysis of four studies which included parent or family member participation and found improved body weight (SMD: -0.20; 95% CI: -0.41,0.00; P = 0.05). @ Ingimage 32 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE On the other hand, three meta-analyses found positive but insignificant effects of studies with a parental involvement component. In a 2015 systematic review and meta-analysis of 49 school-based studies of children aged 5–12 years (n = 38 001), Dudley, Cotton and Peralta (92) examined different teaching strategy approaches, such as enhanced curriculum, cross-curricula, parental involvement, experimental, contingent reinforcement, literary abstraction and games- or web-based approaches. They found that most of the teaching strategies led to positive changes in healthy eating behaviours (e.g. fruit and vegetable consumption, reduced sugar consumption), but in the pooled results, fewer strategies had mean effect sizes over 0.4 (Md > 0.4), above which was considered the cut-off for positive effect in this review. The 10 studies which included a parental involvement strategy generally led to good results in individual studies (91% positive outcomes), but the pooled mean effect size (Md = 0.39) was just below the cut-off for significant positive effect. Similarly, Wang et al. (93) conducted a systematic review and meta-analysis in 2015 of 139 studies of obesity prevention strategies targeting healthy diets and/or physical activity, 115 of which were conducted in the school setting. Statistically significant and beneficial effects were reported by about half of the school-based studies, especially those with a home involvement component. However, in the meta-analysis of eight studies of school-based interventions with a home component, the weighted mean BMI difference was beneficial but not significant. In the 2013 comparative effectiveness review and meta- analysis of 124 obesity prevention interventions in the United States described under essential criterion 1.1, Wang et al. (58) found greatest strength of evidence for the benefits of school- based approaches from physical activity alone with a home component and combined diet and physical activity with home and community components. However, in the meta-analyses, only the combined diet and physical activity interventions without a home component achieved significant reductions in BMI, whereas those with a home component did not. Three of the abovementioned meta-analyses found a greater beneficial influence on BMI or other weight-related outcomes due to a higher level of parental involvement. In the meta- analyses by Niemeier, Hektner and Enger (87) and Vasques et al. (50) described above, there were significant differences between levels of parental involvement with higher levels of involvement showing a greater effect on BMI reduction. In the meta-analysis by Krishnaswami et al. (90) also described above, the authors noted that the two studies with the highest community engagement performance scores achieved the highest outcome performance. Lastly, in their 2013 meta-analysis described under essential criterion 1.1 of 21 combined diet and physical activity programmes among children aged 4–11 years in primary schools and their effect on BMI and other anthropometric measures, Williams et al. (67) found that the combined @ Ingimage “All five meta-analyses that compared studies with or without parental involvement found better results in BMI and blood pressure among the studies with parental involvement. Parental involvement was identified as a moderating factor leading to greater success in three of these meta-analyses” 33RESULTS diet and physical activity interventions with stakeholder involvement and family engagement were particularly effective, but they were unable to pool results for this aspect in particular due to heterogeneity of studies. In addition to these meta-analyses, 18 reviews analysed parental involvement in school-based nutrition programmes. Six of these reviews only included studies that included parental involvement (70, 94-98), whereas 12 reviews compared studies with or without such involvement or investigating parental involvement as a feature of successful programmes (74, 78-80, 83, 99-105). Four reviews had analysis of parental involvement in school-based nutrition programmes as a main objective (94-96, 98), and four reviews examined the effect of different levels of parental involvement (94, 96, 99, 102). Only one of the reviews focused specifically on low- or middle-income countries and/or disadvantageous populations (70). Among the six reviews only including studies of interventions with a parental involvement (70, 94-98), the four reviews where study of parental involvement was a main objective noted beneficial effect on BMI, dietary and other outcomes (94- 96, 98). Hingle et al. in 2010 (94) systematically reviewed the effect of parental involvement in 24 obesity and NCD prevention studies, 16 of which were conducted in the school setting, comparing dietary interventions for children with and without a parental involvement component. Although limited conclusions could be drawn from this review because there were not enough studies making such comparisons, the authors observed that a greater proportion (100%) of studies using direct methods (e.g. parent attendance of educational or counselling sessions) achieved at least some dietary change (positive or mixed), while only 64% of studies using indirect methods (e.g. newsletters, tip sheets, invitations to attend events) achieved any changes in dietary outcomes. In 2018, Verjans- Janssen et al. (95) reviewed 25 school-based nutrition and physical activity intervention studies with a direct parental involvement component (e.g. requesting that parents attend education sessions, asking parents to attend or participate in family behaviour counselling or parent training sessions) in primary school children. They found favourable results in 11 of 18 studies measuring BMI or BMI z-score, though most studies had small effect sizes. Results on nutrition behaviour were inconclusive, with five out of 11 studies reporting favourable results for all outcomes measured (e.g. SSB intake, fruit and vegetable intake, added sugar intake, red meat consumption). In a 2016 systematic review of 11 studies of participatory, school- and community- based NCD prevention interventions, Jourdan et al. (98) found positive effects at the organizational level (e.g. development of policies) and in healthier eating among adult stakeholders and children. The authors concluded that involvement of children and parents or communities has potential benefits; however, the number of relevant studies was small and there were methodological challenges. As part of the 2012 series of systematic reviews on various aspects of nutrition education in countries with a high or very high human development index mentioned in essential criterion 1.1, the USDA (96) conducted another systematic review of 10 RCTs @ In gi m ag e 34 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE of parental involvement in nutrition education, of which seven were school-based. The authors concluded that limited and inconsistent evidence was available to assess the effects of parental involvement in nutrition education on children’s dietary behaviour. Six of the 10 studies found an added benefit from parental involvement in at least one subgroup, of which three studies found this effect in girls but not in boys. One of two studies measuring the effect of increasing levels of parental involvement found a significant dose-response relationship. Studies conducted outside of the United States were more often effective. Positive intervention effects in terms of reducing BMI, enhancing physical activity and improving fruit and vegetable consumption were also found in a 2015 systematic review by Langford et al. (97) of 26 diet and physical activity studies following the Health Promoting School approach, and therefore including a parent component. However, involving families — despite being an essential part of the Health Promoting School approach — was reported as highly challenging due to factors such as language barriers, misunderstanding the purpose of the programme and lack of time or opportunity. In the 2019 scoping review of 15 preschool multicomponent interventions in low socioeconomic settings in high-income countries described under essential criterion 1.1, Luybli, Schmillen and Sotos-Prieto (70) found that the nine successful studies, as well as the six inconclusive studies, included a parental involvement component. The authors noted that parental involvement is important in this age group, and such multicomponent interventions can address obesity in low socioeconomic settings. Among the 12 reviews analysing the effect of parental involvement as a feature of successful programmes eight reviews (74, 79, 80, 99-103) pointed to family or community involvement as a factor leading to positive outcomes in terms of weight- related outcomes, fruit and vegetable consumption or NCD risk factors, whereas four reviews found either no or mixed results (78, 83, 104, 105). In the 2018 systematic review of 56 weight gain prevention and management studies in children 2-19 years in various settings described under essential criterion 1.1, Bleich et al. (80) found that all 17 of the 24 school-based RCTs that reported positive results in weight-related outcomes used combined diet and physical activity interventions and included a secondary home setting. Among younger children, a 2012 review of 12 obesity prevention studies among children aged 4–6 years old in the school or preschool setting by Nixon et al. (99), found that all four studies reporting improved weight outcomes had high or moderate parental involvement. The authors noted children’s (and parents’) perceived competence in making healthy changes in their diet and physical activity levels as a key element for success. In a 2017 systematic review of 43 preschool-based obesity prevention studies, Ward et al. (100) found that 13 of 18 studies including a dietary intake measure and 10 of 24 studies with an anthropometric measure demonstrated at least one successful intervention effect. They constructed an intervention strength based on approaches common in community health efforts for obesity (e.g. use of multiple strategies, policy or environment changes, frequent exposures or longer durations) and found this to be positively correlated with reporting of positive anthropometric outcomes for physical activity, diet and combined interventions. Furthermore, parent engagement components increased the strength of these relationships. Among adolescents, the review described under essential criterion 1.1 by Meiklejohn, Ryan and Palermo (74) of 11 nutrition education studies in children aged 10–18 years, found that parental involvement (e.g. provision of newsletters, factsheets, meetings and shared homework tasks) was an additional component in all four studies with beneficial effect on BMI and in eight out of nine studies with beneficial effect on dietary intake. 35RESULTS In the 2006 review of 15 studies of fruit and vegetable promotion studies among primary and secondary school children described under essential criterion 1.1, Knai et al. (79) found that eight out of 11 studies including the active involvement of parents (at school and home) and three out of four studies including a community involvement component were associated with increased consumption of fruit and vegetables. Likewise, Silveira et al. (101) noted that parental involvement accentuated the positive effects on BMI, BMI z-score and fruit and vegetable consumption in a 2011 systematic review of 24 school-based studies among children aged 5–18 years focusing largely on nutrition education. In a 2018 systematic review of 41 school or preschool based nutrition education studies, Murimi et al. (102) found that 19 studies were successful in meeting their stated primary research objectives in terms of nutrition-related outcomes. Active parental involvement (e.g. face-to-face meetings, participation in nutrition or cooking classes, health fairs or tasting sessions) was associated with higher success than passive involvement (e.g. provision of messages, websites, homework) or no parental involvement, particularly in the preschool setting. In a broader systematic review of 37 school-based NCD risk factor prevention intervention studies, including healthy diets, Saraf et al. (103) in 2012 found that 80% of such interventions had positive effects on reducing NCD risk factors (e.g. nutrition knowledge, consumption of healthy foods, BMI, physical activity level) and that positive results were more likely when interventions included families and communities rather than schools alone. On the other hand, four reviews included did not find improved effects as result of parental involvement. In a 2012 systematic review, Van Lippevelde et al. (104) investigated five studies on school-based obesity prevention programmes focusing on diet and/or physical activity to determine the effect of parental involvement on diverse outcomes (dietary knowledge, health behaviours, BMI, BMI z-score, fat intake). While some positive effects of parental involvement were found, the evidence was inconclusive due to the lack of studies comparing interventions with and without a parent component. Likewise, Black, Matvienko-Sikar and Kearney (105) in 2017 found insufficient evidence to determine the impact of parental involvement in school and preschool interventions to improve dietary outcomes. In their review of 39 studies, of which 31 were conducted in schools or in preschools, substantial and statistically significant effects on dietary intake (fruit and vegetable or fat intake) were found in only one third of outcomes assessed, with no clear pattern appearing between the groups with and without a parent component. In the 2006 systematic review of 25 largely school-based obesity prevention studies focusing on diet and/or physical activity described under essential criterion 1.1, Doak et al. (83) found that slightly (but insignificantly) more of the non-effective interventions actively involved parents (57% vs. 47%) or the broader community (14% vs. 6%). In the systematic review of 14 school- based fruit and vegetable promotion studies also described in essential criterion 1.1, Aloia et al. (78) did not find positive associations between family or parental involvement and increases in fruit and vegetable consumption. Four systematic reviews examined the effect of different levels of parental involvement, all of which are described above. Hingle et al. (94) and Murimi et al. (102) both found active methods (e.g. parent attendance of educational or counselling sessions, face-to-face meetings, participation in nutrition or cooking classes, health fairs or tasting sessions) to be more effective than passive methods (e.g. newsletters, tip sheets, invitations to attend events, provision of messages, websites, homework). USDA (96) found a significant dose-response relationship in one of two nutrition education studies measuring parental involvement. In children aged 4–6 years old, Nixon et al. (99) found that all four studies reporting improved weight outcomes had high or moderate parental involvement. 36 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Essential Criterion 2.3: School staff training in nutrition and health-related issues School staff play a critical role in the implementation of many school-based health promotion interventions. While none of the systematic reviews in this report specifically assessed the effectiveness of school staff training in nutrition and health-related issues, several reviews of school-based nutrition interventions delivered by teachers mention the importance of adequate staff training and many of the primary studies included teacher training conducted alongside other activities. However, these aspects were seldom analysed to evaluate the impact of staff training apart from narrative descriptions from findings of individual studies. Five systematic reviews were identified that examined aspects of school staff training or identified it as a success factor (79, 97, 102, 106, 107). Two Cochrane reviews by Wolfenden et al. (106, 107) in 2016 and 2017 systematically reviewed the effectiveness of strategies to improve implementation by regular staff of health promoting programmes including healthy diet and physical activity from 27 studies in schools and 10 studies in child care centres (e.g. preschools). Such strategies included staff education and training, performance feedback, prompts and reminders or implementation resources. While many studies reported significant improvements as a result of the implementation strategy compared to usual practice, the authors found insufficient evidence to draw conclusions regarding their effectiveness at improving implementation or improving staff knowledge, attitudes and skills. The three other systematic reviews looked at teachers’ training when assessing study designs or components of those studies that had shown effect, and identified training of teachers as an enabling factor. In the 2015 review of 26 school-based diet and physical activity programmes aligning with the Health Promoting School approach described previously under essential criterion 2.2, Langford et al. (97) identified working with teachers to develop programmes and increase ownership and providing ongoing training, support and communication as moderators of positive results in terms of reducing BMI, enhancing physical activity and improving fruit and vegetable consumption. Furthermore, in the systematic review by Knai et al. (79) of 15 fruit and vegetable promotion interventions described under essential criteria 1.1 and 2.2, eight out of 11 studies using teacher involvement (e.g. training) as an intervention component had significant positive results on fruit and vegetable consumption. The authors observed that special training of teachers appeared to be associated with successful results in increasing fruit and vegetable intake. Finally, in the 2018 systematic review of 41 school or preschool based nutrition education studies described under essential criterion 2.2 above, Murimi et al. (102) found that studies that trained existing teachers or engaged nutrition and physical activity experts in the implementation were more likely to succeed. NFSI Component 3: Developing a nutrition and health promoting curriculum The third component of the NFSI framework concerns the school curriculum, which can be particularly important for nutrition. The essential criteria in this component relate to the development and implementation of curricula that include nutrition education, physical education and life-skills education, as well as to the regular monitoring and evaluation of these aspects of the curricula. Most of the reviews identified concerned different aspects of nutrition education in schools. The number of reviews related to physical education was lower since this review of reviews only considered physical education activities that were implemented in combination with nutrition programmes. Likewise, few reviews were identified for nutrition-related life-skills education. No reviews were identified for monitoring and evaluation of school curricula. 37RESULTS NFSI Component 3: Developing a nutrition and health promoting curriculum • Ten meta-analyses and 28 reviews were identified relevant to essential criterion 3.1 that examined various aspects of nutrition education in schools, including one systematic review of studies from low- and middle-income countries. • Eight of the 10 meta-analyses found significant improvements in weight-related or dietary outcomes. • Four systematic reviews highlighted the benefits of integrating nutrition education into the formal curriculum, while one meta-analysis found that curriculum-based approaches were only effective when implemented alongside other teaching strategies. • Effective nutrition education programmes were often identified as those that are behaviourally- focused, of longer duration and intensity, facilitated by trained staff, integrated into the curricula and/or implemented in combination with other approaches (e.g. changes to the environment, parental involvement). • Effective approaches included enhanced curriculum, cross-curricular approaches, experiential and hands-on learning strategies (e.g. cooking classes, school gardening), contingent reinforcement approaches and peer-led education programmes. • Information and communication technology (ICT) may have potential to improve dietary habits among adolescents in some settings. • Thirteen meta-analyses and four systematic reviews were identified relevant to essential criterion 3.2 comparing physical activity education alone or in combination with nutrition activities. • Nine meta-analyses and two reviews found better effects on weight-related, dietary or other outcomes for combined programmes than for those that only targeted physical activity. • Eight reviews were identified relevant to essential criterion 3.3 examining life-skills education in the context of nutrition. • Effective intervention components were suggested to be experiential learning, such as cooking or gardening skills. • Self-efficacy and children’s perceived competence in making healthy changes to their diet and physical activity levels were found to be associated with better outcomes. Essential criterion 3.1: Culturally appropriate and effective nutrition education Nutrition education is a common strategy within programmes to improve nutrition and dietary behaviours, practices and habits. It aims to achieve sustained changes in behaviour and build food and nutrition capacities. No particular WHO guideline exists for nutrition education, in schools or other settings, but such activities should be age- appropriate, context-specific and underpinned by guidance on healthy and safe diets (108-110). FAO has developed a comprehensive planning guide for establishing context-relevant and age-appropriate effective nutrition education in primary schools (20), and is currently developing a white paper and a series of tools to enhance the methodological quality, 38 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE impact and scope of food and nutrition education in low- and middle-income countries (111). Ten meta-analyses (53, 60, 92, 112-118) and 28 reviews (45, 68, 69, 74, 75, 78, 79, 96, 101, 102, 119-136) were identified that assessed the impact or effective components of school-based nutrition education. Of the 10 meta-analyses relevant to nutrition education, four focused explicitly on school- based nutrition education (92, 112-114), whereas six considered the effect of nutrition education amongst other approaches (53, 60, 115-118). All the four meta-analyses that explicitly focused on school-based nutrition education showed significant improvements in BMI, other weight-related outcomes, biochemical measures, dietary outcomes and knowledge. One of these was a 2015 review by Dudley, Cotton and Peralta (92) of 49 school-based studies to promote healthy eating that also examined the effect of different teaching approaches and strategies. In their meta-analysis of data involving 20 234 children they found that most teaching strategies led to positive changes, but in the pooled results, fewer strategies had mean effect sizes Md > 0.4, above the selected cut-off for positive effect in this review. Curriculum-based approaches were most popular but only had mean effect sizes Md > 0.4 in the meta-analyses when implemented alongside other teaching strategies. Enhanced curriculum approaches were effective at increasing knowledge (Md = 0.75) but not at reducing sugar intake from beverages with added sugar or fruit juices (Md = 0.28). Cross-curricular approaches were effective to improve fruit and vegetable consumption (Md = 0.63) as well as decreasing sugar intake from beverages with added sugar or fruit juices (Md = 0.42). Experiential learning strategies (e.g. school garden, cooking demonstrations) were associated with the largest pooled mean effect size for preference and increased nutritional knowledge outcomes (Md = 1.35), and were also effective for food consumption/energy intake (Md = 1.31) and for fruit and vegetable consumption (Md = 0.68). In a 2018 meta-analysis of 30 studies, Wang et al. (112) found that educational interventions on health as single or multicomponent interventions among 35 296 children had significant effect on a range of adiposity-related outcomes, including BMI (-0.15kg/m2; 95% CI: -0.24, -0.05, P = 0.003), BMI z-score (-0.03; 95% CI: -0.05, -0.02, P < 0.001), waist circumference (-0.97; 95% CI: -1.95, 0.00, P = 0.050), triceps skinfold (-1.39; 95% CI: -2.41, -0.37, P = 0.008), SBP (-1.13; 95% CI: -2.20, -0.07, P = 0.037), total cholesterol (-4.04; 95% CI: -7.18. -0.90, P = 0.012), and triglycerides (-2.62; 95% CI: 4.33, -0.90, P = 0.003). However, educational interventions were found to have little or no significant impact on the waist-to-hip ratio, DBP or high- or low-density lipoprotein. In another meta- analysis of eight studies on nutrition education interventions among 8 722 children in primary and secondary school, Silveira et al. (113) found in 2013 that, while only two individual studies demonstrated significant reductions in BMI, the pooled results from the eight studies gave a significant effect on BMI (-0.33 kg/m2; 95% CI: -0.55, -0.11). The largest effect was observed among the four studies with a duration greater than one year (-0.48 kg/m2; 95% CI: -0.76, -0.19). In China, Kong, Liu and Tao (114) performed a systematic review and meta-analysis in 2016 of 17 studies on school- based nutrition education interventions for obesity prevention in primary schools in the country. They found significant impact of interventions on obesity prevalence among the 14 studies of higher methodological quality involving 17 277 children (OR: 0.73; 95% CI: 0.55, 0.98) and in two studies involving 6 029 children where the duration was longer than 2 years (OR: 0.49; 95% CI: 0.42, 0.58). Four of the six meta-analyses that considered the effect of nutrition education combined with other approaches or components showed beneficial effect on dietary outcomes (60, 115-117), one only showed 39RESULTS “ Effective nutrition education programmes were often identified as those that are behaviourally focused, of longer duration and intensity, facilitated by trained staff, integrated into the curricula and/or implemented in combination with other approaches (e.g. changes to the environment, parental involvement)” @ In gi m ag e 40 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE an impact when nutrition education was combined with physical education (53) and one had mixed results (118). In the 2007 meta-analysis of seven school-based studies described under essential criterion 1.1, Howerton et al. (60) found increased fruit and vegetable consumption in the intervention group compared to the control group (+0.45 servings; 95% CI: +0.33, +0.59). All studies included a classroom-based component with lessons or workshops, coupled with a variety of other components (e.g. family involvement, food service changes). In a 2018 meta-analysis of 30 studies aiming to increase vegetable intake in children aged 2-5 years of age, 24 of which were conducted in preschool or child care settings, Nekitsing et al. (115) found a small-to-moderate effect (Hedge’s g = 0.40) from pooling results of 30 studies and a slightly higher effect (g: 0.42; 95% CI: 0.33, 0.51, P < 0.001) from pooling results of all 44 intervention arms. In subgroup analyses of intervention strategies, the effect of educational strategies alone was lower (g: 0.30; 95% CI: 0.15, 0.46). The studies using taste exposure had a significantly higher impact on intake (g: 0.79; 95% CI: 0.53, 1.05) than educational (g: 0.30; 95% CI: 0.15, 0.46) strategies alone. In another meta-analysis of various interventions to promote fruit and vegetable consumption in primary schools (e.g. curricula, environment, teacher/parental involvement, fruit and vegetable schemes), Delgado-Noguera et al. (116) found in 2011 that the use of computer-based nutrition education was the only intervention strategy found to be effective at increasing fruit and vegetable consumption. Pooled results of two such programmes involving 606 primary school children showed effectiveness in improving consumption of fruit and vegetables (SMD: 0.33; 95% CI: 0.16, 0.50). In a 2017 meta-analysis of interventions to reduce SSB consumption and obesity which included 10 964 children, Vargas-Garcia et al. (117) found that interventions significantly decreased consumption (-76 mL/day; 95% CI: -105, -46). The subgroup analysis of 15 school-based interventions which primarily consisted of nutrition education also resulted in a significant decrease in SSB consumption (-28 mL/day; 95% CI: -42, -12). On the other hand, in the 2012 meta-analysis of 23 school-based physical activity and nutrition education interventions of 17 693 children between 7 and 17 years described under essential criterion 1.1, Friedrich, Schuch and Wagner (53) found that neither physical activity nor nutrition education interventions in isolation showed a significant reduction in BMI, but that the reduction was significant for combined interventions (SMD: 0.37; 95% CI: 0.63, 0.12). Furthermore, in their 2018 meta-analysis of 16 studies, 12 of which were school-based, Abdel Rahman et al. (118) only found a modest effect of educational and behavioural interventions in reducing SSB intake from two of the school-based studies (MD: 26.53 mL/day; 95% CI: 53.72, 0.66; P = 0.06) and found no effect on reduction of BMI z-scores. In addition to these meta-analyses, 28 reviews investigated different aspects of nutrition education in schools (45, 68, 69, 74, 75, 78, 79, 96, 101, 102, 119-136). Among these, five reviews examined integration into formal curricula or education resources (69, 79, 101, 119, 120), five considered the benefits of nutrition education delivered as part of multicomponent interventions (74, 75, 96, 102, 121), six examined or noted the use of hands-on activities (e.g., gardening or cookery classes) in nutrition education (102, 121-125), six focused on the use of ICT technologies as platforms for school- based nutrition education activities (126-131), one focused on the type of educator (132), one focused on peer-led nutrition education (133) and six considered the effectiveness of nutrition education as one approach amongst several school-based health and nutrition interventions (45, 68, 78, 134-136). Only one systematic review focused on studies from low- and middle-income countries (69). 41RESULTS Of the five reviews that examined integration into formal curricula or education resources, four systematic reviews highlighted the benefits of integrating nutrition education into the formal curriculum (69, 79, 101, 119). In a 2017 systematic review of nine studies of nutrition education integrated into the school curriculum and delivered by trained teachers in elementary schools in the United States, Price et al. (119) found significant reductions in BMI in five out of nine studies. Interventions with a duration greater than 1 year showed more pronounced results, with positive impact on reducing overweight, obesity and BMI outcomes. In the 2011 systematic review of 24 school-based nutrition education programmes described under essential criterion 2.2, Silveira et al. (101) found evidence of positive effects on rates of overweight and obesity and on fruit and vegetable consumption from interventions that were greater than one year in duration, introduced into the regular activities of the school or the curriculum, included parental involvement and/or provided fruit and vegetables as part of the school food services. In the 2006 systematic review of 15 fruit and vegetable promotion interventions by Knai et al. (79) described under essential criteria 1.1, 2.2 and 2.3, six out of nine studies integrating fruit and vegetable knowledge or preparation into the curriculum as an intervention component had significant positive results on consumption. In the 2012 systematic review of 25 school- based diet and physical activity interventions in low- and middle-income countries described under essential criterion 1.1, Verstraeten et al. (69) noted that effective interventions integrated educational activities into the school curriculum. None of the intervention studies using nutrition education alone had a significant effect on mean BMI, but two high-quality studies conducted using combined diet and physical activity interventions through integrated curriculum delivered by teachers and accompanied by nutrition education of parents each had large effect sizes. In a 2016 scoping review of studies of 32 nutrition education resources, Peralta, Dudley and Cotton (120) found that most nutrition education resources included curriculum-based aspects and were delivered in and outside classrooms, through online or web-based learning. However, the resources were less likely to embed cross-curricular, experiential learning approaches and contingent reinforcement activities. Five systematic reviews with an explicit focus on nutrition education considered the benefits of this approach delivered as part of multicomponent interventions (74, 75, 96, 102, 121). In the 2018 systematic review of 41 school or preschool based nutrition education studies described under essential criteria 2.2 and 2.3, Murimi et al. (102) found that 19 studies were successful in meeting their stated primary research objectives in terms of nutrition-related outcomes (weight-related outcomes; biochemical markers such as HDL-C; dietary intake of calories and/or various macro- and micronutrients; dietary consumption or 42 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE preference of, for example, fruits and vegetables, whole grains or unhealthy beverages or snacks; dietary diversity; children’s nutrition knowledge; and food availability in cafeterias), whereas 19 studies only partially met and three did not meet their stated objectives on nutrition-related outcomes. Successful interventions were more often multicomponent and multilevel, implemented for more than 6 months with frequent exposures, engaged parents actively, trained existing teachers or engaged nutrition and physical activity experts in the implementation. The authors identified as a key factor of effectiveness the targeting of specific behaviours and making nutrition education behaviour-focused through age-appropriate and/ or experiential activities. In the 2016 systematic review of 11 RCT studies of nutrition education implemented as part of multicomponent strategies for children 10–18 years described in essential criterion 1.1, Meiklejohn, Ryan and Palermo (74) reported that nine studies demonstrated significant changes in dietary intake (e.g. reduced consumption of SSBs, increased consumption of fruit and vegetables) and four studies obtained significant changes in anthropometric measures (e.g. BMI, weight change, percentage body fat). Other contributing factors were parental involvement, behaviourally-focused nutrition education and the use of theories to underpin the intervention design, incorporation of policy changes within @ In gi m ag e school settings (changes in canteens, food supply and vending machines) and combining nutrition education with physical activity programmes. As mentioned under essential criteria 1.1 and 2.2, the USDA conducted a series of systematic reviews in 2012 on various aspects of nutrition education in countries with a high or very high human development index (75, 96, 121, 130, 132). In a systematic review of 14 studies, 10 of which were school-based, USDA (121) noted that six studies found multicomponent interventions to be more effective than single component interventions for at least one dietary outcome, three studies found single component nutrition education interventions to be more effective than multicomponent interventions for at least one dietary outcome, and eight did not find any difference between them. However, in another systematic review in the same series of 10 school-based nutrition education studies, USDA (75) found consistent evidence that combining nutrition education with changes to the school food environment is more effective for improving children’s and adolescents’ dietary intake than either component alone. Furthermore, in the same series USDA (96) found limited and inconsistent evidence from 10 studies that combining multiple approaches by involving parents in nutrition education improves dietary behaviour, though more positive effects were found among girls and in studies conducted outside the United States. 43RESULTS Six reviews examined or noted the use of hands-on activities, such as gardening or cookery classes, in nutrition education (102, 121-125). In the systematic review of 41 school- or preschool- based nutrition education studies described above, Murimi et al. (102) found that successful interventions included age-appropriate and/ or experiential activities (e.g. hands-on learning, cooking or tasting session, play and games on topics that are important to the age group). Also described above, the systematic review of 14 multiple vs. single component nutrition education studies by USDA (121) noted that the three studies that consistently found better effect of multicomponent interventions were also the only ones to combine nutrition education with a hands-on component (e.g. cooking or gardening). In a 2019 systematic literature review of 44 school- based studies, Bailey, Drummond and Ward (122) found that nutrition education interventions to promote food literacy in adolescents improved dietary intake in eight out of nine studies, but the effect on long-term healthy dietary behaviour was mixed. Teaching of hands-on cooking skills and use of chefs or school gardens were identified as beneficial components. Four of five studies on school-based gardening showed a positive impact on fruit and vegetable consumption. All four studies that utilised experimental cooking and nutrition education programmes led by chef instructors showed positive results in overall cooking confidence, cooking skills and tasting new foods. The qualitative studies found that while adolescents value food literacy lessons and home economics teachers rate various aspects of food literacy as important skills, the school food environments are seldom comprehensively supportive of food literacy. In a 2009 systematic review of 11 studies on garden-based nutrition education programmes — eight of which were school-based — Robinson-O’Brien, Story and Heim (123) found that these programmes have the potential to promote fruit and vegetable intake among youth and to increase willingness to taste fruits and vegetables among younger children. Among studies in the school setting, improvements were reported as result of exposure to garden-based nutrition education in two out of three studies measuring fruit and vegetable intake, two out of five studies measuring willingness to try vegetables, three out of three studies reporting on willingness to taste vegetables and three out of five measuring nutrition knowledge. In a 2015 systematic review of 102 quantitative and qualitative studies of school architecture and design, Frerichs et al. (124) analysed 21 studies of school garden on-site food production. They found that these could easily be integrated into curricula and have benefits beyond improved dietary behaviour, however, barriers such as intensive time and resource requirements were identified. They also analysed 14 studies relating to school teaching kitchens and found that these were popular among staff and students and could easily be integrated with other academic topics. Five of seven studies measuring dietary outcomes found significant increases in the consumption of, for example, vegetables. In a narrative review of six school-based hands-on culinary skill interventions — often combined with components such as nutrition education, gardening, tasting sessions or trips to farmers’ markets — Muzaffar, Metcalfe and Fiese (125) in 2018 identified too few studies to draw conclusive findings. However, all six studies were successful in meeting their objectives and the authors noted quantitatively significant effects on food preferences, cooking skills, cooking self-efficacy, cooking behavioural intentions, food-preparation frequency, knowledge, healthy dietary intake, BMI and blood pressure from three of the studies. Furthermore, qualitative evaluations available from all six studies indicated substantial effects in terms of programme appeal and improvements in cooking skills and healthy eating. 44 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Another six reviews focused on the use of ICT technologies as platforms for school- based nutrition education activities and these generally showed positive results, at least in the short term (126-131). In 2014, Ajie and Chapman-Novakofski (126) reviewed 15 studies on computer-based nutrition education interventions targeting overweight and obesity in adolescents 12–18 years, 10 of which took place in the school setting. Most of these 10 studies found significant results in at least one of the dietary or anthropometric outcome measures post-intervention, albeit some were only successful among at-risk subgroups or among girls. Likewise, Whittemore et al. (127) in 2013 reviewed 12 school-based internet obesity prevention programmes among adolescents and found that 10 of these improved dietary behaviour or physical activity in the short term. However, only one out of four studies that examined the effect on BMI showed a significant decrease. Two of the three studies that compared web-based nutrition education to traditional nutrition education noted a greater effect on health behaviours and BMI for the web-based education interventions. Hamel and Robbins (128) conducted a 2013 M r. Ea gl e’ s Ki ds C hi ld ho od O be si ty P ro je ct 45RESULTS systematic review of 15 computer- and web- based interventions to promote healthy eating among children and adolescents, 10 of which were conducted in the school setting. Although 11 of the 15 interventions included in the review resulted in statistically significant positive changes in weight- or diet-related outcomes, these changes were not maintained in studies that included post- intervention follow-up. Interventions conducted in schools or using individually tailored feedback enhanced success. The authors suggest that the closely supervised environment of a classroom setting, along with the potential for social support from teachers or fellow classmates, may enhance the effect of a school-based intervention over a home-based approach that may be completed alone. In another systematic review of 11 school-based interventions using ICT including computer games, programmes, text messages and interactive CD-ROMs, do Amaral e Melo et al. (129) found in 2017 statistically significant positive effects on diet in five of the studies. The authors were unable to determine the most effective type of intervention due to the heterogeneity of studies. However, they suggest that long-term interventions aimed at changing a single health behaviour (e.g. fruit and vegetable intake) and with frequent exposure may improve nutrition behaviours. Furthermore, interventions that target diet and physical activity together seem to be have greater impact, as well as interventions using games. As part of the series of systematic reviews on various aspects of nutrition education in countries with a high or very high human development index described above, the USDA (130) conducted a systematic review of 24 studies on nutrition education delivered via digital media and/or technology, 15 of which were school-based. They found moderate evidence that that nutrition education delivered via digital media/technology (computer- and internet-based programmes) may be effective for improving dietary intake-related behaviours among children and adolescents: 21 of 24 studies found significant effect on dietary intake-related behaviours in at least one population subgroup, and in 15 of these the intervention was more effective than the control (no intervention or other delivery method). In a 2019 systematic review of 29 various school-based healthy eating intervention studies in adolescents, Calvert, Dempsey and Povey (131) found that, overall, 24 of the studies were successful in @ In gi m ag e 46 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE promoting dietary behaviour change. Among the interventions that appeared to be more effective, were those that used educational media to deliver health messages (all seven studies with this component were successful) and incorporated computer-based individualized feedback with normative information on eating behaviours (four of five studies with this component were successful). One systematic review by the USDA (132) in the series of systematic reviews in 2012 on various aspects of nutrition education mentioned above focused on the type of educator. However, not more than one study was considered eligible for inclusion, thus there was insufficient evidence to draw conclusions. Moreover, the results from this study were mixed, resulting for the most part in no significant differences in dietary intake between the intervention groups, although the changes in dietary intake were significantly greater in the teacher group compared to the nutritionist group. Another review focused on peer-led nutrition education. In this 2016 systematic review of 17 school-based, peer-led nutrition education interventions, Yip et al. (133) found such approaches to be effective in improving knowledge, self-efficacy and attitudes towards healthy eating. Such programmes also generated favourable outcomes for anthropometric measures (BMI, waist circumference) or other biological measures (blood pressure) in all four studies measuring these outcomes. Furthermore, a dose-impact relationship was observed for varying levels of participation in all three studies investigating this aspect. Among the six reviews that considered nutrition education as one approach amongst several school-based school health and nutrition interventions, six reviews found positive results on various weight-related and dietary intake indicators, but generally concluded that it is most effective when delivered as part of a more comprehensive package (45, 78, 134), whereas three reviews found moderate or inconclusive evidence (68, 135, 136). In the 2016 systematic review of 14 interventions targeting fruit and vegetable consumption among children in pre- and primary schools in the United States described under essential criteria 1.1 and 2.2, Aloia et al. (78) found that eight of the studies showed significant increases in consumption of fruit and vegetables among pre- and primary school-aged children. The authors identified nutrition education as one of the most promising components, although further research would be needed to strengthen the evidence. In the 2017 systematic review of 36 school-based and largely educational interventions to reduce SSB consumption among adolescents aged 12–17 years described under essential criterion 1.1, Vézina-Im et al. (45) found that although the 10 legislative/environmental studies (e.g. ban on sugary drinks in schools) had the highest success rate (90%), the 20 educational/behavioural interventions (e.g. nutrition education) and the six interventions that were both educational/ behavioural and legislative/environmental were also successful (65% and 66.7%, respectively). In a 2015 systematic review of eight RCT studies to reduce SSB consumption, seven of which were school-based, Avery, Bostock and McCullough (134) found that four of six studies focusing on educational approaches reduced the risk of being overweight, though one of these also provided a replacement drink. The authors concluded that school-based education programmes to reduce SSB consumption, and which include follow-up modules, offer opportunities for implementing effective, sustainable interventions. On the other hand, three of the eight reviews comparing nutrition education to other approaches did not find positive results. In the 2011 systematic review of 11 school-based obesity prevention interventions described under essential criterion 47RESULTS 1.1, De Bourdeaudhuij et al. (68) found that interventions focusing on education alone have less of an effect on obesity prevention than multicomponent programmes where education is combined with environmental modifications. In a 2010 review of programmes targeting healthy diets in primary and secondary schools in Europe, Van Cauwenberghe et al. (135) found moderate evidence that educational interventions (i.e. a nutrition education curriculum that was delivered by the teachers) can improve dietary behaviour in adolescents and limited evidence that it can improve dietary behaviour in children, while evidence of the effect on body composition was inconclusive. For both children and adolescents, combined educational and environmental interventions were more effective than those targeting education alone. Likewise, in a 2018 systematic review of 48 school cafeteria environment studies, Gordon, Dynan and Siegel (136) found that nutrition education interventions (“system 1” type of interventions) were less effective than environmental interventions (“system 2” type of interventions) within the school cafeteria to impact nutrition-related outcomes (e.g. BMI, consumption and/or selection of fruits, vegetables, sugary drinks, plain milk and/or whole grains) (four of eight system 1 studies vs. 24 of 27 system 2 studies, P = 0.033). The review is further described under essential criterion 4.1. Essential criterion 3.2: Age-, sex- and culturally- appropriate physical education curriculum WHO guidelines recommend that children and youth aged 5–17 should accumulate at least 60 minutes of moderate- to vigorous-intensity physical activity daily, but longer durations would provide additional health benefits (137). WHO guidelines also exist for preschool age children, which, among other things, recommend that children between 3 and 4 years should spend at least 180 minutes in a variety of types of physical activity at any intensity per day, and that at least 60 minutes of those should be on moderate- to vigorous intensity physical activity (138). For children and young people, physical activity includes play, games, sports, transportation, chores, recreation, physical education or planned exercise and can be done in the context of the school. Both the 2004 WHO Global Strategy on Diet, Physical Activity and Health (139) and the 2018 WHO Global Action Plan on Physical Activity (140) call for physical education and enabling environments for physical activity in schools. A wealth of reviews exists on the effect of promoting physical activity in schools. However, to limit the scope of the evidence base for this review of reviews for the NFSI framework, only those reviews that assessed the effectiveness of combined physical activity and nutrition versus physical activity alone were considered. This resulted in 13 meta-analyses (46-50, 53-58, 61, 62) and four reviews (69, 81, 83, 85). As described under essential criterion 1.1, five of the 13 meta-analyses comparing physical activity education alone or in combination with nutrition activities found significant improvements of the combined programmes on weight-related outcomes (48, 50, 53, 54) or blood pressure (61). These improvements, however, were not found for programmes that focused on physical activity alone. Another four meta-analyses described under essential criterion 1.1 found significant effect of both combined diet and physical activity programmes and those focusing on physical activity alone on weight-related outcomes (47, 49, 55) or blood lipids (62), though the pooled effect sizes were larger for the combined programmes. In the 2011 Cochrane review of 55 long- and short-term childhood obesity prevention programme studies, 43 of which included an educational setting, Waters et al. (47) found the greatest overall effect on BMI or BMI z-score 48 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE from combined physical activity and dietary interventions (SMD: -0.18; 95% CI: -0.27,-0.09), and a slightly smaller effect size from physical activity interventions alone (SMD: – 0.11; 95% CI: -0.19, -0.02). In 2017, updating the review by Waters et al. (47), Gori et al. (49) reviewed 72 studies in overweight or obese primary school age children, 47 of which were school-based, and found significant results for combined diet and physical activity interventions in children aged 6–12 years (mean BMI-SDS: -0.13; 95% CI: -0.19, -0.06) with a smaller effect size in the same age group for physical activity alone (mean BMI-SDS: -0.11; 95% CI: -0.16, -0.06). In the 2012 meta-analysis of 43 studies on school-based interventions, Lavelle, Mackay and Pell (55) found significant reduction in BMI from combined diet and physical activity interventions (-0.17 kg/m2; 95% CI: -0.29 to -0.06, P < 0.001) and a smaller effect from the physical activity interventions used in isolation (-0.13 kg/m2; 95% CI: -0.22 to -0.04, P = 0.001). In the 2014 meta- analysis of 17 child obesity prevention studies focusing on blood lipids, Cai et al. (62), found significant improvements in LDL-C from combined programmes (WMD: -8.49 mg/dL; 95% CI: -12.77, -4.21) which was larger than from programmes focusing on physical activity alone (WMD: 2.67mg/ dL; 95% CI: -4.30, -1.03). Furthermore, two other meta-analyses described under essential criterion 1.1 had mixed results when comparing physical activity alone or in combination with diet (56, 57). In the 2019 Cochrane systematic review and meta-analysis of 153 obesity prevention studies, of which 91 were school-based and 22 were in kindergartens or preschools, Brown et al. (57) found an effect of combined diet and physical activity interventions in children aged 6–12 years on BMI z-score (-0.04; 95% CI: 0.08, -0.01) and on BMI from physical sightandlife.org 49RESULTS activity alone (-0.10 kg/m2; 95% CI: -0.14, -0.06). In adolescents between 13 and 18 years, significant effect was found of physical activity alone on BMI z-score (-0.20; 95% CI: -0.30, -0.10) and on BMI (-1.53 kg/m2; 95% CI: -2.67, -0.39), with no effect from the combined interventions. In China, Feng et al. (56) reviewed 76 school-based obesity intervention studies in 2017 and found that combined diet and physical activity interventions with health education had a significant effect and slightly larger effect on BMI than physical activity only interventions in treatment studies ((SMD: -1.80; 95% CI: -2.15, -1.44) vs. (SMD: -0.91; 95% CI: 1.15, 0.67)) whereas in prevention studies only the combined approach was effective (SMD: 0.19; 95% CI: -0.27, -0.11). In contrast, one meta-analysis found better effect of interventions employing physical activity alone than of programmes combining diet and physical activity. In this 2014 meta- analysis, Langford et al. (46) did find an effect of multicomponent interventions employing multiple approaches (i.e. educational, environmental and family involvement) aligned with the Health Promoting School approach. Examining programme components however, they found no effect of the combined diet and physical activity interventions but significant effect on BMI (MD: -0.38 kg/m²; 95% CI: 0.73, 0.03) and BMI z-score (MD: -0.47; 95% CI: -0.69, -0.25) of interventions focusing on physical activity alone. Finally, in the 2013 comparative effectiveness review and meta-analysis of 124 school-based interventions in the United States, Wang et al. (58) found strong evidence for school-based combined diet and physical activity interventions with a home component, whereas moderate strength of evidence was found for diet or physical activity interventions alone. However, results were only pooled for the combined diet and physical activity interventions, but not for interventions that addressed either one alone. In addition to these meta-analyses, findings from two of the four reviews included under essential criterion 1.1 support the usefulness of integrating physical activity within nutrition programmes for better effect (69) as well as for long-term results (81). In the 2009 systematic review of 34 “lifestyle” intervention studies to prevent obesity among schoolchildren, Brown and Summerbell (81) found that the efficacy of interventions targeting physical activity alone was inconsistent, whereas combined interventions which targeted both diet and physical activity may help prevent children from becoming overweight in the long term. In low- and middle-income countries, Verstraeten et al. (69) conducted in 2012 a systematic review of 15 studies targeting physical activity and dietary behaviours for the primary prevention of obesity in children and adolescents aged 6–18 years. While four out of the five physical activity intervention studies had positive effect, the largest effect size on BMI came from two high- quality studies of combined diet and physical activity interventions. In contrast, two other systematic reviews did not find the combined physical activity and nutrition programmes more effective than physical activity programmes alone (83, 85). In the 2006 systematic review of 25 largely school-based obesity prevention studies focusing on diet and/or physical activity, Doak et al. (83) found that slightly (but insignificantly) more of the non-effective interventions intervened on both diet and activity (88% vs. 82%). Similarly, in the 2008 systematic review of 14 school-based obesity prevention programmes of at least 6 months duration, Kropski, Keckley and Jensen (85) did not find that combined diet and physical activity interventions were more effective than interventions with either component alone. Significant reduction in overweight was reported in at least one subgroup in five of 11 combined studies and one of two physical activity studies. 50 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Essential criterion 3.3: Healthy living and life-skills education curriculum In addition to nutrition education and physical education, healthy living and life-skills education should be included in the curriculum of nutrition- friendly schools. WHO defines life skills as “abilities for adaptive and positive behaviour that enable individuals to deal effectively with the demands and challenges of everyday life” (WHO, 2003). These skills include the psychosocial competencies and interpersonal skills necessary for decision- making, problem solving, communication, critical thinking and self-management, as well as, for example, critical evaluation of nutrition advertisements or countering social pressure to adopt unhealthy eating habits. In this way, life- skills education enables children and adolescents to manage a healthy lifestyle, including nutrition. Eight reviews were identified that analysed life-skills education related to nutrition (98, 99, 122, 123, 125, 127, 133, 141). Three reviews attributed programme success or effectiveness to the development of life skills (99, 122, 141). In the 2012 systematic review of strategies to prevent overweight and obesity in children aged 4–6 years described under essential criterion 2.2, Nixon et al. (99) noted that children’s self-perceived competence in making healthy changes in their diet and physical activity levels was a key element for success. Furthermore, interventions with the highest success included development of skills, behavioural capability and self-efficacy, in addition to parental involvement. A 2014 systematic review by Hersch et al. (141) examined evidence from eight studies — seven of which were school- based — measuring the impact of cooking programmes on food preparation skills, dietary intake, cooking confidence, fruit and vegetable preferences, attitudes toward food and cooking- or food-related knowledge. Findings suggested that cooking programmes may have a positive impact on children’s food-related preferences, attitudes and behaviours. Similarly, as described under essential criterion 3.1 earlier, the 2019 systematic literature review by Bailey, Drummond and Ward (122) of 44 school-based food literacy studies noted that four out of five studies on school- based gardening showed a positive impact on fruit and vegetable consumption. Furthermore, all four studies utilising experimental cooking and nutritional education programmes led by chef instructors showed positive results in overall cooking confidence, cooking skills and tasting new foods. As highlighted above, the authors noted from qualitative studies that both adolescents and home economics teachers value food literacy skills, but that school food environments are seldom supportive. Five reviews reported on self-efficacy of students as an intervention outcome (98, 123, 125, 127, 133). In the 2016 systematic review by Yip et al. (133) of 17 peer-led nutrition education programmes described under essential criterion 3.1 above, 13 studies that evaluated self-efficacy, knowledge, intentions and attitudes related to health behaviours showed improvement and two studies that measured self-esteem also reported improvements. One study that reported on body image found improvement in both the intervention and control groups. Despite the potential of peer-led programmes to improve knowledge, self-efficacy and attitudes towards healthy eating, the authors highlight the need for more research to determine the impact of these changes on diet quality and for more studies to use validated tools to measure attitudes and self- efficacy so as to strengthen the quality of evidence. In the 2013 review of 12 school-based Internet obesity prevention programmes also described under essential criterion 3.1 above, Whittemore et al. (127) found improved adolescents’ self-efficacy for healthy eating and physical activity in all three studies that targeted self-efficacy. An additional study included training in coping skills as a part of 51RESULTS the interactive Internet programme, but there was no difference in outcome measures between the Internet education and behavioural programme with or without the addition of coping skills training. In the 2009 review of 11 garden-based nutrition education studies also described in essential criterion 3.1 above, Robinson O’Brien, Story and Heim (123) included two studies that measured self-efficacy to consume fruit and vegetables. They noted that self-efficacy increased in the intervention group in one study, but it did not improve in either the intervention or control group in the other study. In the 2016 systematic review of the involvement of young people in school- and community-based NCD prevention interventions described under essential criteria 1.1 and 2.2, Jourdan et al. (98) found that the main objective of some interventions was to empower children and increase their leadership and advocacy skills for health. The active involvement of children as stakeholders in these school-based interventions led to outcomes related to self- efficacy and leadership skills that support healthy behaviours. Likewise, in the narrative review of six school-based hands-on culinary skill interventions also described under essential criterion 3.1 above, Muzaffar, Metcalfe and Fiese (125) noted in 2018 significant effects on outcomes such as cooking skills, cooking self-efficacy, cooking behavioural intentions, food-preparation frequency and knowledge in three studies reporting quantitative results. They also noted substantial effects reported qualitatively from all the six studies on, for example, cooking skills. NFSI Component 4: Creating a supportive environment The essential criteria included under component 4 are intended to create a school environment that is supportive of a healthy diet. These include healthy foods and beverages through school meals, food vendors and snack bars, as well as adequate facilities to prepare and eat those foods, positive messaging and absence of marketing of unhealthy foods, water and sanitation facilities, space for physical activities and a supportive environment which has positive role models and is free from bullying. Most of the evidence identified concerned the availability of foods and beverages in schools. Some evidence was identified for WASH, marketing of foods and beverages in schools and role-modelling of healthy behaviours. Few reviews were identified that looked at school-based nutrition programmes and positive messaging or adequate eating places or cooking facilities, while no reviews were identified that specifically focused on the impact on nutrition of clean and separate toilets for boys and girls, spaces for physical activity or action against bullying, stigmatization and discrimination. @ Ingimage 52 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE NFSI Component 4: Creating a supportive environment • Nine meta-analyses and 24 reviews were identified relevant to essential criterion 4.1 considering the promotion of healthy eating through meals and snacks available or provided in schools. Many of these reviewed current policies and standards implemented in countries. Three of the meta-analyses and five of the reviews were conducted in low- and middle-income countries or in disadvantaged populations. • School meal standards, competitive food and beverage policies and school food procurement policies had beneficial effects on food and beverage availability, purchase or consumption and/ or dietary intake, but the results were less conclusive about the effect on weight-related outcomes unless physical activity was simultaneously addressed. • School feeding programmes in low-income or disadvantaged settings improved weight gain, micronutrient status and energy intake, while participation in school breakfast programmes in high-income countries helped prevention of obesity. In the same settings, provision of fortified foods or beverages through schools improved micronutrient status, particularly anaemia reduction, and in some cases anthropometric outcomes. • Providing free or subsidized fruit and vegetables often increased consumption, including in one study after the scheme had finished. • Two meta-analyses and seven systematic reviews were identified relevant to essential criterion 4.2 considering messaging, incentivising desired behaviour or use of labelling to encourage healthy eating in the school, which resulted in improvements in outcomes related to diet, food purchases and preferences, with mixed results for weight-related outcomes. Incentives should not lead to increased food waste. • One review examined food and beverage marketing specifically in schools, relevant to essential criterion 4.3. Children are commonly exposed to marketing of unhealthy foods and beverages in and around schools through a variety of strategies, especially in high schools and areas with lower socioeconomic status. Restrictive policies or guidelines were not common and, where they did exist, were often breached. • One systematic review which examined eating places and cooking facilities in schools relevant to essential criteria 4.4 and 4.5, identified overcrowding in the cafeteria or lack of appropriate kitchen facilities and equipment as barriers to healthy school environments for eating. • Three systematic reviews were identified relevant to essential criterion 4.6, analysing aspects of access to safe drinking water relating to nutrition outcomes. They found positive effects on water consumption and mixed effects on SSB consumption and weight-related outcomes. • One meta-analysis and one systematic review relevant to essential criterion 4.7 examined water, sanitation and hygiene (WASH) in the school setting relating to nutrition outcomes. Adequate hand washing and sanitation reduced risk of diarrhoea, amongst other outcomes. • One meta-analysis and two systematic reviews discussed evidence for modelling healthy dietary behaviours relevant to essential criterion 4.11, focusing on improved intervention effectiveness to reduce the consumption of SSBs and other healthy eating behaviours. Modelling could be, for example, by parents, peers or teachers. 53RESULTS Essential criterion 4.1: School meals, food vendors and snack bars (if present) promote healthy eating No specific guidelines or recommendations exist from WHO regarding the quality or composition of foods and beverages in schools. However, the general guidance related to healthy diets and nutrient intake also applies to the school setting and recommends the establishment of standards to foster healthy dietary practices through ensuring the availability of healthy, safe and affordable food in preschools, schools, other public institutions and in the workplace (110). Four of the five keys to a healthy diet apply to school-age children and should be encouraged in the school environment: 1) eat a variety of foods, 2) eat plenty of fruit and vegetables, 3) eat moderate amounts of fats and oils, and 4) eat less salt and sugars (108). Furthermore, school foods can also help students to reach dietary goals for nutrients such as sodium and potassium. The WHO guideline on sodium recommends a reduction in sodium intake to control blood pressure in children and the WHO guideline on potassium suggests increased potassium intake from food to control blood pressure in children (142, 143). In areas where anaemia is 20% or higher in school age children, WHO recommends point-of-use fortification of foods with iron-containing micronutrient powders in children aged 2–12 years (144). WHO also recommends fortification of staple foods such as wheat flour, maize flour and corn meal with iron to prevent iron deficiency in populations, particularly vulnerable groups such as children and women (145, 146). The WHO School Policy Framework (147) notes that food served in schools should adhere to nutrition standards based on national or regional dietary guidelines. This document proposes that in countries where meals are provided at school, 30% of recommended nutrient intake for age is used as a reference value to plan the content of the meals served, which can also provide guidance for the provision of adequate serving sizes in school meals. The implementation plan for the recommendations of the commission on ending childhood obesity (28) reiterates that settings such as schools should be required to create healthy food environments by setting standards for foods that can be provided or sold in schools. Nine meta-analyses (36, 59, 67, 91, 115, 116, 148-150) and 24 reviews (35, 37-45, 76, 77, 79, 124, 131, 135, 136, 151-157) were identified relevant to promotion of healthy eating through school meals or snacks available or provided in schools. Among these, four meta-analyses (36, 67, 91, 115) and 17 reviews (35, 37-45, 79, 124, 131, 136, 151-153) considered the availability of healthy foods and beverages implemented through policy or other measures, whereas six meta-analyses (59, 67, 116, 148-150) and nine reviews (37, 44, 76, 77, 135, 154- 157) considered the provision of foods, beverages or snacks such as fruits and vegetables in schools. The four meta-analyses considering the availability of foods and beverages (36, 67, 91, 115) generally found beneficial effect on dietary outcomes, but less often on weight-related outcomes. In the 2018 systematic review and meta-analysis of 91 studies on different types of school food environment policies described under essential criterion 1.1, Micha et al. (36) found effects on various outcomes. From 29 studies of competitive food and beverage standards or policies (restricting, for example, SSBs, unhealthy snacks, foods or beverages high in calories or specific nutrients, and/or large portion sizes), positive effects were found on total consumption of SSBs ( 0.18 servings/day; 95% CI: -0.31, -0.05), total consumption of unhealthy snacks (-0.17 servings/day; 95% CI: -0.22, -0.13) and in-school consumption of unhealthy snacks (-0.05 servings/ day; 95% CI: 0.08, -0.02), with no effect found on in-school SSB consumption, total calorie intake or overweight/obesity. From 39 studies of school meal standards (mainly for lunch meals), positive effects were found on fruit consumption (+0.76 servings/day; 95% CI: +0.37, +1.16), total fat intake 54 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE (1.49%E; 95% CI: -2.42, -0.57), saturated fat intake (-0.93%E; 95% CI: -1.15, -0.70), sodium intake (170 mg/day; 95% CI: -242, -98) and BMI percentile (-1.0; 95% CI: -1.62, -0.39). Positive effects were also seen on increased milk consumption and carbohydrate intake in these studies, but no effect was seen on intakes of total calories, dietary fibre or whole grains, or on other adiposity and metabolic measures. They also studied the effect of direct or indirect fruit and vegetable provision, which is described below. In the 2018 meta-analysis of 30 studies aiming to increase vegetable intake in children 2-5 years of age described under essential criterion 3.1, 24 of which were conducted in preschool or child care settings, Nekitsing et al. (115) found a small-to- moderate effect (Hedge’s g = 0.40) from pooling results of 30 studies and a slightly higher effect (g: 0.42; 95% CI: 0.33, 0.51, P < 0.001) from pooling results of all 44 intervention arms. In subgroup analyses of intervention strategies, the studies using taste exposure had a significantly higher impact on intake (g: 0.79; 95% CI: 0.53, 1.05) than food service modifications alone (g: 0.30; 95% CI: 0.10, 0.50). In the 2019 systematic review and meta- analysis described under essential criterion 2.2, Nathan et al. (91) examined the effect of 10 school- or centre-based studies aiming to improve the quality of foods brought from home, e.g. lunch boxes. In addition to parental involvement, interventions included components such as physical resources (e.g. lunch packs, containers) and/or were linked to development of a policy and the communication of this policy to parents. A meta-analysis of four studies (two in schools and two in centres) found a moderate increase in vegetables brought from home (SMD: 0.40; 95% CI: 0.16, 0.64, P = 0.001) but not fruit. Sub- analyses by setting revealed that a significant @ Ingimage 55RESULTS increase remained in centre-based studies, but not in school-based studies. Mixed results were reported for this and other nutrition-related outcomes (e.g. provision and/or consumption of fruit and vegetables, discretionary foods and/or SSBs). Two multicomponent studies, conducted in a centre and school respectively, reported effects on weight-related outcomes. In the 2013 meta- analysis by Williams et al. (67) on the effect of 21 school diet and physical activity studies described under essential criteria 1.1 and 2.2, five of the studies focused on diet-related policies aimed at increasing availability of healthy foods and beverages or decreasing availability of unhealthy foods and beverages within schools (e.g. removing low nutrient and/or energy-dense foods, fried potato products, desserts and whole or 2% milk from cafeterias, ensuring fruit and vegetables are available in the cafeteria, etc.). The pooled results of these five studies, however, gave a non- significant decrease in BMI z-score. Among the 17 reviews that evaluated studies that addressed the availability of foods and beverages in or around schools, 10 reviews concerned school food and nutrition policies described under essential criterion 1.1 (35, 37-45), five reviews focused on other measures such as standards that may not have been implemented by policies (79, 124, 131, 151, 152) and two focused on softer measures such as nudging (136, 153). The 10 school food and nutrition policy studies described under essential criterion 1.1 generally found positive effects on outcomes related to food and beverage availability in schools and diet-related outcomes, including intake, consumption or purchase, while results were mostly mixed regarding the effect on weight- related outcomes. In the 2019 systematic review of 33 natural experiments on childhood obesity prevention and control, Bramante et al. (35) found that school-based policies focusing on both the food and beverage and physical activity environments (versus targeting only one of those environments) consistently showed improvements in BMI. Furthermore, three of four school-based studies that focused on the food and beverage environment showed reduced SSB consumption. These three studies evaluated state- or school-wide policies to decrease access to SSBs. One of three school-based studies that focused on the food and beverage environment showed increased fruit and vegetable consumption, as result of implementing a policy on the sale of competitive foods and beverages. In the 2014 systematic review of effects of 24 studies of state and district competitive food and beverage laws and policies, Chriqui, Pickel and Story (41) found that 15 of the interventions had outcomes in the expected direction. Such positive results were found in two out of four studies measuring weight or BMI, in six out of 11 measuring consumption, purchase or dietary intake, and in 11 out of 13 measuring availability and access. In the 2018 systematic review of 26 studies on competitive foods and beverage in schools in the United States, Sildén (42) found that competitive foods and beverages are pervasive in schools, but evidence was lacking that such products cause overweight or obesity in children. On the other hand, the author noted that there was a positive association between competitive food and beverages policies and weight-related and/or dietary outcomes, and “ Providing free or subsidized fruit and vegetables often increased consumption, including in one study after the scheme had finished” 56 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE provided examples of several cases where caloric intake and/or overweight/obesity measures had improved. In the 2015 systematic review of 37 studies of policy and built environment changes, Mayne, Auchincloss and Michael (39) included five school-based nutrition policy studies concerning restrictions on sugary foods and beverages or higher fat foods, and/or increases in availability of milk or fruit/vegetables. Four of five studies measuring dietary outcomes (e.g. energy density, consumption of low-nutrient- density food or SSBs, and calorie, fat or fruit and vegetable intakes) reported results in the expected directions, whereas one had mixed results. The only study measuring weight-related outcomes reported mixed results. Similarly, in the 2014 systematic review of 18 studies of school food environment policies or interventions, Driessen et al. (38) found that all but one involved changes to the availability of foods in school canteens and produced clear, positive changes in diet or purchases. Fifteen out of 16 studies measuring purchasing behaviour or dietary outcomes reported a significant intervention effect. Two out of four studies measuring BMI found positive effects, however both studies were uncontrolled and therefore it is unknown whether changes in BMI resulted from the policy or some other environmental change. The authors suggested that high-level policy changes at district, state or national level can positively impact the school food environment and improve children’s diets. In the 2019 Cochrane review of 58 studies, von Philipsborn et al. (44) conducted a subanalysis on seven studies of school nutrition standards and found that reduced SSB availability within the school was associated with reduced intake in children in five of the studies, three of which involved a policy-level measure. The authors noted an indication that reduced availability of SSBs in schools may increase compensatory sugary drink consumption outside school among older students, but may not affect body dissatisfaction or dieting behaviour. They also noted that one study of improved placement of plain milk in school cafeterias did not affect the selection and consumption of sugar-sweetened or plain milk. In another subanalysis of healthier selections in vending machines in workplaces and schools, the one study in schools reported significant effect on SSB intake. Likewise, in the 2017 systematic review of 36 school-based SSB reduction interventions in adolescents described under essential criteria 1.1 and 3.1, Vézina-Im et al. (45) noted that the 10 studies involving legislative measures or structural changes to the environment (e.g. banning SSBs from school cafeterias or replacing them with healthier alternatives) were the most effective. Nine of 10 such studies resulted in decreased consumption of SSBs. In the 2009 systematic review of 27 studies of school food and nutrition policies worldwide targeting the food environment (e.g. nutrition guidelines, regulation of food and beverage @ Ingimage 57RESULTS availability, price intervention), Jaime and Lock (37) reported that all three studies measuring fat intake found significant decreases in total fat or saturated fat, and both studies measuring fruit and vegetable intake showed a positive impact. Three out of four studies that examined the effect of policies on menu composition showed a significant decrease in total fat and saturated fat on the school menus. All four studies of interventions to increase fruit and vegetables offered at school lunch showed that guidelines led to increased fruit and vegetable availability. Both studies that measured the impact of restrictive nutrition policies on sales of food and beverages suggested a significant but limited decrease in the sales of banned foods, such as chips and SSBs; and both studies that measured the impact of reducing low-fat food prices showed significant increases in low-fat snacks and fruit and vegetable sales. Only one study reported effects on BMI, but found no difference after 1 year of follow-up. In 2015, Afshin et al. (40) reviewed the evidence for specific policies to improve dietary habits and reduce cardiovascular and metabolic risk factors, including 76 studies of school regulation and procurement policies. Among 31 studies to increase the availability of healthy foods and beverages — such as fruit and vegetables, low-fat snacks, milk, whole grain products or water — the interventions that aimed to increase fruit and vegetable intake appeared effective. As noted in essential criterion 1.1, the 26 studies on restrictions to unhealthy foods and beverages in schools through bans or nutrient standards for competitive foods appeared generally effective in reducing intake of unhealthy competitive foods and beverages and decreasing incidence and prevalence of overweight/obesity, and more so in governmental and school policies than in voluntary programmes. However, among 22 studies on the effect of standards based on food, nutrients or portion size for lunch and/or breakfast, results were inconsistent across studies and no consistent patterns were identified to explain the heterogeneous findings. The authors concluded that changes in school procurement policies appear effective for either increasing healthy or reducing unhealthy choices, while setting of nutrition standards has less consistent benefits. In the United States, Mansfield and Savaiano (43) in 2017 reviewed 31 studies of the U.S. National School Lunch Program nutrition standards under three different legislative eras and concluded that increasing access to healthy foods during school lunch improved students’ dietary intakes. Improved food consumption behaviours — including increased selection, intake and sales of healthy foods, and decreased plate waste — were reported in 14 of the 19 intervention and longitudinal observation studies. However, only two of the 12 one-time observation studies reported food consumption behaviours meeting target school nutrition standards. 58 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE @ Ingimage The five reviews which were not described under essential criterion 1.1 (because the status of policy was not clear) also found positive effects of school food standards or other measures to improve the availability of foods and beverages in schools on diet-related outcomes (79, 124, 131, 151, 152). In a 2014 systematic review of 34 food procurement policy and programme studies, Niebylski et al. (151) found all the 19 studies conducted in the school setting demonstrated increases in healthy food purchasing. Furthermore, many of the school interventions that included an additional education component were effective at increasing the intake of healthy foods and decreasing intakes of foods high in fat, sodium or sugar. This review also included two school- based studies that assessed health outcomes, and these found reductions in blood pressure and BMI. In the 2015 systematic review of 102 quantitative and qualitative studies of school architecture and design described under essential criterion 3.1, Frerichs et al. (124) found improved dietary behaviours, as measured by sales, self- reported data or plate waste, from 12 of 16 studies of increased accessibility of healthy items or decreased accessibility of less healthy items. On the other hand, studies on the effect of different foods in vending machines had mixed results. In the 2006 systematic review of 15 fruit and vegetable promotion interventions described under essential criteria 1.1, 2.2, 2.3 and 3.1, Knai et al. (79) reported six out of nine studies with increased exposure to fruit and vegetables in the school setting (e.g. canteens) and six out of 10 studies modifying lunches or snacks provided had significant positive results on consumption. In the 2019 systematic review of 29 various school-based studies in adolescents described under essential criterion 3.1, Calvert, Dempsey and Povey (131) noted that increased availability of healthy foods in school was a promising intervention component and improved healthy eating outcomes in all six studies with this component. In a 2008 systematic review of 22 school-based obesity prevention or reduction studies in China, Li et al. (152) found that 10 of the 13 studies that included a dietary modification component (e.g. increase vegetable, fruit, legumes; restrict SSBs; portion size control) reported on weight-related outcomes and all of these studies reported effectiveness on overweight/ obesity prevalence and/or BMI. Two systematic reviews looked at the set- up of school cafeterias rather than the content of the menus (136, 153). In the 2018 systematic review of 48 school cafeteria environment studies mentioned under essential criterion 3.1, Gordon, Dynan and Siegel (136) found that interventions that change the environment to favour fast and intuitive thinking for selecting healthier options were both more common and more effective than interventions that favour slow and cognitively-demanding choices through educating children. Using behavioural economics theory, they characterized school cafeterias and approaches to improving nutrition into “system 1” 59RESULTS characterized as “fast and intuitive thinking” (e.g. price changes, placement) and/or “system 2” characterized as “slow and cognitively demanding” (e.g. nutritional education, information posting), and assessed the impact on nutrition-related outcomes (e.g. BMI, consumption and/or selection of fruit, vegetables, sugary drinks, plain milk and/ or whole grains). The authors found that 89% of 27 “system 1” studies were successful, compared to 62% of 13 mixed “system 1” and “system 2” studies and 50% of eight “system 2” interventions. “System 1” studies were significantly more effective than “system 2” studies (P = 0.033). In another systematic review of 12 studies on the effects of choice architectural nudging interventions on adolescent vegetable consumption where diet- related behaviour is influenced without limiting choice sets or making alternatives costlier (e.g. provision of free vegetables, serving styles and changes in the physical environment), Nørnberg et al. (153) in 2016 found overall inconclusive results. They noted that providing free vegetables resulted in more favourable attitudes towards vegetable consumption, and offering a greater variety of vegetables resulted in increased vegetable consumption. Six meta-analyses (59, 67, 116, 148-150) and nine reviews (37, 44, 76, 77, 135, 154-157) considered the provision of foods, beverages or snacks such as fruit and vegetables in schools. Among these, two meta-analyses (67, 148) and four reviews (76, 154-156) analysed school feeding programmes, two meta-analyses (149, 150) and one systematic review (157) examined the effect of micronutrient fortification of foods or beverages provided in schools, while three meta-analyses (59, 116) and four systematic reviews (37, 44, 77, 135) examined fruit and vegetable schemes in schools. Regarding school feeding programmes, one meta-analysis focused on disadvantaged children in low-resource settings in the context of undernutrition (148), while another focused on school lunch and breakfast programmes in the context of obesity reduction (67). Both meta- analyses showed mixed results. In a 2007 Cochrane review and meta-analysis of 18 studies in schools in disadvantaged settings in low- and higher- and lower-income countries, Kristjansson et al. (148) found that school meals may have some small benefits for nutrition-related outcomes. Evidence from the highest quality studies (RCTs) from low- income countries found that children who were fed at school gained more weight than controls over an average study duration of 19 months (+0.39 kg; 95% CI: 0.11, 0.67). In the lower quality studies (controlled before and after trials), children who were fed at school also gained more weight than controls over an average study duration of 11.3 months (0.71 kg; 95% CI: 0.48, 0.95). For height, results were mixed; in RCTs, differences in gains were significant only for younger children (+0.40cm; 95% CI: 0.03, 0.77), whereas the results from the controlled before and after trials were large and significant overall (+1.43cm; 95% CI: 0.46, 2.41). Positive results were also noted for school attendance and educational and cognitive outcomes. Results from higher-income countries were mixed, but generally positive. In the 2013 systematic review and meta-analysis on the effect of school diet and physical activity polices on child obesity in the United States described above, Williams et al. (67) found that participation in the U.S. National School Lunch Program targeted at children at risk of undernutrition was associated with a non-significant rise in BMI z-score, while participation in the School Breakfast Program, also targeted at children at risk of undernutrition, was associated with a significantly lower BMI z-score (-0.08; 95% CI: -0.143, -0.017). Of the four reviews on school feeding programmes, three were conducted in low- and middle-income countries (154-156) and one was conducted in a high-income country but with many studies targeted at disadvantaged children (76). In 2015 Watkins et al. (154) did a literature review of 60 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 10 school feeding programme studies in primary schools in low- and middle-income countries in Africa, Asia and Latin America. Among the 10 studies measuring anthropometric outcomes, small but significant improvements were seen in height/ stunting in seven of the studies and on weight/ underweight in six of the studies. Nearly all the food supplementation studies that reported significant height and weight gains included an animal-based product not usually included in school feeding programmes in low-income countries. Spillover benefits were observed for younger siblings, indicating that school feeding could have an important role in promoting the health of the next generation. Incorporating findings from other related systematic reviews, the authors concluded that the largest effects were seen for in-school meals on height/stunting, weight/underweight and for multiple micronutrient fortification on iron status, anaemia/haemoglobin, vitamin A status and zinc status. Take-home rations, multiple micronutrient fortification or powder had some effect on height/stunting and weight/underweight, whereas in-school meals and micronutrient powders had some effect on various micronutrient- related outcomes. In 2008 Adelman, Gilligan and Lehrer (155) reviewed 20 studies of food-for-education programmes (e.g. breakfast, mid-day meal, micronutrient fortified meals/snacks, animal- source foods) implemented in primary schools in low- and middle-income countries in Africa, Asia and Latin America. Three of four studies measuring consumption-related outcomes found that school feeding increased caloric intake among school-aged children, particularly when the baseline was well below their age- or weight- recommended consumption level. Furthermore, three of four studies found that increasing access W FP /R au l S ae nz 61RESULTS @ In gi m ag e to calories over a sufficient period increased body size or composition (e.g. height, weight, BMI, mid-upper arm circumference (MUAC)). However, provision of micronutrient-rich foods or deworming seemed to have contributed to improvements in some outcomes (e.g. height, MUAC). Improvements were typically small, but the authors note that the effect may have been mitigated by increased activity levels. Iron status was improved as a result of fortified school meals in three out of four studies, vitamin B12 status was improved as result of meat-based meals in one study, vitamin A was improved as result of vitamin A fortification in one out of two studies, prevalence of iodine deficiency disorders (IDD) was reduced as result of iodine fortification in one study. Similar conclusions were drawn by Jomaa, McDonnell and Probart (156) in a 2011 review of 13 school feeding programmes in low- and middle-income countries, including meals based on staples, snacks, fortified snacks and drinks. The authors found positive effects on energy intake, micronutrient status, school enrolment and attendance of children participating in school feeding programmes compared to non-participants. As described in essential criterion 1.1, Colley et al. in 2019 (76) found that multicomponent school food programmes in Canada — many of which were implemented in indigenous communities or targeted at disadvantaged children — were positively associated with children’s development of nutrition knowledge in four out of six studies which measured it, with changes in dietary behaviour and food preferences in all seven studies measuring this and in intake of healthy foods in six out of eight studies which assessed intakes. The effects of micronutrient fortification of foods or beverages provided in schools were reviewed in two meta-analyses (149, 150) and three systematic reviews (154, 155, 157), which all found beneficial effects on micronutrient status and sometimes on anthropometric outcomes. In the 2017 Cochrane review and meta-analysis by De-Regil, Jefferds and Peña-Rosas (149) @ Ingimage 62 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE underpinning the WHO guideline (144) on point- of-use fortification with micronutrient powders (MNP) in preschool- and school-age children, eight out of 13 studies included were conducted in the school setting among low- and middle-income populations in Asia, Africa and Latin America. The overall results from the meta-analysis found lower risk of anaemia (prevalence ratio (PR): 0.66; 95% CI: 0.49, 0.88) and iron deficiency (PR: 0.35; 95% CI: 0.27, 0.47) as well as higher haemoglobin (MD: +3.37 g/L; 95% CI: +0.94, +5.80) among preschool- and school-age children receiving iron-containing MNP for point-of-use fortification of foods. In a 2015 systematic review and meta-analysis of multiple micronutrient-fortified non-dairy beverage interventions in school-aged children in low- and middle-income countries, Aaron, Dror and Yang (150) found positive effects of multiple micronutrient-fortified beverages on haemoglobin (+2.76 g/L; 95% CI: +1.19, +4.33) and serum ferritin (+15.42 pmol/L; 95% CI: +5.73, +25.12); as well as reduced risk of anaemia (RR: 0.58; 95% CI: 0.29, 0.88), iron deficiency (RR: 0.34; 95% CI: 0.21, 0.55) and iron deficiency anaemia (RR: 0.17; 95% CI: 0.06, 0.53). Furthermore, they found a moderate but significant effect for weight gain in the intervention group compared with the control (+0.30 kg; 95% CI: +0.01, +0.58), but neither height nor weight-for-age or height-for-age Z-scores were significantly different between groups. In a 2011 systematic review of 12 studies to examine the effects of multiple micronutrient fortification in the school setting, Best et al. (157) found that such fortification, with the exception of zinc, improved micronutrient status in 10 of 11 intervention groups. However, the impacts of multiple micronutrient fortification on functional health outcomes, such as growth, morbidity and cognitive development, was inconsistent. Four of seven studies reported a significant beneficial effect on weight gain, increased BMI or MUAC, but only two showed an effect on height. Four of seven studies showed some beneficial effects on children’s health including reduced incidence or duration of diarrhoea and respiratory diseases. The impacts on cognitive performance were inconsistent and varied between cognitive domains (e.g. memory, fluid intelligence, school performance), whereas the impact on working memory across studies was more positive. As described above, the 2015 literature review by Watkins et al. (154) noted the largest effects for, among other things, multiple micronutrient fortification on iron status, anaemia/haemoglobin, vitamin A status and zinc status. Multiple micronutrient fortification or powder also had some effect on height/ stunting and weight/underweight, whereas micronutrient powders had some effect on various micronutrient-related outcomes. Also described above, Adelman, Gilligan and Lehrer (155) noted that iron status was improved as result of fortified school meals in three out of four studies, vitamin A status was improved as result of vitamin A fortification in one out of two studies, and prevalence of IDD was reduced as result of iodine fortification in one study. Finally, the effect of providing fruit and vegetables through schools was reviewed in three meta-analyses (36, 44, 59, 116) and four systematic reviews (37, 44, 77, 135). In their 2018 systematic review and meta-analysis of 91 school food environment policies mentioned above, Micha et al. (36) found from 40 studies of direct or indirect provision of healthy food policies (largely targeting fruit and vegetables) that positive effects were seen on consumption of fruits (+0.27 servings/ day; 95% CI: +0.17, +0.36), fruits and vegetables (+0.28 servings/day; 95% CI: +0.17, +0.40) and vegetables (+0.04 servings/day; 95% CI: +0.01, +0.08). No effect was seen on total calorie intake, water consumption or any adiposity or metabolic measures from the provision policies. In a 2012 systematic review and meta-analysis of 27 school- based interventions to improve daily fruit and vegetable intake in children aged 5–12 years by 63RESULTS Evans et al. (59), 10 of the studies included free school fruit and vegetable distribution or fruit and vegetable subscription programmes either alone or along with other components. While it is not possible to separate the effect of these various components from other changes, the meta-analysis of 21 studies found that fruit and vegetable intake increased (+0.25 portions/day; 95% CI: +0.06, +0.43), mainly due to increases in fruit consumption. Separate analyses for fruit and vegetable consumption showed a significant increase in fruit (+0.24 portions/day; 95% CI: +0.05, +0.43) but a non-significant increase in vegetables. On the other hand, a 2011 systematic review and meta-analysis by Delgado-Noguera et al. (116) of 19 studies on primary school interventions to promote fruit and vegetable consumption found no significant effect of free or subsidized fruit and vegetables in a pooled analysis of two of three such programmes studied with a RCT design. In the systematic review of 27 studies of school food and nutrition policies by Jaime and Lock (37) described above, four studies measuring the impact of providing free or subsidized fruit and vegetables showed a statistically significant increase in fruit and vegetable consumption. In the 2010 systematic review of 42 European school-based interventions to promote a healthy diet described under essential criterion 3.1, Van Cauwenberghe et al. (135) found that all five studies on fruit and vegetable provision programmes increased fruit and vegetable consumption. However, only one study showed a lasting effect. Similarly, in a 2008 systematic review of 30 school fruit and vegetable programmes, 10 of which included free or subsidized fruit and vegetables amongst other components (e.g. classroom activities, school policies, involvement of teachers, peer leaders, parents, community and school food service), de Sa and Lock (77) found that 70% of studies reported positive associations between interventions and nutrition outcome. Two of the national programmes providing free or subsidized fruit and/or vegetables had been evaluated and had shown significant increase consumption during the scheme, with one of them also showing sustained increases after the scheme finished. In the 2019 Cochrane review of interventions to reduce SSB consumption mentioned above, von Philipsborn et al. (44) found that both studies measuring impact of fruit and vegetable provision on SSB consumption reported a significant effect, however, the overall quality of evidence was judged as very low certainty. 64 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Essential criterion 4.2: Positive messaging towards nutrition and active living Two meta-analyses (92, 115) and seven reviews (35, 44, 105, 124, 136, 158, 159) were identified that considered implicit or explicit messaging, incentivising desired behaviour or specific labelling to encourage healthy eating. Many other reviews included studies with an element of positive messaging toward nutrition and active living — often as part of nutrition education activities — but the review authors seldom analysed the effect of this specifically. In the 2015 systematic review and meta- analysis by Dudley, Cotton and Peralta (92) of teaching strategies to promote healthy eating in primary school children described under essential criteria 2.2 and 3.1, four of the included studies used contingent reinforcement (i.e. reward for behaviour), and all four of the studies reported statistically significant results. Based on the two studies that reported effect sizes, the average effect size for contingent reinforcement in promoting fruit and vegetable consumption or preference was Md = 1.34. Although this is a large effect size (> 0.4 is considered a large effect), more studies are needed to determine the average effect size with less variance. In the 2018 meta-analysis of 30 studies aiming to increase vegetable intake in children 2–5 years of age described under essential criteria 3.1 and 4.1, 24 of which were conducted in preschool or child care settings, Nekitsing et al. (115) found a small-to-moderate effect (Hedge’s g = 0.40) from pooling results of 30 studies and a slightly higher effect (g: 0.42; 95% CI: 0.33, 0.51, P < 0.001) from pooling results of all 44 intervention arms. In subgroup analyses of intervention strategies, the studies using taste exposure had a significantly higher impact on intake (g: 0.79; 95% CI: 0.53, 1.05) than food service modifications alone (g: 0.30; 95% CI: 0.10, 0.50), especially when coupled @ In gi m ag e 65RESULTS with reward and modelling in studies (g: 1.08; 95% CI: 0.50, 1.66), with the largest effect size observed for unfamiliar/disliked vegetables. In the 2018 systematic review of 48 school cafeteria environment studies described under essential criteria 3.1 and 4.1, Gordon, Dynan and Siegel (136) found that messaging that support fast and intuitive thinking for selecting healthier options appeared effective. Nine of 11 studies using incentives were successful, as were all five studies using emoticons, whereas one out of three studies using messaging was effective. However, the authors noted that as the latter was a single component study, this demonstrates that messaging alone may significantly increase consumption of vegetables. In the 2019 systematic review of 33 natural experiments for childhood obesity prevention and control described under essential criteria 1.1 and 4.1, Bramante et al (35) found that the four studies of multicomponent interventions in the school setting focusing on food/beverage and physical activity/built environments had favourable BMI outcomes. Two of these four studies that also focused on healthy messaging achieved the largest improvement in BMI. Similarly, in 2017, Black, Matvienko-Sikar and Kearney (105) also noted that effective school-based programmes to improve children’s diet and health incorporated role-models including peers, teachers and heroic figures, as well as rewards for healthy eating and increased access to healthy foods in their systematic review of 39 school-based diet and physical activity interventions described under essential criterion 2.2. In the 2019 Cochrane review of interventions to reduce SSB consumption described under essential criterion 4.1, von Philipsborn et al. (44) found low certainty evidence that prizes and tokens or emoticons may be effective. All three studies reported that small prizes and token rewards for the selection of plain milk in elementary school cafeterias were associated with decreased selection of sugar-sweetened milk, though the effect on total milk selection varied. They also identified one study that found a positive effect of emoticon (smileys) labels on plain milk and small prizes in reward for selecting such milk over sugar-sweetened milk. The authors noted that this may drive up food waste if children choose the plain milk but do not drink it. In a 2011 systematic review of 30 school-based studies, Jensen et al. (158) found that economic incentives may influence school children’s choice of foods, snacks and beverages. In particular, direct price incentives were effective for altering consumption in school cafeterias or from vending machines and for increasing fruit/vegetable consumption in schools, in contexts where foods or beverages are offered (for sale) in schools and students (typically 10–12 years or older) bring money to school to buy some of these items. The authors were not able to assess the effectiveness of small rewards — such as pens, other small items or participation in a raffle — for healthier food choices, since such interventions were usually implemented in combination with extensive educational activities. In 2017 Sacco et al. (159) conducted a systematic review of 11 real-world and artificial studies on the effect of menu labelling on food choice of children aged between 6 and 18 years. The studies of hypothetical food purchases in artificial environments suggest that menu labelling may be efficacious in reducing calories purchased for or by children and adolescents. The real-world studies were less supportive, although two of the three school-based studies suggested a potential impact of utilizing menu labelling in school cafeterias. In the 2015 systematic review of 102 quantitative and qualitative studies of school architecture and design described under essential criteria 3.1 and 4.1, Frerichs et al. (124) found evidence from analysis of 15 studies of educational signage, wayfinding and marketing, that these approaches are acceptable strategies to students, parents and teachers for addressing healthy eating in schools, though results were mixed and often not sustained. 66 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Essential criterion 4.3: Absence of marketing of foods and beverages at school WHO developed a set of 12 recommendations, endorsed by the Sixty-third World Health Assembly in 2010, aimed at reducing the impact of marketing foods high in saturated fats, trans- fatty acids, free sugars or salt to children (160). The Recommendations recognize that “the nutritional well-being of children within schools should be paramount and the foundation stone for children’s well-being at this formative age” and recommends that such “settings where children gather should be free from all forms of marketing of foods high in saturated fats, trans-fatty acids, free sugars, or salt.” (160) This includes marketing of foods and beverages as well as promotions or sponsorships, during or outside of school hours (161). Several reviews have demonstrated how marketing of unhealthy foods and beverages to children are widespread across the world and are influencing their food preferences, purchase requests and/or consumption patterns (162-166). For this review of reviews, however, only one review was identified that specifically examined marketing in schools. This 2017 review by Velazquez, Black and Potvin Kent (167) of 27 studies of food and beverage marketing in schools found that students are commonly exposed to food and beverage marketing through a variety of strategies, especially in secondary schools and schools with lower socioeconomic status. Marketing exposure is often associated with food purchasing and consumption, especially for food items of low nutritional quality (e.g. salty snacks, sweet confectionery, SSBs). Furthermore, many schools do not have, or are not subject to, laws, policies or guidelines regulating such marketing. Two studies evaluating compliance with laws or policies restricting school food and beverage marketing found numerous instances of non-compliant marketing. Essential criteria 4.4: Access to an adequate eating place and 4.5: Adequate school cooking facilities The WHO School Policy Framework (147) provides some suggestions regarding food service areas. School food service areas should be friendly, welcoming and consist of multiple points of service in cafeterias to avoid long queues and give students time to choose foods. In addition, they should provide a clean eating area and food ought to be provided in a non-stigmatizing manner, especially to students participating in free or reduced-price meal programmes. Maintaining adequate eating and cooking spaces is also an important factor in prevention of foodborne illness and diarrhoeal disease in schoolchildren. This is particularly important because diarrhoeal disease can exacerbate undernutrition, while children who are malnourished are more susceptible to infection (168). Cooking spaces should include all the necessary facilities to adhere to @ In gi m ag e 67RESULTS WHO’s Five Keys to Safer Food: 1) Keep clean, 2) Separate raw and cooked, 3) Cook thoroughly, 4) Keep food at safe temperatures, and 5) Use safe water and raw materials (108). Ensuring the proper handling of food is an essential step in the provision of safe and nutritious foods to children in schools. One review was identified that specifically examined studies related to eating places and cooking facilities in schools. In this 2015 systematic review of 102 quantitative and qualitative studies of school architecture and design described under essential criteria 3.1, 4.1 and 4.2, Frerichs et al. (124) found evidence that a range of physical factors contribute to creating supportive school environments for healthy eating. Analysis of studies of the dining or kitchen domain found that overcrowding in the cafeteria contributed to students’ negative social experiences, while lack of appropriate kitchen facilities and equipment hindered schools’ ability to provide meals with high nutritional quality and appeal. Essential Criterion 4.6: Access to safe drinking water Access to safe water has been a high priority on the global health and development agenda for years, first in the Millennium Development Goals and now as one of the Sustainable Development Goals (SDGs). Goal six of the SDGs is to ensure access to water and sanitation to all, with a specific target (6.1) for universal and equitable access to safe and affordable drinking water (169). Given the significant time children spend at school, it is important that they have access to adequate sanitation and safe drinking water. The WHO WASH standards for schools in low-cost settings highlight some of the positive effects including reduced disease burden, more effective learning, gender equity and an opportunity to learn and practice lifelong hygiene behaviours (170). Three systematic reviews were identified that analysed aspects of access to safe drinking water relating to nutrition outcomes (44, 124, 171). In a 2012 review of 47 water and sanitation studies in schools, Jasper, Le and Bartram (171) identified a subset of 11 studies focused on access to clean drinking water. Of the 11 studies, three noted a statistically significant increase in water consumption when children had free access to water in schools. One of the RCTs included from Europe reported a 31% reduction in the risk of overweight associated with providing drinking water and education in schools. In the 2019 Cochrane review of interventions to reduce SSB consumption described under essential criteria 4.1 and 4.2, von Philipsborn et al. (44) found four studies that improved availability of drinking water in schools as part of their school nutrition standards. Two of these studies reported significant effect on total intake of SSBs or sugar-sweetened milk, though there was indication that improved water availability may decrease total milk selection. Furthermore, two out of three studies that also measured weight-related outcomes reported significant beneficial effects. In the 2015 systematic review of 102 quantitative and qualitative studies of school architecture and design described in essential criteria 3.1, 4.1, 4.2, 4.4 and 4.5, Frerichs et al. (124) found that six of nine studies on water access found improvements in water intake, but not in terms of reduced SSB consumption. @ In gi m ag e 68 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE Essential criterion 4.7: Promotion of safe hygiene and sanitary behaviour One of the guidelines in the WHO standards for WASH in low-cost school settings pertains to hygiene promotion (170). The correct use and maintenance of water and sanitation facilities is ensured through sustained hygiene promotion and used as resources for improving hygiene behaviours. Hygiene education ought to be provided to teachers and schoolchildren, and positive hygiene behaviours should be systematically promoted throughout the school setting through clear rules and the participation of school staff, students and parents. WHO also recently released guidelines on sanitation and health, where schools are mentioned as an important setting where access to safe toilets should be ensured, and the rationale provided includes nutritional outcomes (172). UNICEF supports efforts to provide every child with the benefits of clean water for drinking and washing, dignity and safety through ample toilets and washstands (separated for girls and boys), education for good hygiene and healthy school environments through safe waste disposal. UNICEF finds that providing schoolchildren with access to drinking water, sanitation facilities and hygiene education is associated with educational and health benefits for both schoolchildren and the greater community (173). One meta-analysis (174) and one systematic review (171) that analysed the effectiveness of interventions to promote good sanitation and hygiene in the school setting in relation to nutrition-related outcomes were identified. In a 2015 Cochrane review and meta-analysis of 22 RCT studies, of which 12 were conducted in child day-care centres or schools, Ejemot-Nwadiaro et al. (174) found that handwashing promotion prevented about one third of diarrhoea episodes in these settings (RR: 0.70; 95% CI: 0.58, 0.85). Most of these studies were from high-income countries, but the authors suggest that the intervention may also prevent a similar burden of diarrhoea in middle- and low-income countries based on the result of two trials from urban areas in such countries (RR: 0.66; 95% CI: 0.43, 0.99). In the 2012 systematic review of 47 water and sanitation interventions in schools described under essential criterion 4.6 above, Jasper, Le and Bartram (171) found that two of three RCT studies aimed at improving handwashing and three of six studies providing water for both drinking and handwashing reported reduced illness and absenteeism as a result of the intervention. Furthermore, scarce supplies (e.g. water, soap, paper towels, etc.) were commonly reported to be associated with less handwashing. Essential criterion 4.8: Availability of clean and separate toilets, for boys and girls WHO standards for school water and sanitation include a guideline pertaining to toilet facilities, indicating that they should be private, secure, accessible, clean and culturally appropriate (170). Under this guideline, it is advised that male and female toilets be completely separated. While reviews exist that consider separation of toilet facilities in schools or adequate facilities for females, no review was identified that considered outcomes relevant to nutrition. Thus, no reviews were identified for essential criterion 4.8 (Availability of clean and separate toilets, for boys and girls). @ In gi m ag e 69RESULTS Essential criterion 4.9: Opportunity for all age groups to access space and school sporting facilities for physical activity in and outside of curriculum WHO recommends that adequate facilities are available on school premises and in public spaces for physical activity during recreational time for all children (28, 137). Two reviews examined the effects of space for physical activities, within and outside of the school curriculum. While reviews exist that consider the built environment for physical activity in school, these were not conducted in relation to school-based nutrition programmes. Thus, no reviews were identified for essential criterion 4.9 (Opportunity for all age groups to access space and school sporting facilities for physical activity in and outside of curriculum). Essential criterion 4.10: Affirmative action against bullying, stigmatization and discrimination Interventions to reduce bullying, stigmatization and discrimination are important for improving the school environment and student well-being, and many reviews have been conducted on these issues. However, no reviews were identified that examined such interventions in the context of diet or nutrition. Thus, no reviews were identified for essential criterion 4.10 (Affirmative action against bullying, stigmatization and discrimination). Essential criterion 4.11: School staff as role-models in encouraging healthy eating and healthy lifestyle While many of the reviews included school staff and teachers in the implementation of health interventions, few reviews specifically examined the effectiveness of using school staff as role-models. Most evidence exists in demonstrating the influential role of parents and families in modelling healthy lifestyle, as discussed under essential criterion 2.2. One meta-analysis (117) and two systematic reviews (99, 105) were identified that discussed evidence for modelling healthy dietary behaviours improving intervention effectiveness. In the 2017 systematic review and meta-analysis of 40 interventions to reduce SSB consumption and obesity – 15 of which were school-based — by Vargas-Garcia et al. (117) described under essential criterion 3.1, there was evidence that modelling or demonstrating the behaviour helped to reduce SSB intake in children. Univariate meta-regressions showed that modelling was associated with greater effectiveness in reducing intakes across all age groups (-124 mL/day; 95% CI: -221, -27) and in children specifically (-173 mL/day; 95% CI: -315, -31). In the 2012 systematic review of 12 obesity prevention studies in children aged 4–6 years described under essential criteria 2.2 and 3.3, Nixon et al. (99) found that interventions with the highest levels of success involved modelling healthy eating and physical activity. Although the study interventions focused on parental modelling of healthy diet behaviours, it is suggested that teachers could also be influential role-models in shaping young children’s perceptions of healthy living. Similarly, Black, Matvienko-Sikar and Kearney in 2017 (105) also noted that effective school-based programmes to improve children’s diet and health incorporated role-models including peers, teachers and heroic figures, as well as rewards for healthy eating and increased access to healthy foods, in their systematic review of 39 school-based diet and physical activity interventions described under essential criteria 2.2 and 4.2. NFSI component 5: Providing supportive school nutrition and health services NFSI component 5 concerns the services that schools may provide to support nutrition and health, in relation to detecting nutrition problems and as a delivery channel for nutrition interventions. Some evidence was identified that reviewed growth monitoring and screening for overweight and obesity. Some evidence was also identified for nutrition interventions delivered through school health services, such as micronutrient supplementation, deworming and counselling for overweight and obesity. 70 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE NFSI component 5: Providing supportive school nutrition and health services • One meta-analysis and one review examined regular monitoring and feedback on children’s growth relevant to essential criteria 5.1 and 5.2. • School-based weight monitoring and guidance was found to be effective in treatment of overweight and obesity in children. School-based screening programmes were also found to be useful in obesity prevention programmes, reporting positive effects in both children and their families. • Three meta-analyses were identified relevant to essential criterion 5.3 examining the impact of delivering nutrition interventions through school health services. Two of these were conducted in low- and middle-income countries. • School-based delivery of micronutrient supplementation had positive effects on micronutrient status, including anaemia, iron and zinc status. • School-based deworming had positive effects on haemoglobin and anaemia, especially when combined with hygiene promotion and/or co-administration of iron and retinol. • Counselling by school health nurses as part of school-based obesity prevention and treatment reduced BMI in children. Essential criteria 5.1: Regular monitoring of children’s growth development and 5.2: Effective feedback system for parents and children of findings of the regular monitoring Regular monitoring of children’s growth and development using tools such as the WHO Growth Reference for children and adolescents 5–19 years (175) can help identify underweight, overweight and obesity, as well as other growth-related conditions. One meta-analysis (56) and one review (176) were identified that examined the impact of school-based growth monitoring and BMI screening programmes. In the 2017 systematic review and meta-analysis of 76 school-based obesity treatment and prevention studies in China described under essential criteria 1.1 and 3.2, Feng et al. (56) found that multicomponent interventions were more successful than single component interventions. All four of the studies combining health education with weight management, which included weight monitoring and guidance among overweight and obese children, were effective in improving weight- related outcomes. In a 2015 review of 21 mandatory, school-based BMI screening programmes in the United States, Ruggieri and Bass (176) found that screening programmes can be valuable as part of school-based obesity prevention programmes and promotion of healthy lifestyles in schools. Despite confidentiality concerns shared by parents, school nurses and school administrators, most parents were supportive, provided that respect for student privacy was maintained at all levels. BMI screening and referral lead to increased follow-up action by parents in some studies, whereas other studies reported that only a small percentage of parents actually took action. Other reported outcomes were reduced use of diet-pills, encouraging a healthy breakfast, limiting family consumption of unhealthy foods and beverages, decreasing family meals in fast food restaurants, limiting unhealthy fats for meals cooked at home, limiting children’s screen time, giving more time for physical activity and reducing use of 71RESULTS physical activity as punishment. One study reported stabilization of BMI over the years of the programme. Resources that are required — for staff, measurement tools, obtaining parental permission and reporting data — were reduced by introducing biennial rather than annual screening. Essential criterion 5.3: A supportive school health service, including a referral system School health services, a key feature of the WHO Health Promoting Schools approach, may provide a range of supportive nutrition interventions, such as micronutrient supplements or anthelminthic treatment, depending on the country context. WHO recommends iron or iron and folic acid supplementation for the prevention of anaemia and iron deficiency in areas where anaemia is a public health problem. Daily supplementation is recommended for school-age children (177) and adolescent girls (178) in settings where anaemia prevalence is 40% or higher. In settings where anaemia prevalence is 20% or higher, intermittent (e.g. weekly) supplementation is recommended for both school-age children (179) and adolescent girls (180). The guidelines for school-age children (5–12 years) acknowledge schools as an appropriate delivery platform and entry point to reach that age group, however, there also needs to be consideration of how to reach children outside of the school system. The guidelines pertinent to adolescent girls also make reference to schools as a potential delivery system, but this is less specific, probably because they cover adult women as well. WHO also recommends anthelminthic treatment (deworming) as a public health intervention for school-age children and adolescent girls living in areas where the baseline prevalence of any soil-transmitted infection is 20% or more among these target groups, in order to reduce the worm burden of soil- transmitted helminth infection (181). School-based programmes are considered particularly feasible, as it is easy to deliver medicines through teaching staff, with estimated costs varying from USD 0.03 to 0.13 per child per year (182). Traditionally, many countries have used school-based deworming programmes, which should be implemented along with more long-term solutions to prevent soil-transmitted helminth infections such as improvements in water, sanitation and hygiene. Supportive school health services extend beyond deworming and micronutrient supplementation programmes. School health services are involved in regular growth monitoring, can provide counselling on nutrition and diet and make referrals as needed. Three meta-analyses and one review were identified that examined nutrition-related school health services. Two of the meta-analyses reviewed the provision of micronutrient supplements in schools (183, 184), one meta-analysis reviewed school-based deworming (184) and one meta-analysis (88) and one review (132) examined nutrition counselling or education by health workers. Regarding the provision of micronutrient supplements in schools, Salam et al. in 2016 (183) reviewed 30 studies of micronutrient supplementation “ School-based weight monitoring and guidance was found to be effective in treatment of overweight and obesity in children. School-based screening programmes were also found to be useful in obesity prevention programmes, reporting positive effects in both children and their families” 72 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE in adolescents in low- and middle-income countries and found a significant decrease in anaemia prevalence (RR: 0.69; 95% CI: 0.62, 0.76). Further subgroup analysis showed that school- based delivery significantly reduced anaemia (RR: 0.67; 95% CI: 0.60, 0.74). School-based delivery of iron or iron/folic acid supplementation (alone or in combination with other micronutrients) was also associated with improved serum haemoglobin (MD: +1.94 g/dL; 95% CI: +1.48, +2.41), ferritin (MD: +3.80 mcg/L; 95% CI: +2.00, +5.59) and iron (MD: +6.97 mmol/L; 95% CI: +0.19, +13.76). Zinc supplementation also improved serum zinc concentrations (MD: +0.96 mcg/dL; 95% CI: +0.81, +1.12). Furthermore, in the 2018 systematic review and meta-analysis of eight school-based deworming programmes described immediately below, Girum and Wasie (184) noted that micronutrient supplementation may enhance the effect of deworming. Regarding school-based deworming, Girum and Wasie (184) did in 2018 a systematic review and meta-analysis of eight school-based deworming programmes involving 1 005 239 children. They found an overall improvement in haemoglobin after deworming, with best impact on anaemia rates in studies that combined deworming with hygiene promotion and/or co-administration of iron and retinol. Meta-analysis of five studies found an overall improvement in haemoglobin after deworming (+1.62 g/dL; 95% CI: +1.01, +2.25). On nutrition counselling or education delivered by health workers in the school setting, the 2016 systematic review and meta-analysis of 11 childhood obesity prevention or treatment interventions delivered by school health nurses described under essential criterion 2.2 by Schroeder, Travers and Smaldone (88) found positive effects of involving school health nurses in anthropometric measurement and additional activities such as student education and counselling, involvement of parents, staff education or physical activity. The pooled results showed small but significant changes in BMI (-0.48 kg/m2; 95% CI: -0.84, -0.12), BMI z-score (-0.10; 95% CI: -0.15, -0.05) and BMI percentile (-0.41; 95% CI: -0.60, -0.21). Although the findings suggest that school health nurses can play a role in implementing successful school-based health programmes, the authors highlight several barriers to involving them including lack of confidence in counselling methods and understaffing due to budgetary constraints. In contrast, the USDA (132) in 2012 in examining the effect of different types of educators as part of the series of systematic reviews on nutrition education described in essential criterion 3.1, noted that the changes in dietary intake among school children were significantly greater in the teacher-educated group compared to the nutritionist-educated group. However, this was based on assessment of one single study, which was considered insufficient to draw conclusions. WHO/ Yoshi Shimizu 73RESULTS DISCUSSION AND CONCLUSIONS Summary of findings T he purpose of this review was to summarize the evidence base that underpins the NFSI framework. A strong evidence base to support many specific criteria of the NFSI was identified. These included the importance of having nutrition-related school policies, involving parents in school-based nutrition interventions, implementing effective nutrition and physical activity education, creating healthy school food environments and ensuring school-based monitoring of children’s growth. However, meta- analyses related to other criteria (e.g. the role of training staff in nutrition and physical activity, the role of staff as role-models of healthy behaviour, the need to prevent bullying) do not yet exist. Findings from this review suggest that comprehensive school nutrition policies (NFSI Component 1) using multiple components and approaches (e.g. diet, physical activity, educational interventions, environmental changes) are associated with positive weight-related, dietary and other outcomes among school children. The majority of the reviews considering multicomponent programmes combining diet and physical activity interventions found this combination effective in addressing weight-related outcomes. A smaller number of reviews examined combinations of approaches, e.g. nutrition improvement through curriculum, environment and parental involvement. These also found positive effects. A large number of reviews was identified which found positive effects from enhancing awareness and capacity building of the wider school community (NFSI Component 2). The effect of engaging families and communities through community involvement and outreach in the area of nutrition and health-related issues was the most frequently studied intervention. Parental involvement was identified as a moderating factor leading to greater success, with greater impact for a higher level of parental involvement. Direct involvement methods (e.g. face-to-face) where parents are expected to take an active role were more effective than indirect methods (e.g. newsletters) where parents may remain passive recipients of information. Less evidence was identified for the effect of staff training in delivering nutrition and healthy diet interventions, but current evidence suggests beneficial effects from training teachers and involving experts in school-based nutrition activities. Numerous reviews examined effectiveness or different aspects of delivering nutrition and health promoting curricula (NFSI Component 3). Effective nutrition education programmes were behaviourally-focused, context and age-specific, of longer duration and intensity, facilitated by trained staff, integrated into the curricula and/ or implemented in combination with other approaches (e.g. changes to the environment). Hands-on learning strategies and practical skills were considered promising, as were those that built children’s self-efficacy in making healthy changes in their diet and physical activity levels. Physical activity education was more effective when combined with nutrition interventions. A large body of evidence also exists for creating supportive environments for nutrition (NFSI Component 4), particularly through increasing availability of healthy foods and beverages and simultaneously reducing availability of unhealthy foods and beverages. Positive effects were reported on dietary and food availability outcomes from school lunch standards, competitive food and beverage policies and school food procurement 74 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE policies. Provision of school food, including fortified foods, had positive effect in low-income or disadvantaged settings. Provision of fruit and vegetables to children increased their consumption. Less evidence was identified on the impact of messaging and restricting marketing, although positive outcomes were reported in a handful of reviews of such interventions. Some evidence was identified regarding supportive school health services for nutrition (NFSI Component 5). School-based weight monitoring and screening programmes demonstrated positive effects in both children and their families. School health service delivery of interventions such as micronutrient supplementation, deworming and nutrition counselling also had positive results and supported the use of schools as a delivery platform for essential nutrition actions to school-age children and adolescents. Strengths and limitations This comprehensive review which identified more than 100 reviews synthesizing findings from primary studies systematically assesses and summarizes the evidence underpinning the NFSI criteria. The findings of this review may inform governments, policy-makers, and researchers concerned with school-based health and nutrition programmes and initiatives. As demonstrated in the Global Nutrition Policy Review conducted in 2016-2017 (12), there is a compelling need for comprehensive school health and nutrition programmes and policies, implementation of which had declined since the Global Nutrition Policy Review conducted in 2009-2010 (13). A major limitation of the evidence base is the geographical bias towards higher-income countries. Except for 13 reviews in low- or middle- WHO/Yoshi Shimizu 75DISCUSSION AND CONCLUSIONS income countries or among disadvantaged populations in high-income countries, the bulk of the evidence was from high-income countries and related to healthy diets and/or prevention of overweight/obesity. Only one of the low- or middle-income country reviews considered programmes to achieve healthy diets and prevent obesity, whereas other reviews focused on school feeding or interventions to reduce undernutrition. On the other hand, only one review among disadvantaged populations in high-income countries focused on undernutrition, while other reviews focused on healthy diets and obesity prevention. Other limitations of this review include the high level of heterogeneity amongst studies in the reviews and a likelihood of small study bias. Study populations, interventions, controls and outcomes varied across the studies themselves and in the way that these were synthesized for the reviews. Therefore, these findings must be interpreted with caution. It is also important to point out that while a large number of reviews were identified and included, many of these relied on the same primary studies. Regarding interventions, the NFSI is a package that includes multiple components which are likely to have synergistic effects. Therefore, there are limitations to the conclusions that can be drawn about the effects of the components individually and as a whole. No review was identified that assessed the effect of the whole set of components of NFSI on the health and nutrition outcomes of school-age children and adolescents. This review relies instead on evidence that considered one or more components from the NFSI Framework. At the same time, study and review authors may not have been reporting on exposure of the study populations to other components which had already been implemented as part of national policies and programmes. As a consequence, this evidence review is not able to assess the synergistic effects that may result from interventions that include all components of the NFSI. Furthermore, the commonly used categorization of “diet” and “physical activity” interventions described in the reviews may include a range of very different intervention approaches requiring different inputs, resources and engagement. Likewise, commonly studied interventions such as “nutrition education” could mask a range of different approaches (formal curricula vs. ad hoc information sessions) and have different content (food and nutrition theory vs. skills development). The methodologies and strategies used, and how they are used in combination, often determine the efficacy of such interventions. This high-level categorization may, therefore, not unpack the details that would be required in order to understand precisely what kinds of interventions and measures are needed to improve nutrition and diet in schools. Finally, many of the outcomes associated with nutrition, such as body weight, micronutrient status, bone density and eating habits, develop over time. Unfortunately, many of the studies included in this review relied on cross- sectional data which are unable to measure these long-term effects. Future research This review highlighted several areas where research was limited. These include the NFSI criteria related to dissemination of school nutrition policies, monitoring and evaluation of school nutrition-related curricula, the influence of environmental characteristics including provision of access to adequate eating places and cooking facilities, and nutrition-related school health services. Evidence was also scarce 76 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE or missing on the nutritional impact of relevant interventions that are typically not considered part of the nutrition domain. These would include interventions relating to hygiene and sanitary behaviour, separation of toilets or bullying and stigmatization. Adding nutrition outcomes to research on these interventions would provide valuable insights to underlying factors influencing nutrition. Future work is needed to assess the impacts of the components of NFSI, both individually and as a whole, on the health and nutrition outcomes of school children and communities. It is well recognized that effective nutrition policy responses require a comprehensive package of interlinked, and probably synergistic, measures. This, on the other hand, presents challenges for identifying the impact of individual interventions. It is not surprising, therefore, that the NFSI includes a broad spectrum of recommendations, and large effects on student outcomes are likely to be identified only when the full spectrum of activities are adopted and implemented. Moving forward, efforts must be made to support schools in adopting policies that cover the full NFSI set of essential criteria, whilst also ensuring robust arrangements for evaluation and monitoring, in order to build the evidence base required to identify the effectiveness of the NFSI approach or areas where it can be improved. Research on school health and nutrition interventions and the reporting of study and review results is heterogenous. To strengthen research on school W H O /Y os hi S hi m iz u 77DISCUSSION AND CONCLUSIONS health and nutrition programmes, therefore, future research may benefit from collecting individual-level outcome measures from students at various time points, ideally over a number of years, resources permitting. Also, future reviews and individual studies should report clearly on the full set of intervention approaches applied and components addressed, aspects which were poorly documented in the reviews examined. As noted, most of the research in this review—and in the larger area of school-health— relied on evidence from high-income countries. More research is needed from low- and lower- middle income countries, where school health and nutrition programmes may play a particularly important role, on educational outcomes as well as health and nutrition outcomes, but where contexts and resources often limit the quality of recommended intervention implementation. Conclusion Aligned with the pillars of the Health Promoting School approach, the NFSI can be well integrated into existing efforts in countries to increase the number of health promoting schools, for example through the Global School Health Initiative and the ongoing work to develop Global Standards for Health Promoting Schools. The WHO e-Library of Evidence for Nutrition Action (eLENA) (185) and WHO’s recommended essential nutrition actions (186) compile guidance and recommendations relevant for school-age children and adolescents. WHO has recently initiated work to develop two sets of guidelines with relevance to the NFSI following the WHO guideline development process. The first guideline on school food and nutrition policies is being developed by the WHO Department of Nutrition and Food Safety and its Nutrition Guidance Expert Advisory Group (NUGAG) (187, 188). The focus of the guideline is on interventions that influence the school food environment (NFSI component 4), including for example, nutrition standards or rules that determine the quality of food served or sold in and around schools and marketing restrictions of unhealthy foods and non-alcoholic beverages in and around schools. The second guideline is on school health services, being developed by the WHO Department of Maternal, Newborn, Child and Adolescent Health, which focuses on all school health services including nutrition (189). Other supporting WHO resources and tools include the Global Nutrition Policy Review, which tracks school health and nutrition policies and programmes, as well as the Global School-based Student Health Survey, which supports Member States to measure and assess the behavioural risk factors and protective factors among children and adolescents (190). The WHO School Policy Framework provides additional guidance for developing and implementing school policies and programmes to promote healthy diets and increase levels of physical activity (147). This includes guidance related to starting a school policy, policy options, stakeholder involvement and monitoring and evaluation. Several WHO regions have taken action to protect and promote good nutrition in schools. The WHO Regional Office for Europe provides a tool for the development of school nutrition 78 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE programmes (191) which outlines steps involved in designing a nutrition policy and outlines important elements of such a policy. Additionally, it provides guidance on designing children’s diets, recommended nutrient intakes for children and adolescents and nutrition education. The WHO Western Pacific Regional Office developed a guide for school principals relating to restricting the sale and marketing of sugary drinks in and around schools (192). The guide outlines why a focus on school drink environments is important, as well as providing examples of success stories and recommended actions for school principals. Furthermore, all WHO Regional Offices have developed regional nutrient profile models to identify which foods and non-alcoholic beverages for which marketing to children should be restricted. In countries that do not have and are not planning to adopt related nationwide regulations — such as restrictions on print, broadcast or digital advertising and other forms of marketing — these tools could be used to develop regulations that cover marketing and availability of foods and beverages specifically in school settings. Additional guidance may also be found from several other UN initiatives, such as the FRESH initiative, which works to raise awareness in the education sector of the value of implementing effective school health programmes as an advocacy strategy (16). FAO recently developed a framework to guide its work in the area of school food and @ In gi m ag e nutrition (193), while the UN agencies are also working jointly on guidance for setting nutrition standards for school meals. In conclusion, this review provides an extensive summary of current evidence for the effectiveness of many components of the NFSI. Member States are encouraged, therefore, to draw on this evidence base and to use the NFSI framework to build country-specific school nutrition-related policies and programmes. In doing so, they can fully realize the potential of schools as a key setting for improving nutrition throughout childhood and adolescence, while laying good foundations for health and well- being later in life. In doing so, they can fully realize the potential of schools as a key setting for improving nutrition throughout childhood and adolescence, while laying good foundations for health and wellbeing later in life. 79DISCUSSION AND CONCLUSIONS REFERENCES 1. Darnton-Hill I, Nishida C, James W. A life course approach to diet, nutrition and the prevention of chronic diseases. Public health nutrition. 2004;7:101-21. doi: 10.1079/phn2003584. 2. Sawyer SM, Afifi RA, Bearinger LH, Blakemore S-J, Dick B, Ezeh AC et al. Adolescence: a foundation for future health. The Lancet. 2012;379:1630-40. doi: 10.1016/S0140-6736(12)60072-5. 3. Das JK, Salam RA, Thornburg KL, Prentice AM, Campisi S, Lassi ZS et al. Nutrition in adolescents: physiology, metabolism, and nutritional needs. Ann N Y Acad Sci. 2017;1393:21-33. doi:10.1111/nyas.13330. 4. Bryan J, Osendarp S, Hughes D, Calvaresi E, Baghurst K, van Klinken J-W. Nutrients for cognitive development in school-aged children. Nutrition reviews. 2004;62:295-306. doi:10.1111/j.1753-4887.2004.tb00055.x. 5. Sridhar D. Linkages between nutrition, ill-health and education. Paris: UNESCO; 2008 (https://www. researchgate.net/publication/228927260_Linkages_between_Nutrition_Ill-Health_and_Education, accessed 30 November 2020). 6. Alderman H, Behrman JR, Hoddinott J. Economic and nutritional analyses offer substantial synergies for understanding human nutrition. J Nutr. 2007;137:537-44. doi:10.1093/jn/137.3.537. 7. Hanushek EA, Woessmann L. The Role of Education Quality for Economic Growth. . Washington, D.C.: World Bank; 2007; Policy Research Working Paper; No. 4122 (https://openknowledge.worldbank.org/ handle/10986/7154, accessed 30 November 2020). 8. Bundy DAP, Silva N, Horton S, Patton GC, Schultz L, Jamison DT. Child and Adolescent Health and Development: Realizing Neglected Potential. Chapter in Bundy DAP, Silva N, Horton S, Jamison DT, Patton GC. Optimizing Education Outcomes: High-Return Investments in School Health for Increased Participation and Learning. Washington, D.C: World Bank. doi:10.1596/978-1-4648-0423-6_ch1. 9. UNSCN. Schools as a System to Improve Nutrition. Rome: United Nations System Standing Committee on Nutrition; 2017 (https://www.unscn.org/uploads/web/news/document/School-Paper-EN-WEB.pdf, accessed 12 October 2020). 10. Khambalia A, Dickinson S, Hardy L, Gill T, Baur L. A synthesis of existing systematic reviews and meta‐analyses of school‐based behavioural interventions for controlling and preventing obesity. Obesity Reviews. 2012;13:214- 33. doi:10.1111/j.1467-789X.2011.00947.x. 11. WHO. Global Accelerated Action for the Health of Adolescents (AA-HA!): guidance to support country implementation. Geneva: World Health Organization; 2017 (https://www.who.int/maternal_child_adolescent/ topics/adolescence/framework-accelerated-action/en/, accessed 30 November 2020). 12. WHO. Global nutrition policy review 2016-2017. Country progress in creating enabling policy environments for promoting healthy diets and nutrition. Geneva: World Health Organization; 2018 (https://www.who.int/ nutrition/publications/policies/global_nut_policyreview_2016-2017/en/, accessed 12 October 2020). 13. WHO. Global nutrition policy review. What does it take to scale up nutrition action? Geneva: World Health Organization; 2013 (https://www.who.int/publications/i/item/9789241505529, accessed 3 December 2020). 14. FAO. Nutrition guidelines and standards for school meals. A report from 33 low and middle-income countries. Rome: Food and Agriculture Organization of the United Nations; 2019 (http://www.fao.org/3/CA2773EN/ ca2773en.pdf, accessed 3 December 2020). 15. WHO. Healthy Nutrition: An Essential Element of a Health- Promoting School. WHO Information Series on School Health. Document 4. Geneva: World Health Organization; 1998 (https://www.who.int/school_youth_health/ media/en/428.pdf?ua=1, accessed 3 December 2020). 16. UNESCO, UNICEF, WHO, the World Bank. Focusing Resources on Effective School Health: a FRESH start to enhancing the quality and equity of education. (https://unesdoc.unesco.org/ark:/48223/pf0000124086, 12 October 2020). 80 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 17. WFP, UNICEF. The Essential Package. Twelve interventions to improve the health and nutrition of school-age children. Rome & New York: World Food Programme & United Nations Children’s Fund; 2002 (https://documents. wfp.org/stellent/groups/public/documents/newsroom/wfp212806.pdf, accessed 12 October 2020). 18. UNICEF. Child-friendly schools. United Nations Children’s Fund; 2009. (https://www.unicef.org/publications/ index_49574.html , accessed 30 October 2020). 19. WHO. Health-Promoting Schools. A healthy setting for living, learning and working. Geneva: World Health Organization; 1998 (https://www.who.int/school_youth_health/media/en/92.pdf?ua=1, accessed 12 October 2020). 20. FAO. Nutrition Education in Primary Schools. Rome: Food and Agriculture Organization of the United Nations; 2005 (http://www.fao.org/3/a0333e/a0333e00.htm, accessed 30 October 2020). 21. WHO. WHO Expert Meeting on Childhood Obesity, Kobe, Japan, 20–24 June, 2005. World Health Organization; 2005. (https://www.who.int/nmh/media/obesity_expert_meeting/en/, accessed 3 December 2020). 22. WHO. Report of the brainstorming meeting on the development of a framework on the Nutrition-Friendly Schools Initiative, Feb 27–28, 2006, Montreux, Switzerland. World Health Organization; 2006 (https://www.who. int/nutrition/publications/Montreux_Meeting_Report.pdf, accessed December 2020). 23. Summerbell CD, Waters E, Edmunds L, Kelly S, Brown T, Campbell KJ. Interventions for preventing obesity in children. Cochrane Database Syst Rev. 2005;3. doi:10.1002/14651858.CD001871.pub2. 24. Centre for Public Health Excellence at NICE (UK), National Collaborating Centre for Primary Care (UK). Obesity: The Prevention, Identification, Assessment and Management of Overweight and Obesity in Adults and Children. In:2006: https://www.ncbi.nlm.nih.gov/pubmed/22497033, accessed 30 October 2020). 25. Flynn M, McNeil D, Maloff B, Mutasingwa D, Wu M, Ford C et al. Reducing obesity and related chronic disease risk in children and youth: a synthesis of evidence with ‘best practice’recommendations. Obesity reviews. 2006;7:7-66. doi:10.1111/j.1467-789X.2006.00242.x. 26. Delisle HF, Receveur O, Agueh V, Nishida C. Pilot project of the Nutrition-Friendly School Initiative (NFSI) in Ouagadougou, Burkina Faso and Cotonou, Benin, in West Africa. Glob Health Promot. 2013;20:39-49. doi:10.1177/1757975913476907. 27. Halk Sağlığı Genel Müdürlüğü (General Directorate of Public Health). Beslenme Dostu Okullar Programı (Nutrition Friendly Schools Program). (https://hsgm.saglik.gov.tr/tr/okul-sagligi/beslenme-dostu-okullar- program%C4%B1.html, accessed 12 November 2020). 28. WHO. Report of the Commission on Ending Childhood Obesity: implementation plan: executive summary. Geneva: World Health Organization; 2017 (https://www.who.int/end-childhood-obesity/publications/echo- plan-executive-summary/en/, accessed 3 December 2020). 29. WHO. Making every school a health promoting school. (https://www.who.int/maternal_child_adolescent/ adolescence/making-every-school-a-health-promoting-school/en/, accessed 12 October 2020). 30. WHO. Thirteenth general programme of Work 2019-2023: Promote health, Keep the world safe, Serve the vulnerable. Geneva: World Health Organization; 2019 (https://apps.who.int/iris/bitstream/ handle/10665/324775/WHO-PRP-18.1-eng.pdf, accessed 12 October 2020). 31. Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N, Hamel C et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med Res Methodol. 2007;7:10. doi:10.1186/1471-2288-7-10. 32. Shea BJ, Bouter LM, Peterson J, Boers M, Andersson N, Ortiz Z et al. External validation of a measurement tool to assess systematic reviews (AMSTAR). PLoS One. 2007;2:e1350. doi:10.1371/journal.pone.0001350. 33. Sequeira-Byron P, Fedorowicz Z, Jagannath VA, Sharif MO. An AMSTAR assessment of the methodological quality of systematic reviews of oral healthcare interventions published in the Journal of Applied Oral Science (JAOS). J Appl Oral Sci. 2011;19:440-7. doi:10.1590/s1678-77572011000500002. 81REFERENCES 34. Shea BJ, Hamel C, Wells GA, Bouter LM, Kristjansson E, Grimshaw J et al. AMSTAR is a reliable and valid measurement tool to assess the methodological quality of systematic reviews. J Clin Epidemiol. 2009;62:1013- 20. doi:10.1016/j.jclinepi.2008.10.009. 35. Bramante CT, Thornton RLJ, Bennett WL, Zhang A, Wilson RF, Bass EB et al. Systematic Review of Natural Experiments for Childhood Obesity Prevention and Control. Am J Prev Med. 2019;56:147-58. doi:10.1016/j.amepre.2018.08.023. 36. Micha R, Karageorgou D, Bakogianni I, Trichia E, Whitsel LP, Story M et al. Effectiveness of school food environment policies on children’s dietary behaviors: A systematic review and meta-analysis. PLoS One. 2018;13:e0194555. doi:10.1371/journal.pone.0194555. 37. Jaime PC, Lock K. Do school based food and nutrition policies improve diet and reduce obesity? Prev Med. 2009;48:45-53. doi:10.1016/j.ypmed.2008.10.018. 38. Driessen CE, Cameron AJ, Thornton LE, Lai SK, Barnett LM. Effect of changes to the school food environment on eating behaviours and/or body weight in children: a systematic review. Obes Rev. 2014;15:968-82. doi:10.1111/ obr.12224. 39. Mayne SL, Auchincloss AH, Michael YL. Impact of policy and built environment changes on obesity-related outcomes: a systematic review of naturally occurring experiments. Obes Rev. 2015;16:362-75. doi:10.1111/ obr.12269. 40. Afshin A, Penalvo J, Del Gobbo L, Kashaf M, Micha R, Morrish K et al. CVD Prevention Through Policy: a Review of Mass Media, Food/Menu Labeling, Taxation/Subsidies, Built Environment, School Procurement, Worksite Wellness, and Marketing Standards to Improve Diet. Curr Cardiol Rep. 2015;17:98. doi:10.1007/s11886-015- 0658-9. 41. Chriqui JF, Pickel M, Story M. Influence of school competitive food and beverage policies on obesity, consumption, and availability: a systematic review. JAMA Pediatr. 2014;168:279-86. doi:10.1001/ jamapediatrics.2013.4457. 42. Sildén KE. Impact of competitive foods in public schools on child nutrition: effects on adolescent obesity in the United States an integrative systematic literature review. Glob Health Action. 2018;11:1477492. doi:10.1080/16549716.2018.1477492. 43. Mansfield JL, Savaiano DA. Effect of school wellness policies and the Healthy, Hunger-Free Kids Act on food- consumption behaviors of students, 2006-2016: a systematic review. Nutr Rev. 2017;75:533-52. doi:10.1093/ nutrit/nux020. 44. von Philipsborn P, Stratil JM, Burns J, Busert LK, Pfadenhauer LM, Polus S et al. Environmental interventions to reduce the consumption of sugar-sweetened beverages and their effects on health. Cochrane Database Syst Rev. 2019;6:CD012292. doi:10.1002/14651858.CD012292.pub2. 45. Vézina-Im LA, Beaulieu D, Bélanger-Gravel A, Boucher D, Sirois C, Dugas M et al. Efficacy of school-based interventions aimed at decreasing sugar-sweetened beverage consumption among adolescents: a systematic review. Public Health Nutr. 2017;20:2416-31. doi:10.1017/S1368980017000076. 46. Langford R, Bonell CP, Jones HE, Pouliou T, Murphy SM, Waters E et al. The WHO Health Promoting School framework for improving the health and well‐being of students and their academic achievement. The Cochrane Library. 2014. doi:10.1002/14651858.CD008958.pub2. 47. Waters E, de Silva-Sanigorski A, Hall BJ, Brown T, Campbell KJ, Gao Y et al. Interventions for preventing obesity in children. Cochrane Database Syst Rev. 2011:CD001871. doi:10.1002/14651858.CD001871.pub3. 48. Peirson L, Fitzpatrick-Lewis D, Morrison K, Ciliska D, Kenny M, Usman Ali M et al. Prevention of overweight and obesity in children and youth: a systematic review and meta-analysis. CMAJ Open. 2015;3:E23-33. doi:10.9778/ cmajo.20140053. 82 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 49. Gori D, Guaraldi F, Cinocca S, Moser G, Rucci P, Fantini MP. Effectiveness of educational and lifestyle interventions to prevent paediatric obesity: systematic review and meta-analyses of randomized and non-randomized controlled trials. Obes Sci Pract. 2017;3:235-48. doi:10.1002/osp4.111. 50. Vasques C, Magalhães P, Cortinhas A, Mota P, Leitão J, Lopes VP. Effects of intervention programs on child and adolescent BMI: A meta-analysis study. J Phys Act Health. 2014;11:426-44. doi:10.1123/jpah.2012-0035. 51. Luckner H, Moss JR, Gericke CA. Effectiveness of interventions to promote healthy weight in general populations of children and adults: a meta-analysis. Eur J Public Health. 2012;22:491-7. doi:10.1093/eurpub/ckr141. 52. Sobol-Goldberg S, Rabinowitz J, Gross R. School-based obesity prevention programs: a meta-analysis of randomized controlled trials. Obesity (Silver Spring). 2013;21:2422-8. doi:10.1002/oby.20515. 53. Friedrich RR, Schuch I, Wagner MB. Effect of interventions on the body mass index of school-age students. Rev Saude Publica. 2012;46:551-60. doi:10.1590/s0034-89102012005000036. 54. Katz DL, O’Connell M, Njike VY, Yeh MC, Nawaz H. Strategies for the prevention and control of obesity in the school setting: systematic review and meta-analysis. Int J Obes (Lond). 2008;32:1780-9. doi:10.1038/ijo.2008.158. 55. Lavelle HV, Mackay DF, Pell JP. Systematic review and meta-analysis of school-based interventions to reduce body mass index. J Public Health (Oxf). 2012;34:360-9. doi:10.1093/pubmed/fdr116. 56. Feng L, Wei DM, Lin ST, Maddison R, Ni Mhurchu C, Jiang Y et al. Systematic review and meta-analysis of school-based obesity interventions in mainland China. PLoS One. 2017;12:e0184704. doi:10.1371/journal.pone.0184704. 57. Brown T, Moore TH, Hooper L, Gao Y, Zayegh A, Ijaz S et al. Interventions for preventing obesity in children. Cochrane Database Syst Rev. 2019;7:CD001871. doi:10.1002/14651858.CD001871.pub4. 58. Wang Y, Wu Y, Wilson RF, Bleich S, Cheskin L, Weston C et al. Childhood Obesity Prevention Programs: Comparative Effectiveness Review and Meta-Analysis. Rockville (MD): Agency for Healthcare Research and Quality (US); 2013 (https://www.ncbi.nlm.nih.gov/pubmed/23865092, accessed 3 December 2020). 59. Evans CE, Christian MS, Cleghorn CL, Greenwood DC, Cade JE. Systematic review and meta-analysis of school-based interventions to improve daily fruit and vegetable intake in children aged 5 to 12 y. Am J Clin Nutr. 2012;96:889-901. doi:10.3945/ajcn.111.030270. 60. Howerton MW, Bell BS, Dodd KW, Berrigan D, Stolzenberg-Solomon R, Nebeling L. 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;39:186-96. doi:10.1016/j.jneb.2007.01.010. 61. Cai L, Wu Y, Wilson RF, Segal JB, Kim MT, Wang Y. Effect of childhood obesity prevention programs on blood pressure: a systematic review and meta-analysis. Circulation. 2014;129:1832-9. doi:10.1161/ CIRCULATIONAHA.113.005666. 62. Cai L, Wu Y, Cheskin LJ, Wilson RF, Wang Y. Effect of childhood obesity prevention programmes on blood lipids: a systematic review and meta-analysis. Obes Rev. 2014;15:933-44. doi:10.1111/obr.12227. 63. Mead E, Brown T, Rees K, Azevedo LB, Whittaker V, Jones D et al. Diet, physical activity and behavioural interventions for the treatment of overweight or obese children from the age of 6 to 11 years. Cochrane Database Syst Rev. 2017;6:CD012651. doi:10.1002/14651858.CD012651. 64. Al-Khudairy L, Loveman E, Colquitt JL, Mead E, Johnson RE, Fraser H et al. Diet, physical activity and behavioural interventions for the treatment of overweight or obese adolescents aged 12 to 17 years. Cochrane Database Syst Rev. 2017;6:CD012691. doi:10.1002/14651858.CD012691. 65. Hung LS, Tidwell DK, Hall ME, Lee ML, Briley CA, Hunt BP. A meta-analysis of school-based obesity prevention programs demonstrates limited efficacy of decreasing childhood obesity. Nutr Res. 2015;35:229-40. doi:10.1016/j.nutres.2015.01.002. 66. Guerra PH, Nobre MR, da Silveira JA, Taddei JA. School-based physical activity and nutritional education interventions on body mass index: a meta-analysis of randomised community trials - project PANE. Prev Med. 2014;61:81-9. doi:10.1016/j.ypmed.2014.01.005. 83REFERENCES 67. Williams AJ, Henley WE, Williams CA, Hurst AJ, Logan S, Wyatt KM. Systematic review and meta-analysis of the association between childhood overweight and obesity and primary school diet and physical activity policies. Int J Behav Nutr Phys Act. 2013;10:101. doi:10.1186/1479-5868-10-101. 68. De Bourdeaudhuij I, Van Cauwenberghe E, Spittaels H, Oppert JM, Rostami C, Brug J et al. School-based interventions promoting both physical activity and healthy eating in Europe: a systematic review within the HOPE project. Obes Rev. 2011;12:205-16. doi:10.1111/j.1467-789X.2009.00711.x. 69. Verstraeten R, Roberfroid D, Lachat C, Leroy JL, Holdsworth M, Maes L et al. Effectiveness of preventive school-based obesity interventions in low- and middle-income countries: a systematic review. Am J Clin Nutr. 2012;96:415-38. doi:10.3945/ajcn.112.035378. 70. Luybli M, Schmillen H, Sotos-Prieto M. School-Based Interventions in Low Socioeconomic Settings to Reduce Obesity Outcomes among Preschoolers: A Scoping Review. Nutrients. 2019;11. doi:10.3390/nu11071518. 71. Johnson T, Weed LD, Touger-Decker R. School-based interventions for overweight and obesity in minority school children. J Sch Nurs. 2012;28:116-23. doi:10.1177/1059840511426147. 72. Wang D, Stewart D. The implementation and effectiveness of school-based nutrition promotion programmes using a health-promoting schools approach: a systematic review. Public Health Nutr. 2013;16:1082-100. doi:10.1017/S1368980012003497. 73. Brown EC, Buchan DS, Baker JS, Wyatt FB, Bocalini DS, Kilgore L. A Systematised Review of Primary School Whole Class Child Obesity Interventions: Effectiveness, Characteristics, and Strategies. Biomed Res Int. 2016;2016:4902714. doi:10.1155/2016/4902714. 74. Meiklejohn S, Ryan L, Palermo C. A Systematic Review of the Impact of Multi-Strategy Nutrition Education Programs on Health and Nutrition of Adolescents. J Nutr Educ Behav. 2016;48:631-46.e1. doi:10.1016/j. jneb.2016.07.015. 75. USDA. Chapter 4-D. Food Environment. In: A Series of Systematic Reviews on the Effects of Nutrition Education on Children’s and Adolescents’ Dietary Intake. Virginia: U.S. Department of Agriculture; 2012 (https://nesr.usda. gov/sites/default/files/2019-04/NutritionEducation-FullReport.pdf, accessed 29 October 2020). 76. Colley P, Myer B, Seabrook J, Gilliland J. The Impact of Canadian School Food Programs on Children’s Nutrition and Health: A Systematic Review. Can J Diet Pract Res. 2019;80:79-86. doi:10.3148/cjdpr-2018-037. 77. de Sa J, Lock K. Will European agricultural policy for school fruit and vegetables improve public health? A review of school fruit and vegetable programmes. Eur J Public Health. 2008;18:558-68. doi:10.1093/eurpub/ckn061. 78. Aloia CR, Shockey TA, Nahar VK, Knight KB. Pertinence of the recent school-based nutrition interventions targeting fruit and vegetable consumption in the United States:a systematic review. Health Promot Perspect. 2016;6:1-9. doi:10.15171/hpp.2016.01. 79. Knai C, Pomerleau J, Lock K, McKee M. Getting children to eat more fruit and vegetables: a systematic review. Prev Med. 2006;42:85-95. doi:10.1016/j.ypmed.2005.11.012. 80. Bleich SN, Vercammen KA, Zatz LY, Frelier JM, Ebbeling CB, Peeters A. Interventions to prevent global childhood overweight and obesity: a systematic review. Lancet Diabetes Endocrinol. 2018;6:332-46. doi:10.1016/S2213- 8587(17)30358-3. 81. Brown T, Summerbell C. Systematic review of school-based interventions that focus on changing dietary intake and physical activity levels to prevent childhood obesity: an update to the obesity guidance produced by the National Institute for Health and Clinical Excellence. Obes Rev. 2009;10:110-41. doi:10.1111/j.1467- 789X.2008.00515.x. 82. Stevens CJ. Obesity prevention interventions for middle school-age children of ethnic minority: a review of the literature. J Spec Pediatr Nurs. 2010;15:233-43. doi:10.1111/j.1744-6155.2010.00242.x. 84 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 83. Doak CM, Visscher TL, Renders CM, Seidell JC. The prevention of overweight and obesity in children and adolescents: a review of interventions and programmes. Obes Rev. 2006;7:111-36. doi:10.1111/j.1467- 789X.2006.00234.x. 84. Lissau I. Prevention of overweight in the school arena. Acta Paediatr. 2007;96:12-8. doi:10.1111/j.1651- 2227.2007.00164.x. 85. Kropski JA, Keckley PH, Jensen GL. School-based obesity prevention programs: an evidence-based review. Obesity (Silver Spring). 2008;16:1009-18. doi:10.1038/oby.2008.29. 86. McIsaac JD, Spencer R, Chiasson K, Kontak J, Kirk SFL. Factors Influencing the Implementation of Nutrition Policies in Schools: A Scoping Review. Health Educ Behav. 2019;46:224-50. doi:10.1177/1090198118796891. 87. Niemeier BS, Hektner JM, Enger KB. Parent participation in weight-related health interventions for children and adolescents: a systematic review and meta-analysis. Prev Med. 2012;55:3-13. doi:10.1016/j.ypmed.2012.04.021. 88. Schroeder K, Travers J, Smaldone A. Are School Nurses an Overlooked Resource in Reducing Childhood Obesity? A Systematic Review and Meta-Analysis. J Sch Health. 2016;86:309-21. doi:10.1111/josh.12386. 89. Oosterhoff M, Joore M, Ferreira I. The effects of school-based lifestyle interventions on body mass index and blood pressure: a multivariate multilevel meta-analysis of randomized controlled trials. Obes Rev. 2016;17:1131- 53. doi:10.1111/obr.12446. 90. Krishnaswami J, Martinson M, Wakimoto P, Anglemeyer A. Community-engaged interventions on diet, activity, and weight outcomes in U.S. schools: a systematic review. Am J Prev Med. 2012;43:81-91. doi:10.1016/j. amepre.2012.02.031. 91. Nathan N, Janssen L, Sutherland R, Hodder RK, Evans CEL, Booth D et al. The effectiveness of lunchbox interventions on improving the foods and beverages packed and consumed by children at centre-based care or school: a systematic review and meta-analysis. Int J Behav Nutr Phys Act. 2019;16:38. doi:10.1186/s12966-019- 0798-1. 92. Dudley DA, Cotton WG, Peralta LR. Teaching approaches and strategies that promote healthy eating in primary school children: a systematic review and meta-analysis. Int J Behav Nutr Phys Act. 2015;12:28. doi:10.1186/ s12966-015-0182-8. 93. Wang Y, Cai L, Wu Y, Wilson RF, Weston C, Fawole O et al. What childhood obesity prevention programmes work? A systematic review and meta-analysis. Obes Rev. 2015;16:547-65. doi:10.1111/obr.12277. 94. Hingle MD, O’Connor TM, Dave JM, Baranowski T. Parental involvement in interventions to improve child dietary intake: a systematic review. Prev Med. 2010;51:103-11. doi:10.1016/j.ypmed.2010.04.014. 95. Verjans-Janssen SRB, van de Kolk I, Van Kann DHH, Kremers SPJ, Gerards SMPL. Effectiveness of school-based physical activity and nutrition interventions with direct parental involvement on children’s BMI and energy balance-related behaviors - A systematic review. PLoS One. 2018;13:e0204560. doi:10.1371/journal.pone.0204560. 96. USDA. Chapter 4-B. Parental Involvement. In: A Series of Systematic Reviews on the Effects of Nutrition Education on Children’s and Adolescents’ Dietary Intake. Virginia: U.S. Department of Agriculture; 2012 (https:// nesr.usda.gov/sites/default/files/2019-04/NutritionEducation-FullReport.pdf, accessed 29 October 2020). 97. Langford R, Bonell C, Jones H, Campbell R. Obesity prevention and the Health promoting Schools framework: essential components and barriers to success. Int J Behav Nutr Phys Act. 2015;12:15. doi:10.1186/s12966-015-0167-7. 98. Jourdan D, Christensen JH, Darlington E, Bonde AH, Bloch P, Jensen BB et al. The involvement of young people in school- and community-based noncommunicable disease prevention interventions: a scoping review of designs and outcomes. BMC Public Health. 2016;16:1123. doi:10.1186/s12889-016-3779-1. 99. Nixon CA, Moore HJ, Douthwaite W, Gibson EL, Vogele C, Kreichauf S et al. Identifying effective behavioural models and behaviour change strategies underpinning preschool- and school-based obesity prevention interventions aimed at 4-6-year-olds: a systematic review. Obes Rev. 2012;13 Suppl 1:106-17. doi:10.1111/j.1467-789X.2011.00962.x. 85REFERENCES 100. Ward DS, Welker E, Choate A, Henderson KE, Lott M, Tovar A et al. Strength of obesity prevention interventions in early care and education settings: A systematic review. Prev Med. 2017;95 Suppl:S37-S52. doi:10.1016/j. ypmed.2016.09.033. 101. Silveira JA, Taddei JA, Guerra PH, Nobre MR. Effectiveness of school-based nutrition education interventions to prevent and reduce excessive weight gain in children and adolescents: a systematic review. J Pediatr (Rio J). 2011;87:382-92. doi:10.2223/JPED.2123. 102. Murimi MW, Moyeda-Carabaza AF, Nguyen B, Saha S, Amin R, Njike V. Factors that contribute to effective nutrition education interventions in children: a systematic review. Nutr Rev. 2018;76:553-80. (https://academic. oup.com/nutritionreviews/article/76/8/553/5003346, accessed 9 December 2020). 103. Saraf DS, Nongkynrih B, Pandav CS, Gupta SK, Shah B, Kapoor SK et al. A systematic review of school-based interventions to prevent risk factors associated with noncommunicable diseases. Asia Pac J Public Health. 2012;24:733-52. doi:10.1177/1010539512445053. 104. Van Lippevelde W, Verloigne M, De Bourdeaudhuij I, Brug J, Bjelland M, Lien N et al. Does parental involvement make a difference in school-based nutrition and physical activity interventions? A systematic review of randomized controlled trials. Int J Public Health. 2012;57:673-8. doi:10.1007/s00038-012-0335-3. 105. Black L, Matvienko-Sikar K, Kearney PM. The association between childcare arrangements and risk of overweight and obesity in childhood: a systematic review. Obes Rev. 2017;18:1170-90. doi:10.1111/obr.12575. 106. Wolfenden L, Jones J, Williams CM, Finch M, Wyse RJ, Kingsland M et al. Strategies to improve the implementation of healthy eating, physical activity and obesity prevention policies, practices or programmes within childcare services. Cochrane Database Syst Rev. 2016;10:CD011779. doi:10.1002/14651858.CD011779.pub2. 107. Wolfenden L, Nathan NK, Sutherland R, Yoong SL, Hodder RK, Wyse RJ et al. Strategies for enhancing the implementation of school-based policies or practices targeting risk factors for chronic disease. Cochrane Database Syst Rev. 2017;11:CD011677. doi:10.1002/14651858.CD011677.pub2. 108. WHO. 5 keys to a healthy diet. Geneva: World Health Organization; 2019 (https://www.who.int/nutrition/ topics/5keys_healthydiet/en/, accessed 30 October 2020). 109. WHO. Food safety. Fact sheet. Geneva: World Health Organization; 2019 (https://www.who.int/news-room/fact- sheets/detail/food-safety, accessed 30 October 2020). 110. WHO. Healthy diet. Fact sheet. Geneva: World Health Organization; 2018 (https://www.who.int/news-room/fact- sheets/detail/healthy-diet, accessed 30 October 2020). 111. FAO, United Arab Emirates University. Stepping up school-based food and nutrition education: Exploring challenges, finding solutions and building partnerships. Rome: Food and Agriculture Organization of the United Nations; 2019 (http://www.fao.org/3/CA3063EN/CA3063EN.pdf, accessed 30 October 2020). 112. Wang X, Zhou G, Zeng J, Yang T, Chen J, Li T. Effect of educational interventions on health in childhood: a meta- analysis of randomized controlled trials. Public Health. 2018;164:134-47. doi:10.1016/j.puhe.2018.04.013. 113. Silveira JA, Taddei JA, Guerra PH, Nobre MR. The effect of participation in school-based nutrition education interventions on body mass index: a meta-analysis of randomized controlled community trials. Prev Med. 2013;56:237-43. doi:10.1016/j.ypmed.2013.01.011. 114. Kong K, Liu J, Tao Y. Limitations of studies on school-based nutrition education interventions for obesity in China: a systematic review and meta-analysis. Asia Pac J Clin Nutr. 2016;25:589-601. doi:10.6133/ apjcn.092015.19. 115. Nekitsing C, Blundell-Birtill P, Cockroft JE, Hetherington MM. Systematic review and meta-analysis of strategies to increase vegetable consumption in preschool children aged 2-5 years. Appetite. 2018;127:138-54. doi:10.1016/j.appet.2018.04.019. 86 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 116. Delgado-Noguera M, Tort S, Martínez-Zapata MJ, Bonfill X. Primary school interventions to promote fruit and vegetable consumption: a systematic review and meta-analysis. Prev Med. 2011;53:3-9. doi:10.1016/j. ypmed.2011.04.016. 117. Vargas-Garcia EJ, Evans CEL, Prestwich A, Sykes-Muskett BJ, Hooson J, Cade JE. Interventions to reduce consumption of sugar-sweetened beverages or increase water intake: evidence from a systematic review and meta-analysis. Obes Rev. 2017;18:1350-63. doi:10.1111/obr.12580. 118. Abdel Rahman A, Jomaa L, Kahale LA, Adair P, Pine C. Effectiveness of behavioral interventions to reduce the intake of sugar-sweetened beverages in children and adolescents: a systematic review and meta-analysis. Nutr Rev. 2018;76:88-107. doi:10.1093/nutrit/nux061. 119. Price C, Cohen D, Pribis P, Cerami J. Nutrition Education and Body Mass Index in Grades K-12: A Systematic Review. J Sch Health. 2017;87:715-20. doi:10.1111/josh.12544. 120. Peralta LR, Dudley DA, Cotton WG. Teaching Healthy Eating to Elementary School Students: A Scoping Review of Nutrition Education Resources. J Sch Health. 2016;86:334-45. doi:10.1111/josh.12382. 121. USDA. Chapter 4-E. Multi- vs. Single-Component Interventions. In: A Series of Systematic Reviews on the Effects of Nutrition Education on Children’s and Adolescents’ Dietary Intake. Virginia: U.S. Department of Agriculture; 2012 (https://nesr.usda.gov/sites/default/files/2019-04/NutritionEducation-FullReport.pdf, accessed 12 October 2020). 122. Bailey CJ, Drummond MJ, Ward PR. Food literacy programmes in secondary schools: a systematic literature review and narrative synthesis of quantitative and qualitative evidence. Public Health Nutr. 2019;22:2891-913. doi:10.1017/S1368980019001666. 123. Robinson-O’Brien R, Story M, Heim S. Impact of garden-based youth nutrition intervention programs: a review. J Am Diet Assoc. 2009;109:273-80. doi:10.1016/j.jada.2008.10.051. 124. Frerichs L, Brittin J, Sorensen D, Trowbridge MJ, Yaroch AL, Siahpush M et al. Influence of school architecture and design on healthy eating: a review of the evidence. Am J Public Health. 2015;105:e46-57. doi:10.2105/ AJPH.2014.302453. 125. Muzaffar H, Metcalfe JJ, Fiese B. Narrative Review of Culinary Interventions with Children in Schools to Promote Healthy Eating: Directions for Future Research and Practice. Curr Dev Nutr. 2018;2:nzy016. doi:10.1093/cdn/ nzy016. 126. Ajie WN, Chapman-Novakofski KM. Impact of computer-mediated, obesity-related nutrition education interventions for adolescents: a systematic review. J Adolesc Health. 2014;54:631-45. doi:10.1016/j. jadohealth.2013.12.019. 127. Whittemore R, Chao A, Popick R, Grey M. School-based internet obesity prevention programs for adolescents: a systematic literature review. Yale J Biol Med. 2013;86:49-62. (https://pubmed.ncbi.nlm.nih.gov/23482347, accessed 3 December 2020). 128. Hamel LM, Robbins LB. Computer- and web-based interventions to promote healthy eating among children and adolescents: a systematic review. J Adv Nurs. 2013;69:16-30. doi:10.1111/j.1365-2648.2012.06086.x. 129. do Amaral E Melo GR, de Carvalho Silva Vargas F, Dos Santos Chagas CM, Toral N. Nutritional interventions for adolescents using information and communication technologies (ICTs): A systematic review. PLoS One. 2017;12:e0184509. doi:10.1371/journal.pone.0184509. 130. USDA. Chapter 4-A. Digital Media and Technology. In: A Series of Systematic Reviews on the Effects of Nutrition Education on Children’s and Adolescents’ Dietary Intake. Virginia: U.S. Department of Agriculture; 2012 (https:// nesr.usda.gov/sites/default/files/2019-04/NutritionEducation-FullReport.pdf, accessed 29 October 2020). 131. Calvert S, Dempsey RC, Povey R. Delivering in-school interventions to improve dietary behaviours amongst 11- to 16-year-olds: A systematic review. Obes Rev. 2019;20:543-53. doi:10.1111/obr.12797. 87REFERENCES 132. USDA. Chapter 4-C. Type of Educator. In: A Series of Systematic Reviews on the Effects of Nutrition Education on Children’s and Adolescents’ Dietary Intake. Virginia: U.S. Department of Agriculture; 2012: (https://nesr.usda. gov/sites/default/files/2019-04/NutritionEducation-FullReport.pdf, accessed 29 October 2020). 133. Yip C, Gates M, Gates A, Hanning RM. Peer-led nutrition education programs for school-aged youth: a systematic review of the literature. Health Educ Res. 2016;31:82-97. doi:10.1093/her/cyv063. 134. Avery A, Bostock L, McCullough F. A systematic review investigating interventions that can help reduce consumption of sugar-sweetened beverages in children leading to changes in body fatness. J Hum Nutr Diet. 2015;28 Suppl 1:52-64. doi:10.1111/jhn.12267. 135. Van Cauwenberghe E, Maes L, Spittaels H, van Lenthe FJ, Brug J, Oppert JM et al. Effectiveness of school- based interventions in Europe to promote healthy nutrition in children and adolescents: systematic review of published and ‘grey’ literature. Br J Nutr. 2010;103:781-97. doi:10.1017/S0007114509993370. 136. Gordon K, Dynan L, Siegel R. Healthier Choices in School Cafeterias: A Systematic Review of Cafeteria Interventions. J Pediatr. 2018;203:273-9.e2. doi:10.1016/j.jpeds.2018.07.031. 137. WHO. Global Recommendations on Physical Activity for Health. Geneva: World Health Organization; 2010 (https://www.who.int/dietphysicalactivity/publications/9789241599979/en/, accessed 31 October 2020). 138. WHO. Guidelines on physical activity, sedentary behaviour and sleep for children under 5 years of age. Geneva: World Health Organization; 2019 (https://apps.who.int/iris/handle/10665/311664, accessed 31 October 2020). 139. WHO. Global Strategy on Diet, Physical Activity and Health. Geneva: World Health Organization; 2004 (https:// www.who.int/dietphysicalactivity/strategy/eb11344/strategy_english_web.pdf, accessed 31 October 2020). 140. WHO. Global action plan on physical activity 2018–2030: more active people for a healthier world. Geneva: World Health Organization; 2018 (https://www.who.int/ncds/prevention/physical-activity/global-action- plan-2018-2030/en/, accessed 3 December 2020). 141. Hersch D, Perdue L, Ambroz T, Boucher JL. The impact of cooking classes on food-related preferences, attitudes, and behaviors of school-aged children: a systematic review of the evidence, 2003-2014. Prev Chronic Dis. 2014;11:E193. doi:10.5888/pcd11.140267. 142. WHO. WHO guideline. Sodium intake for adults and children. Geneva: World Health Organization; 2012 (https:// apps.who.int/iris/handle/10665/77985, accessed 3 December 2020). 143. WHO. WHO guideline. Potassium intake for adults and children. Geneva: World Health Organization; 2012 (https://apps.who.int/iris/handle/10665/77986, accessed 30 October 2020). 144. WHO. WHO guideline: Use of multiple micronutrient powders for point-of-use fortification of foods consumed by infants and young children aged 6–23 months and children aged 2–12 years. Geneva: World Health Organization; 2016 (https://www.who.int/nutrition/publications/micronutrients/guidelines/ mmpowders-infant6to23mons-children2to12yrs/en/, accessed 30 October 2020). 145. WHO. WHO guideline: fortification of maize flour and corn meal with vitamins and minerals. Geneva: World Health Organization; 2016 (https://apps.who.int/iris/handle/10665/251902, accessed 30 October 2020). 146. WHO. Recommendations on wheat and maize flour fortification. Meeting report: interim consensus statement. Geneva: World Health Organization; 2009 (https://www.who.int/nutrition/publications/micronutrients/wheat_ maize_fortification/en/, accessed 30 October 2020). 147. WHO. School policy framework: implementation of the WHO global strategy on diet, physical activity and health Geneva: World Health Organization; 2008 (https://www.who.int/dietphysicalactivity/schools/en/, accessed 30 October 2020). 148. Kristjansson B, Petticrew M, MacDonald B, Krasevec J, Janzen L, Greenhalgh T et al. School feeding for improving the physical and psychosocial health af disadvantaged students (Review). Cochrane Database of Systematic Reviews. 2007. doi:10.1002/14651858.CD004676.pub2. 88 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 149. De-Regil LM, Jefferds MED, Peña-Rosas JP. Point-of-use fortification of foods with micronutrient powders containing iron in children of preschool and school-age. Cochrane Database Syst Rev. 2017;11:CD009666. doi:10.1002/14651858.CD009666.pub2. 150. Aaron GJ, Dror DK, Yang Z. Multiple-Micronutrient Fortified Non-Dairy Beverage Interventions Reduce the Risk of Anemia and Iron Deficiency in School-Aged Children in Low-Middle Income Countries: A Systematic Review and Meta-Analysis (i-iv). Nutrients. 2015;7:3847-68. doi:10.3390/nu7053847. 151. Niebylski ML, Lu T, Campbell NR, Arcand J, Schermel A, Hua D et al. Healthy food procurement policies and their impact. Int J Environ Res Public Health. 2014;11:2608-27. doi:10.3390/ijerph110302608. 152. Li M, Li S, Baur LA, Huxley RR. A systematic review of school-based intervention studies for the prevention or reduction of excess weight among Chinese children and adolescents. Obes Rev. 2008;9:548-59. doi:10.1111/j.1467-789X.2008.00495.x. 153. Nørnberg TR, Houlby L, Skov LR, Peréz-Cueto FJ. Choice architecture interventions for increased vegetable intake and behaviour change in a school setting: a systematic review. Perspect Public Health. 2016;136:132-42. doi:10.1177/1757913915596017. 154. Watkins KL, Gelli A, Hamdani S, Masset E, Mersch C, Nadazdin N et al. Sensitive to Nutrition? A Literature Review of School Feeding Effects in the Child Development Lifecycle. Working Paper Series #16. Home Grown School Feeding. 2015 (https://www.semanticscholar.org/paper/Sensitive-to-nutrition-A-literature-review-of-in- Watkins-Gelli/f0c65a7257f0ea6de2914876489e624e2c90032e, accessed 3 December 2020). 155. Adelman SW, Gilligan DO, Lehrer K. How Effective are Food for Education Programs? A Critical Assessment of the Evidence from Developing Countries Washington, D.C.: International Food Policy Research Institute; 2008. doi:10.2499/0896295095FPREV9. 156. Jomaa LH, McDonnell E, Probart C. School feeding programs in developing countries: impacts on children’s health and educational outcomes. Nutr Rev. 2011;69:83-98. doi:10.1111/j.1753-4887.2010.00369.x. 157. Best C, Neufingerl N, Del Rosso JM, Transler C, van den Briel T, Osendarp S. Can multi-micronutrient food fortification improve the micronutrient status, growth, health, and cognition of schoolchildren? A systematic review. Nutr Rev. 2011;69:186-204. doi:10.1111/j.1753-4887.2011.00378.x. 158. Jensen JD, Hartmann H, de Mul A, Schuit A, Brug J, Consortium E. Economic incentives and nutritional behavior of children in the school setting: a systematic review. Nutr Rev. 2011;69:660-74. doi:10.1111/j.1753- 4887.2011.00422.x. 159. Sacco J, Lillico HG, Chen E, Hobin E. The influence of menu labelling on food choices among children and adolescents: a systematic review of the literature. Perspect Public Health. 2017;137:173-81. doi:10.1177/1757913916658498. 160. WHO. Set of recommendations on the marketing of foods and non-alcoholic beverages to children. WHA63.14. Geneva: World Health Organization; 2010 (https://www.who.int/dietphysicalactivity/publications/ recsmarketing/en/, accessed 30 October 2020). 161. WHO. A framework for implementing the set of recommendations on the marketing of foods and non-alcoholic beverages to children. Geneva: World Health Organization; 2012 (https://www.who.int/dietphysicalactivity/ framework_marketing_food_to_children/en/, accessed 31 October 2020). 162. Sadeghirad B, Duhaney T, Motaghipisheh S, Campbell NR, Johnston BC. Influence of unhealthy food and beverage marketing on children’s dietary intake and preference: a systematic review and meta-analysis of randomized trials. Obes Rev. 2016;17:945-59. doi:10.1111/obr.12445. 163. Kraak VI, Story M. Influence of food companies’ brand mascots and entertainment companies’ cartoon media characters on children’s diet and health: a systematic review and research needs. Obes Rev. 2015;16:107-26. doi:10.1111/obr.12237. 164. Folkvord F, van ‘t Riet J. The persuasive effect of advergames promoting unhealthy foods among children: A meta-analysis. Appetite. 2018;129:245-51. doi:10.1016/j.appet.2018.07.020. 89REFERENCES 165. Buchanan L, Kelly B, Yeatman H, Kariippanon K. The Effects of Digital Marketing of Unhealthy Commodities on Young People: A Systematic Review. Nutrients. 2018;10. doi:10.3390/nu10020148. 166. Boyland EJ, Nolan S, Kelly B, Tudur-Smith C, Jones A, Halford JC et al. Advertising as a cue to consume: a systematic review and meta-analysis of the effects of acute exposure to unhealthy food and nonalcoholic beverage advertising on intake in children and adults. Am J Clin Nutr. 2016;103:519-33. doi:10.3945/ajcn.115.120022. 167. Velazquez CE, Black JL, Potvin Kent M. Food and Beverage Marketing in Schools: A Review of the Evidence. Int J Environ Res Public Health. 2017;14. doi:10.3390/ijerph14091054. 168. WHO. Diarrhoeal disease. Fact sheet. Geneva: World Health Organization; 2017 (https://www.who.int/news- room/fact-sheets/detail/diarrhoeal-disease, accessed 12 November 2020). 169. WHO. Drinking water. Fact sheet. Geneva: World Health Organization; 2019 (https://www.who.int/en/news- room/fact-sheets/detail/drinking-water, accessed 31 October 2020). 170. WHO. Water, sanitation and hygiene standards for schools in low-cost settings. Edited by John Adams, Jamie Bartram, Yves Chartier, Jackie Sims. Geneva: World Health Organization; 2009 (https://www.who.int/water_ sanitation_health/publications/wsh_standards_school/en/, accessed 31 October 2020). 171. Jasper C, Le TT, Bartram J. Water and sanitation in schools: a systematic review of the health and educational outcomes. Int J Environ Res Public Health. 2012;9:2772-87. doi:10.3390/ijerph9082772. 172. WHO. Guidelines on sanitation and health. Geneva: World Health Organization; 2018 (https://www.who.int/ water_sanitation_health/publications/guidelines-on-sanitation-and-health/en/, accessed 31 October 2020). 173. UNICEF. WASH in schools. Call to action. United Nations Children’s Fund; (https://www.unicef.org/wash/schools/ washinschools_53108.html, accessed 30 October 2020). 174. Ejemot-Nwadiaro RI, Ehiri JE, Arikpo D, Meremikwu MM, Critchley JA. Hand washing promotion for preventing diarrhoea. Cochrane Database Syst Rev. 2015:CD004265. doi:10.1002/14651858.CD004265.pub3. 175. WHO. Growth reference 5-19 years. (https://www.who.int/growthref/en/, accessed 30 October 2020). 176. Ruggieri DG, Bass SB. A comprehensive review of school-based body mass index screening programs and their implications for school health: do the controversies accurately reflect the research? J Sch Health. 2015;85:61-72. doi:10.1111/josh.12222. 177. WHO. Guideline: Daily iron supplementation in infants and children. Geneva: World Health Organization; 2016 (https://www.who.int/nutrition/publications/micronutrients/guidelines/daily_iron_supp_childrens/en/, accessed 30 October 2020). 178. WHO. Guideline: Daily iron supplementation in adult women and adolescent girls. Geneva: World Health Organization; 2016 (https://www.who.int/nutrition/publications/micronutrients/guidelines/daily_iron_supp_ womenandgirls/en/, accessed 30 October 2020). 179. WHO. Guideline: Intermittent iron supplementation in preschool and school-age children. Geneva: World Health Organization; 2011 (https://www.who.int/nutrition/publications/micronutrients/guidelines/guideline_ iron_supplementation_children/en/, accessed 30 October 2020). 180. WHO. Guideline: Intermittent iron and folic acid supplementation in menstruating women. Geneva: World Health Organization; 2011 (https://www.who.int/nutrition/publications/micronutrients/guidelines/guideline_ iron_folicacid_suppl_women/en/, accessed 30 October 2020). 181. WHO. Guideline: preventive chemotherapy to control soil-transmitted helminth infections in at-risk population groups. Geneva: World Health Organization; 2017 (https://www.who.int/nutrition/publications/guidelines/ deworming/en/, accessed 31 October 2020). 182. WHO. Helminth control in school age children: a guide for managers of control programmes - 2nd edition. Geneva: World Health Organization; 2011 (https://www.who.int/neglected_diseases/resources/9789241548267/en/, accessed 31 October 2020). 90 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE 183. Salam RA, Hooda M, Das JK, Arshad A, Lassi ZS, Middleton P et al. Interventions to Improve Adolescent Nutrition: A Systematic Review and Meta-Analysis. J Adolesc Health. 2016;59:S29-S39. doi:10.1016/j. jadohealth.2016.06.022. 184. Girum T, Wasie A. The Effect of Deworming School Children on Anemia Prevalence: A Systematic Review and Meta-Analysis. Open Nurs J. 2018;12:155-61. doi:10.2174/1874434601812010155. 185. WHO. e-Library of Evidence for Nutrition Actions (eLENA). (https://www.who.int/elena/en/, accessed 12 October 2020). 186. WHO. Essential nutrition actions: mainstreaming nutrition through the life-course. Geneva: World Health Organization; 2019 (https://www.who.int/nutrition/publications/essential-nutrition-actions-2019/en/, accessed 31 October 2020). 187. WHO. Evidence-informed nutrition guidelines. (https://www.who.int/nutrition/topics/guideline-development/en/, accessed 12 October 2020). 188. WHO. NUGAG - Subgroup on Policy Actions. (https://www.who.int/nutrition/topics/guideline-development/ nugag_policyactions/en/, accessed 12 October 2020). 189. WHO. School Health Services Guideline: Guideline on school-based or school-linked health services provided by a health worker. Guideline in development. (https://www.who.int/maternal_child_adolescent/guidelines/ development/school-interventions-child-adolescent-health/en/, accessed 12 October 2020). 190. WHO. Global school-based student health survey (GSHS). (https://www.who.int/ncds/surveillance/gshs/en/, accessed 12 October 2020). 191. WHO Regional Office for Europe. Food and nutrition policy for schools: a tool for the development of school nutrition programmes in the European Region. Copenhagen: World Health Organization; 2006 (http://www. euro.who.int/en/health-topics/disease-prevention/nutrition/publications/guidance-and-tools/school-age- children-and-adolescents/food-and-nutrition-policy-for-schools-a-tool-for-the-development-of-school- nutrition-programmes-in-the-who-european-region, accessed 30 October 2020). 192. WHO Regional Office for the Western Pacific. Be smart drink water : a guide for school principals in restricting the sale and marketing of sugary drinks in and around schools. Manila: World Health Organization; 2016 (https://iris.wpro.who.int/handle/10665.1/13218, accessed 31 October 2020). 193. FAO. School Food and Nutrition Framework. Rome: Food and Agriculture Organization of the United Nations; 2019 (http://www.fao.org/3/ca4091en/ca4091en.pdf, accessed 4 December 2020). 194. Antonić Degač K, Kaić-Rak A, Laido Z, Perkovac S, Pucarin-Cvetković J, Capak K. Nutrition friendly schools initiative in Croatia. Zagreb, Croatia, 13-16.10.2010 2010 (https://www.bib.irb.hr/541715, accessed 12 November 2020). 195. Baysal SU, İnce T. Recent Developments in School-Based Health Services in Turkey. J Pediatr Res. 2018;5:60-4. doi:10.4274/jpr.98852. 196. Parmar A. Nutrition-Friendly School Initiative – The Indian experience. New Delhi, India: HRIDAY-SHAN; 2009 (http://www.jcsh-cces.ca/upload/PPT%20Parmar%20NFSI_%20Indian%20experience.ppt, accessed 4 December 2020). 197. Elkhouri Edde C, Delisle H, Dabone C, Batal M. Impact of the Nutrition-Friendly School Initiative: analysis of anthropometric and biochemical data among school-aged children in Ouagadougou. Global Health Promotion.0:1757975918789350. doi:10.1177/1757975918789350. 198. Williams J. Evaluating the impact of World Health Organization ‘Nutrition-Friendly Schools Initiative’ approaches on adolescent nutrition outcomes in rural Sri Lanka: University of Oxford; 2017 (https://ora.ox.ac.uk/objects/ uuid:0595aecf-23e7-407a-9eac-57ea4b7bc7f2, accessed 4 December 2020). 91REFERENCES WFP/Boris Heger 92 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE ANNEX I SUMMARY OF EXPERIENCE IN PILOTING, IMPLEMENTING AND EVALUATING THE NFSI IN COUNTRIES T he NFSI framework and a self-appraisal tool (featuring 41 yes/no questions based on the five components of the NFSI) were pilot tested in 18 countries to identify the appropriateness and feasibility of the content. In the initial pilot in 2006-2007, countries felt the NFSI was important but may need adaptation to local contexts. Concerns were raised about the practical application and implementation of the Initiative, particularly in low-resource settings. There were also challenges related to translating the documents into expressions that could be understood in various contexts, as raised by the 10 European pilot countries. Participants suggested that the questionnaire be validated and made more applicable to the specific country contexts. Some countries, such as Croatia (194), highlighted the need to integrate the NFSI into national nutrition policies. Subsequently, several European countries have indeed established national Nutrition-Friendly Schools policies and programmes, such as Turkey where more than 16 000 schools are implementing it (27, 195). Feedback was also provided by India, collected in 2007 on the basis of a pilot with four schools, indicating that schools did not have a written nutrition policy and there was no sustainable health education programming (196). The NFSI was later pilot tested in six schools in Benin and Burkina Faso (26). The self-appraisal by nutrition committees of pilot schools indicated that no schools had a written nutrition policy, while one (private) school had an informal policy. The assessment highlighted inadequate access of children to clean water, hygienic food and toilet facilities in over half the schools. Schools did not monitor growth regularly, due to lack of equipment. As part of the project, nutrition was integrated into the curriculum, nutrition events were held in schools, gardening and poultry raising were conducted, sanitation was improved and street vendors received training to improve hygiene and the nutritional value of food sold to school children (9). An impact evaluation of the NFSI in Burkina Faso in 2014 (197) revealed reductions in thinness and anaemia since the baseline in 2009, but without changes in stunting and with increases in overweight. However, during the same period there has been significant improvement in the nutritional status of school children in Ouagadougou, and the NFSI may only have played a modest, albeit positive, role. These pilot country experiences provided researchers with a valuable opportunity to identify best practices and challenges related to NFSI implementation within schools. For example, one important challenge related to the first component of the NFSI, “developing a school nutrition policy.” Many of the school stakeholders initially believed that policy formation was the mandate of the central government, rather than that of individual schools. However, after they started working on the additional NFSI components, they became aware of the importance of having a school nutrition policy. Other challenges related to teachers’ perceptions of their competence when teaching nutrition and the sense that 93 modifications to curricula, like school policies, fell under the remit of a central administration rather than individual schools. The pilot tests in Burkina Faso also provided the researchers with examples of best practice. For example, it was deemed easier to change the vending in schools working directly with and training the vendors. When such practices were adopted, the types of food available and the hygiene of food vendors improved. Another challenge that the NFSI pilot testing identified was the fact that some vendors were reluctant to make changes to the food that was available due to fears of losing business, which can be an important source of revenue for schools. Making changes related to school nutrition and health services was often stymied by a lack of resources or poor access to personnel from other sectors (such as health care). The authors concluded that the NFSI has the potential to mobilize schools and communities to work to improve nutrition and health, but activities need to be modified to specific school needs and contexts. Lack of resources at the school level (and household poverty) appear to create implementation challenges in low-income countries. In 2014-2015, researchers at the University of Oxford and University of Colombo worked with WHO to generate an NFSI tool for researchers to assess school environments in Sri Lanka. Unlike the NFSI self-appraisal tool, which is completed by committees within schools, this tool was designed for external evaluation. It involves a researcher conducting a semi-structured interview with school principals or a senior staff member to assess school policies and practices related to the NFSI. The Oxford team’s NFSI tool for researchers was used to assess school nutrition policies and practices as they changed over time, as part of a larger World Bank-funded project in rural Sri Lanka entitled Integrating Nutrition Promotion and Rural Development. Multilevel analyses while controlling for confounders at both the school- and student-levels found no significant associations between the implementation of NFSI actions in schools and nutrition-related outcomes among students (198). G.M.B.Akash 94 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AN NE X I I D ET AI LS O F R EV IE W S I NC LU DE D AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Aa ro n, D ro r a nd Ya ng , 2 01 5 ( 15 0) 4.1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 10 sc ho ol- ba se d s tu die s fo un d t ha t M M N fo rti fie d be ve ra ge s p ro vid ed to sch oo l-a ge ch ild re n i m pr ov ed ha em og lob in an d s er um fer rit in lev els an d r ed uc ed ris k o f a na em ia, iro n de fic ien cy an d i ro n d efi cie nc y an ae m ia. Th e r ev iew in clu de d 1 0 do ub le- bli nd ed RC T s tu die s pu bli sh ed be tw ee n 2 00 3 an d 2 01 3. M M N fo rti fie d n on -d air y be ve ra ge in te rv en tio ns to re du ce ris k o f a na em ia an d iro n d efi cie nc y i n s ch oo l-a ge ch ild re n. Ch ild re n ( n= 40 94 ) m ea n ag e r an ge 5– 18 ye ar s i n t he sch oo l s et tin g i n B an gla de sh , Bo tsw an a, In dia , N ige ria , th e P hil ipp ine s, So ut h A fri ca an d t he U nt ied Re pu bli c o f Ta nz an ia. M et a- an aly sis of 9 stu die s f ou nd th at co m pa re d to is o- ca lor ic co nt ro ls, ch ild re n w ho re ce ive d M M N fo rti fie d b ev er ag es fo r 8 w ee ks to 6 m on th s s ho we d s ign ifi ca nt im pr ov em en ts in ha em og lob in (+ 2.7 6 g /L ; 9 5%  CI : 1 .19 , 4.3 3, P= 0.0 04 , 8 st ud ies ) a nd se ru m fe rri tin (+ 15 .42 pm ol/ L; 96 % CI : 5 .73 , 2 5.1 2; P= 0.0 07 ; 8 s tu die s), an d r ed uc ed ris k o f a na em ia (R R 0.5 8; 95 %  CI : 0 .29 , 0 .88 ; P = 0.0 05 ; 6 st ud ies ), iro n d efi cie nc y ( RR : 0 .34 ; 9 5%  CI : 0 .21 , 0 .55 ; P= 0.0 02 ; 7 st ud ies ), an d i ro n d efi cie nc y a na em ia (R R: 0. 17 ; 9 5%  CI : 0 .06 , 0 .53 ; P = 0.0 2; 3 s tu die s). No si gn ifi ca nt in te rv en tio n e ffe cts of fo rti fie d be ve ra ge s w er e f ou nd fo r s er um re tin ol, zi nc or vit am in B1 2. A m od er at e b ut si gn ifi ca nt eff ec t wa s f ou nd fo r w eig ht ga in in th e i nt er ve nt ion co m pa re d w ith th e c on tro l g ro up s ( + 0.3 0 k g; 95 %  CI : 0 .01 , 0 .58 ; P = 0.0 4; 6 s tu die s), w hil e ne ith er he igh t, W AZ no r H AZ di ffe re d s ign ifi ca nt ly be tw ee n g ro up s. Ab de l R ah m an et al. , 2 01 8 ( 11 8) 3.1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 16 st ud ies , 12 of w hic h w er e s ch oo l- ba se d, of th e e ffe cti ve ne ss of ed uc at ion al an d b eh av iou ra l int er ve nt ion s i n r ed uc ing SS B int ak e f ou nd m od es t i m pa ct on SS B c on su m pt ion an d n o eff ec t o n B M I z -sc or e. Th e r ev iew in clu de d 1 6 R CT stu die s p ub lis he d b et we en 20 04 an d 2 01 5. Ed uc at ion al or be ha vio ur al int er ve nt ion s t o r ed uc e t he int ak e o f S SB s i n c hil dr en an d a do les ce nt s. Of th e 1 2 sch oo l-b as ed st ud ies , 8 us ed ed uc at ion al ap pr oa ch es al on e (e .g. cl as sro om le sso ns ) a nd 4 c om bin ed ed uc at ion w ith en vir on m en ta l in te rv en tio ns (e .g. w at er fo un ta ins , pr ov isi on of w at er bo ttl es ). Ch ild re n a nd ad ole sce nt s ag ed 4– 16 ye ar s i n E ur op e, th e U nit ed St at es an d L at in Am er ica , in sc ho ols (1 2 stu die s) or ho m e/ co m m un ity (4 st ud ies ). Of th e 1 2 s ch oo l-b as ed st ud ies , o nl y 2 st ud ies (o ne on ed uc at ion al on e a nd on e o n e du ca tio na l + en vir on m en ta l in te rv en tio ns ) p ro vid ed da ta th at co uld be po ole d i nt o a m et a- an aly sis w hic h re su lte d i n b or de rli ne si gn ifi ca nc e f or re du cti on in SS B c on su m pt ion (M D: -2 6.5 3 m l/d ay ; 9 5% CI: -5 3.7 2, 0.6 6; P= 0.0 6) an d n o e ffe ct on re du cti on of BM I z sc or es . S ix of th e 1 0 r em ain ing tr ial s re po rte d r ed uc tio ns in SS B c on su m pt ion . 95 AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ad elm an , G illi ga n an d L eh re r, 2 00 8 (1 55 ) 4.1 Th is lit er at ur e r ev iew of 20 sc ho ol- ba se d s tu die s of th e i m pa ct of fo od fo r ed uc at ion pr og ra m m es on ed uc at ion an d n ut rit ion ou tco m es fo un d e vid en ce of eff ec ts on fo od co ns um pt ion an d m icr on ut rie nt st at us , pr ov ide d t ha t i nit ial co ns um pt ion an d n ut rie nt de fic ien cie s a re id en tifi ed an d t ha t p ro gr am m es ar e ta ilo re d t o a dd re ss th es e co nd iti on s. Th e r ev iew in clu de d 2 0 stu die s o f v ar iou s d es ign (e .g. RC Ts, fie ld tri als , qu as i-e xp er im en ta l d es ign s) pu bli sh ed be tw ee n 1 98 1 an d 2 00 5. Fo od fo r e du ca tio n ( FF E) pr og ra m m es (e .g. br ea kfa st, m id- da y m ea l, m icr on ut rie nt fo rti fie d m ea ls/ sn ac ks , an im al- so ur ce fo od s). Ch ild re n i n p rim ar y s ch oo ls in low - a nd m idd le- inc om e co un tri es in Af ric a, As ia an d La tin Am er ica . Of th e 2 0 s tu die s, 12 m ea su re d i m pa ct of FF E on nu tri tio n- re lat ed ou tco m es . O f 4 st ud ies m ea su rin g c on su m pt ion re lat ed ou tco m es , 3 fo un d t ha t s ch oo l fe ed ing in cre as ed ca lor ic int ak e am on g s ch oo l-a ge d c hil dr en , p ar tic ula rly w he n ba se lin e w as w ell be low th eir ag e- or w eig ht - re co m m en de d c on su m pt ion le ve l. F ur th er m or e, 3 o ut of 4 stu die s f ou nd th at in cre as ing ac ce ss to ca lor ies ov er a su ffi cie nt pe rio d i nc re as ed bo dy siz e o r c om po sit ion (e .g. he igh t, we igh t, BM I, M UA C) . H ow ev er, pr ov isi on of m icr on ut rie nt ric h fo od s o r d ew or m ing se em ed to ha ve co nt rib ut ed to im pr ov em en ts in so m e o ut co m es (e .g. he igh t, M UA C) . Im pr ov em en ts we re ty pic all y s m all , bu t t he au th or s n ot e t ha t t he eff ec t m ay ha ve be en m iti ga te d b y i nc re as ed ac tiv ity le ve ls. Iro n sta tu s w as im pr ov ed as re su lt of fo rti fie d s ch oo l m ea ls in 3 o ut of 4 stu die s, vit am in B1 2 s ta tu s wa s i m pr ov ed as re su lt of m ea t-b as ed m ea ls im pr ov ed in 1 stu dy , v ita m in A w as im pr ov ed as re su lt of vi ta m in A f or tifi ca tio n i n 1 ou t o f 2 s tu die s, pr ev ale nc e o f io din e d efi cie nc y w as re du ce d a s r es ult of io din e f or tifi ca tio n i n 1 st ud y. 96 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Af sh in et al ., 2 01 5 (4 0) 1.1 , 4 .1 Th is re vie w of th e e vid en ce fo r s pe cifi c p oli cie s t o im pr ov e d iet ar y h ab its an d r ed uc e c ar dio va sc ula r an d m et ab oli c r isk fa cto rs, wh ich in clu de d 7 6 s tu die s o f sch oo l p ro cu re m en t p oli cie s, co nc lud ed th at ch an ge s in sch oo l p ro cu re m en t po lic ies ap pe ar eff ec tiv e f or eit he r in cre as ing he alt hy or re du cin g u nh ea lth y c ho ice s, wh ile se tti ng of nu tri tio n sta nd ar ds ha s l es s c on sis te nt be ne fit s. Th e r ev iew in clu de d s tu die s of va rio us de sig ns ev alu at ing th e e ffe cti ve ne ss of po lic y str at eg ies to im pr ov e d iet pu bli sh ed af te r 1 98 0. Fo r sch oo l p ro cu re m en t p oli cie s, 76 st ud ies of ra nd om ize d o r qu as i-e xp er im en ta l s tu die s we re in clu de d. In te rv en tio ns in clu de d m as s m ed ia ca m pa ign s, fo od an d m en u l ab ell ing , ta xa tio n a nd su bs idi es , loc al bu ilt en vir on m en t, sch oo l p ro cu re m en t po lic ies , w or ks ite w ell ne ss pr og ra m m es , a nd m ar ke tin g s ta nd ar ds . S ch oo l pr oc ur em en t p oli cie s, inc lud ing m ult ico m po ne nt stu die s i f t he se re gu lat ion s we re a m ajo r c om po ne nt of th e i nt er ve nt ion , t o 1 ) inc re as e i n a va ila bil ity of he alt hf ul fo od s a nd be ve ra ge s ( e.g . F &V , low -fa t s na ck s, m ilk , w ho le gr ain pr od uc ts, w at er ), 2) im ple m en t s ta nd ar ds on av ail ab ilit y o f u nh ea lth fu l fo od s a nd be ve ra ge s ( e.g . re str ict ion s o r b an s, nu tri en t sta nd ar ds fo r c om pe tit ive fo od s i n c afe te ria s, ve nd ing m ac hin es , s ch oo l s to re s, sn ac k b ar s, or sn ac k t ru ck s - a bo ut ha lf o f t he se w er e loc al pr og ra m m es ; o th er s we re ba se d o n c ity , s ta te , or na tio na l p oli cie s), an d/ or 3) im ple m en t n ut rit ion sta nd ar ds fo r s ch oo l m ea ls (b as ed on e. g. fo od , nu tri en ts, po rti on si ze ). Al l p op ula tio n g ro up s wo rld wi de . S ch oo l-b as ed stu die s l ar ge ly co ve re d ch ild re n i n s ch oo ls in Eu ro pe an d i n N or th Am er ica . Am on g 3 1 s tu die s t o i nc re as e i n a va ila bil ity of he alt hy fo od s a nd be ve ra ge s, int er ve nt ion s t ha t aim ed to in cre as e F &V in ta ke ap pe ar ed eff ec tiv e. Am on g 2 6 s tu die s o f t he eff ec t o f r es tri cti ng un he alt hy fo od s a nd be ve ra ge s i n s ch oo ls, int er ve nt ion s a pp ea re d g en er all y e ffe cti ve in re du cin g i nt ak e o f u nh ea lth y c om pe tit ive fo od s a nd be ve ra ge s a nd de cre as ing in cid en ce an d p re va len ce of ov er we igh t/o be sit y, wi th go ve rn m en ta l a nd sc ho ol po lic ies be ing m or e eff ec tiv e t ha n v olu nt ar y p ro gr am m es . A m on g 22 st ud ies of th e e ffe ct of st an da rd s f or sc ho ol m ea ls fo r l un ch an d/ or br ea kfa st, re su lts w er e inc on sis te nt ac ro ss stu die s a nd no co ns ist en t pa tte rn s w er e i de nt ifi ed to ex pla in th e he te ro ge ne ou s fi nd ing s. T he au th or s c on clu de d th at ch an ge s i n s ch oo l p ro cu re m en t p oli cie s ap pe ar eff ec tiv e f or ei th er in cre as ing he alt hy or re du cin g u nh ea lth y c ho ice s, wh ile se tti ng of nu tri tio n s ta nd ar ds ha ve le ss co ns ist en t b en efi ts. 97ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Aj ie an d Ch ap m an - No va ko fsk i, 2 01 4 (1 26 ) 3.1 Th is sy ste m at ic re vie w of 15 st ud ies of co m pu te r- ba se d n ut rit ion ed uc at ion int er ve nt ion s, 10 of wh ich w er e s ch oo l-b as ed , co nc lud ed th at su ch pr og ra m m es m ay ha ve po te nt ial to im pr ov e d iet ar y ha bit s a m on g a do les ce nt s. Th e r ev iew in clu de d 1 5 stu die s i nc lud ing RC Ts, qu as i- ex pe rim en ta l t ria ls an d R CT s, pu bli sh ed be tw ee n 2 00 3 an d 2 01 3. Co m pu te r-b as ed ed uc at ion or be ha vio ur al int er ve nt ion s via a sta tio na ry or la pt op co m pu te r t ar ge tin g dis or de re d e at ing be ha vio ur s inc lud ing ov er ea tin g a nd bin ge ea tin g, he alt hy ea tin g an d l ife sty le be ha vio ur s, F& V c on su m pt ion , S SB co ns um pt ion or fa t i nt ak e Ad ole sce nt s a ge d 1 2– 18 ye ar s i n t he U nit ed St at es an d Eu ro pe ; 1 0 o f t he 15 st ud ies we re in th e s ch oo l s et tin g. M os t s tu die s c on du cte d i n t he sc ho ol se tti ng fo un d s ign ifi ca nt re su lts in at le as t o ne of th e die ta ry or an th ro po m et ric ou tco m e m ea su re s po st- int er ve nt ion , a lbe it so m e o nl y a m on g at -ri sk su bg ro up s o r a m on g g irl s. T he au th or s co nc lud ed th at co m pu te r-b as ed nu tri tio n ed uc at ion pr og ra m m es ha ve th e p ot en tia l t o pe rsu ad e a do les ce nt s t o i m pr ov e o ve ra ll d iet ar y ha bit s a nd cu rb th e e xp an sio n o f o ve rw eig ht an d ob es ity . Al -K hu da iry et al ., 20 17 (6 4)  1. 1 Th is Co ch ra ne re vie w an d m et a- an aly sis of 44 ob es ity tr ea tm en t s tu die s, 7 o f w hic h w er e s ch oo l- ba se d, fo un d l ow qu ali ty ev ide nc e t ha t d iet , P A a nd be ha vio ur al int er ve nt ion s re du ce m ea su re s o f B M I an d f ou nd m od er at e q ua lit y ev ide nc e t ha t t he y r ed uc e we igh t i n o ve rw eig ht or ob es e a do les ce nt s. T he re w as no di ffe re nc e o bs er ve d i n su bg ro up an aly se s b y s et tin g. Th e r ev iew in clu de d 4 4 R CT s, pu bli sh ed be tw ee n 1 96 8 an d 2 01 5. Di et , P A a nd /o r b eh av iou r ch an ge in te rv en tio ns (e .g. ed uc at ion , c ou ns ell ing , he alt hy fo od ch oic es , en vir on m en ta l c on tro l) fo r t re at ing ov er we igh t or ob es ity in ad ole sce nt s. Fiv e o f 7 st ud ies w ith a PA co m po ne nt in th e s ch oo l se tti ng al so in clu de d a die ta ry co m po ne nt . Ov er we igh t o r o be se ad ole sce nt s a ge d 1 2– 17 .5 ye ar s ( n= 47 81 ) i n m ult ipl e se tti ng s ( 7 o ut of 44 st ud ies we re co nd uc te d i n t he sc ho ol se tti ng ) w or ldw ide , w ith m os t inc lud ed st ud ies fr om th e Un ite d S ta te s a nd Eu ro pe . Th e p oo led re su lts of al l s ub gr ou ps sh ow ed th at be ha vio ur ch an ge in te rv en tio n p ro gr am m es ga ve si gn ifi ca nt re du cti on s i n B M I ( M D: -1 .18 kg /m 2; 95 % CI : - 1.6 7, -0 .69 ; 2 77 4 p ar tic ipa nt s; 28 tr ial s), BM I z -sc or e ( M D: -0 .13 ; 9 5% CI : -0 .21 , - 0.0 5; 23 99 pa rti cip an ts; 20 tr ial s) an d we igh t c ha ng e ( M D: -3 .67 kg ; 9 5% CI : - 5.2 1, -2 .13 ; 1 99 3 p ar tic ipa nt s; 20 tr ial s). Su bg ro up an aly se s g av e n o d iff er en ce s i n B M I a nd BM I z-s co re be tw ee n t he di ffe re nt se tti ng s. In th e sch oo l-b as ed st ud ies , t he eff ec ts we re po sit ive bu t n ot si gn ifi ca nt fo r b ot h B M I ( -0 .91 kg /m 2; 95 % CI : - 1.9 7, 0.1 5; 61 3 p ar tic ipa nt s; 7 t ria ls) an d BM I z -sc or e ( -0 .7 kg /m 2; 95 % CI : - 2.0 6, 0.6 6; 15 0 pa rti cip an ts; 2 tri als ). 98 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Al oia et al ., 2 01 6 (7 8) 1.1 , 2 .2, 3. 1 Th is sy ste m at ic re vie w of sch oo l-b as ed in te rv en tio ns ta rg et ing F& V c on su m pt ion in th e U S w er e s uc ce ssf ul in 8 o f 1 4 s tu die s, wi th th e m os t p ro m isi ng co m po ne nt s be ing fo od se rv ice st aff inv olv em en t a nd nu tri tio n ed uc at ion . Th e r ev iew in clu de d 14 co nt ro lle d t ria ls (5 ra nd om ize d a nd 9 no n- ra nd om ize d) pu bli sh ed be tw ee n 2 00 6 a nd 20 14 . Sc ho ol- ba se d i nt er ve nt ion s to in cre as e F &V co ns um pt ion inc lud ing pa re nt al inv olv em en t, te ac he r inv olv em en t, nu tri tio n ed uc at ion , fo od se rv ice an d s ch oo l g ar de ns . M os t int er ve nt ion s w er e p ar t o f a m ult ico m po ne nt pr og ra m m e wh ile on e w as si ng le co m po ne nt (i .e. in te ra cti ve bo ok s). Ch ild re n i n k ind er ga rte n th ro ug h 8 th gr ad e i n t he Un ite d S ta te s. Ov er all , 8 ou t o f 1 4 s tu die s ( 13 of w hic h w er e m ult ico m po ne nt ) s ho we d s ign ifi ca nt in cre as es in co ns um pt ion of F& V a m on g s ch oo l c hil dr en . Th e m os t p ro m isi ng in te rv en tio n c om po ne nt wa s f oo d s er vic e s ta ff inv olv em en t a nd nu tri tio n ed uc at ion , w he re as te ac he r in vo lve m en t, pa re nt inv olv em en t o r g ar de ns w er e n ot as so cia te d w ith po sit ive fin din gs . L ike wi se , in te rv en tio n d ur at ion , ty pe of th eo ry us ed , s ty le of in te rv en tio n lea de rsh ip or ai m s t o a ffe ct an te ce de nt s o f F &V co ns um pt ion w er e n ot co rre lat ed w ith hi gh er co ns um pt ion . Av er y, Bo sto ck an d M cC ull ou gh , 20 15 (1 34 ) 3.1 Th is sy ste m at ic re vie w of 8 s tu die s, 7 o f w hic h w er e sch oo l-b as ed , s ug ge ste d th at sc ho ol- ba se d e du ca tio n pr og ra m m es fo cu sin g o n re du cin g S SB co ns um pt ion an d i nc lud ing fo llo w- up m od ule s o ffe r o pp or tu nit ies fo r im ple m en tin g e ffe cti ve , su sta ina ble in te rv en tio ns . Th e r ev iew in clu de d 8 RC Ts pu bli sh ed be tw ee n 2 00 4 an d 2 01 3. In te rv en tio ns to re du ce SS B c on su m pt ion (e .g. sch oo l-b as ed ed uc at ion al pr og ra m m es , s ch oo l- ba se d e du ca tio na l pr og ra m m e c om bin ed w ith en vir on m en ta l c ha ng e, sch oo l-d ist rib ut ed dr ink s a nd ho m e- de liv er ed dr ink s). Ch ild re n a ge d 8 –1 5 y ea rs in Eu ro pe , B ra zil an d t he U nit ed St at es . Six st ud ies ac hie ve d s ign ifi ca nt (P < 0.0 5) re du cti on s i n S SB in ta ke , a lth ou gh th is wa s no t a lw ay s s us ta ine d. Fo ur st ud ies fo cu sin g o n ed uc at ion al ap pr oa ch es re du ce d t he ris k o f b ein g ov er we igh t, th ou gh on e o f t he se al so pr ov ide d a re pla ce m en t d rin k. Th e a ut ho rs co nc lud ed th at sch oo l-b as ed ed uc at ion pr og ra m m es to re du ce SS B c on su m pt ion th at in clu de fo llo w- up m od ule s off er op po rtu nit ies fo r im ple m en tin g e ffe cti ve , su sta ina ble in te rv en tio ns . P ee r s up po rt an d ch an gin g t he sc ho ol en vir on m en t ( e.g . p ro vid ing wa te r o r r ep lac em en t d rin ks ) t o s up po rt ed uc at ion al pr og ra m m es co uld im pr ov e t he ir eff ec tiv en es s. 99ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ba ile y, Dr um m on d a nd W ar d, 20 19 (1 22 ) 3.1 , 3 .3 Th is sy ste m at ic lit er at ur e re vie w of 44 sc ho ol- ba se d stu die s f ou nd th at fo od lit er ac y i n a do les ce nt s im pr ov es di et ar y i nt ak e, an d th at ha nd s-o n c oo kin g s kil ls an d u se of ch efs or sc ho ol ga rd en s w er e b en efi cia l. Th e r ev iew in clu de d 4 4 stu die s, 37 of w hic h w er e qu an tit at ive st ud ies (1 6 cro ss- se cti on al, 13 qu as i- ex pe rim en ta l, 1 pr e- an d po st- te st de sig n, 1 RC T, 1 lon git ud ina l c oh or t s tu dy , 1 o bs er va tio na l, 4 m ixe d m et ho ds ) a nd 7 qu ali ta tiv e stu die s, pu bli sh ed be tw ee n 19 90 an d 2 01 7. In te rv en tio ns to in cre as e f oo d lit er ac y a m on g a do les ce nt s i n se co nd ar y s ch oo ls, in clu din g pr ac tic al co ok ing cl as se s, re ad ing nu tri tio na l la be ls, sch oo l-b as ed ga rd en s, sch oo l s up ple m en ta ry fo od pr og ra m m es , t ec hn olo gy int er ve nt ion ), lec tu re s a bo ut nu tri tio na l e du ca tio n a nd fo od sa fet y. Du ra tio n r an ge d fro m 1 we ek to 10 ye ar s. Ad ole sce nt s a ge d 1 0– 19 ye ar s in se co nd ar y s ch oo ls, m os tly in Au str ali a, Eu ro pe an d t he Un ite d S ta te s, bu t a lso in Ca na da , C hin a, Ira n, So ut h Af ric a, In dia an d K en ya . Ad ole sce nt s w ith gr ea te r n ut rit ion al kn ow led ge an d f oo d s kil ls sh ow ed he alt hie r d iet ar y pr ac tic es . O ut of 9 stu die s m ea su rin g e ffe ct on kn ow led ge , 8 sh ow ed a po sit ive im pa ct on die ta ry in ta ke . H ow ev er, th er e w as a m ixe d as so cia tio n b et we en fo od lit er ac y a nd lo ng -te rm he alt hy di et ar y b eh av iou r. O f p ro gr am m at ic int er ve nt ion s, 4 o ut of 5 stu die s o n s ch oo l- ba se d g ar de nin g s ho we d a po sit ive im pa ct on F& V c on su m pt ion . T he 4 stu die s t ha t u til ize d ex pe rim en ta l c oo kin g a nd nu tri tio na l e du ca tio n pr og ra m m es le d b y c he f in str uc to rs sh ow ed po sit ive re su lts in ov er all co ok ing co nfi de nc e, co ok ing sk ills an d t as tin g n ew fo od s. T he qu ali ta tiv e s tu die s f ou nd th at ad ole sce nt s a nd ho m e e co no m ics te ac he rs va lue fo od lit er ac y sk ills , b ut th e s ch oo l fo od en vir on m en ts ar e se ldo m co m pr eh en siv ely su pp or tiv e o f f oo d lit er ac y. 100 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Be st et al ., 2 01 1 (1 57 ) 4.1 Th is sy ste m at ic re vie w of 12 sch oo l-b as ed st ud ies sh ow ed a p os iti ve eff ec t o f p ro vid ing M M N fo rti fie d f oo ds to ch ild re n o n m icr on ut rie nt sta tu s, bu t e vid en ce fo r th e e ffe cts on gr ow th , m or bid ity an d c og nit ion w er e inc on sis te nt . Th e r ev iew in clu de d 1 2 stu die s ( 5 R CT s, 6 q ua si- ex pe rim en ta l c on tro lle d cli nic al tri als , a nd 1 CB A) . St ud ies w er e c on du cte d i n 19 99 or la te r. Pr ov isi on of M M N fo rti fie d fo od s, us ing ca rri er s s uc h a s be ve ra ge s ( 6 s tu die s), m ilk pr od uc ts (2 st ud ies ), bis cu its (2 st ud ies ), or se as on ing po wd er /sa lt (2 st ud ies ) fo rti fie d w ith a co m bin at ion of at le as t i ro n a nd vi ta m in A o r b et a c ar ot en e a nd of te n iod ine , z inc , B -v ita m ins an d/ or vi ta m in C. Fo rti fie d f oo ds us ua lly pr ov ide d b et we en 50 an d 1 00 % of RD A f or th es e m icr on ut rie nt s. 11 st ud ies co m pa re d p ro vis ion of M M N fo rti fie d f oo d t o u nf or tifi ed fo od , a nd 1 stu dy co m pa re d M M N fo rti fie d f oo d t o s ing le- m icr on ut rie nt fo rti fie d f oo d. Ch ild re n a ge d 6 –1 8 y ea rs, pr im ar ily in th e s ch oo l s et tin g (1 1 o f 1 2 s tu die s), m os tly in low - a nd m idd le- inc om e co un tri es am on g c hil dr en wi th po or nu tri tio na l s ta tu s (B an gla de sh , B ot sw an a, In dia , In do ne sia , P hil ipp ine s, M or oc co , S ou th Af ric a, Th ail an d, Un ite d R ep ub lic of Ta nz an ia, Vi et N am ). On e s tu dy wa s c on du cte d i n a H IC am on g ch ild re n w ith go od nu tri tio na l sta tu s ( Au str ali a) . M M N fo rti fic at ion of fo od s ( wi th th e e xc ep tio n of zi nc ) c on sis te nt ly led to im pr ov em en t of sh or t-t er m m icr on ut rie nt st at us w ith sig nifi ca nt ly im pr ov ed m icr on ut rie nt st at us in 10 of 11 in te rv en tio n g ro up . T he im pa cts of M M N fo rti fic at ion on fu nc tio na l h ea lth ou tco m es (g ro wt h, m or bid ity , c og nit ive de ve lop m en t) we re in co ns ist en t. Ou t o f 7 st ud ies m ea su rin g an th ro po m et ric ou tco m es , 4 st ud ies re po rte d a s ign ifi ca nt be ne fic ial eff ec t o n w eig ht ga in, inc re as ed BM I o r M UA C, bu t o nl y 2 sh ow ed an eff ec t o n h eig ht . 4 of 7 stu die s s ho we d s om e be ne fic ial eff ec ts on ch ild re n’s he alt h, in clu din g re du ce d i nc ide nc e o r d ur at ion or di ar rh oe a a nd re sp ira to ry di se as es . 2 ou t o f 3 st ud ies m ea su rin g im pa ct on di ar rh oe a s ho we d s ign ifi ca nt im pr ov em en t, bu t o nl y 1 ou t o f 3 m ea su rin g im pa ct on re sp ira to ry re lat ed di se as es sh ow ed sig nifi ca nt im pr ov em en t. Im pa cts on co gn iti ve pe rfo rm an ce w er e i nc on sis te nt an d v ar ied be tw ee n c og nit ive do m ain s. Ho we ve r, t he re w er e co ns ist en t p os iti ve re su lts on w or kin g m em or y ac ro ss stu die s, wi th 4 of 6 stu die s r ep or tin g so m e b en efi cia l e ffe ct. Th e a ut ho rs co nc lud e t ha t fo rti fie d s ch oo l m ea ls ca n p ro vid e a sa fe an d eff ec tiv e n ut rit ion al int er ve nt ion an d m ak e a va lua ble co nt rib ut ion to th e n ut rit ion al ne ed s o f sch oo l c hil dr en . 101ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Bl ac k, M at vie nk o- Sik ar an d K ea rn ey , 20 17 (1 05 ) 2.2 , 4 .2, 4. 11 Th is sy ste m at ic re vie w of 39 sc ho ol- ba se d d iet an d PA in te rv en tio ns fo un d ins uffi cie nt ev ide nc e t o de te rm ine th e i m pa ct of pa re nt in vo lve m en t i n t he se int er ve nt ion s. Th e r ev iew in clu de d 3 9 stu die s, all of w hic h w er e RC Ts. Th e d at es of pu bli ca tio n ra ng ed fr om 19 96 to 20 14 , wi th a m ajo rit y p ub lis he d i n th e l as t 1 0 y ea rs. M os t i nt er ve nt ion s w er e m ult ipl e c om po ne nt nu tri tio n pr og ra m m es th at in clu de d co m bin at ion s o f c las s cu rri cu lum ac tiv iti es , s ch oo l fo od se rv ice m od ifi ca tio ns , ho m e a cti vit ies , e nh an ce d PE /P A a nd st ra te gie s t o en ga ge pa re nt s/f am ilie s. Th er e w er e a lso 4 stu die s o f sch oo l m ea l p ro gr am m es , 1 s ch oo l g ar de nin g s tu dy , 1 sch oo l fr uit pr og ra m m e s tu dy an d 2 sc ho ol ca nt ee n s tu die s. Th e m ajo rit y o f t he st ud ies (8 2% ) w er e u nd er ta ke n in sch oo ls (2 5 s tu die s) or pr es ch oo ls (6 st ud ies ) i n th e U nit ed St at es , E ur op e, Au str ali a a nd N ew Ze ala nd . Th e p ar tic ipa nt s i n t he pr im ar y sch oo l-b as ed pr og ra m m es we re pr ed om ina nt ly ag ed 8– 12 ye ar s. Th er e w er e s ub sta nt ial an d s ta tis tic all y sig nifi ca nt im pr ov em en ts in die ta ry in ta ke in on ly 31 % of ou tco m es as se sse d— all re lat ed to in cre as ed F& V o r d ec re as ed fa t i nt ak e. In pr es ch oo l/s ch oo l-b as ed st ud ies , 7 ou t o f 1 3 stu die s w ith a pa re nt in vo lve m en t c om po ne nt vs 4 ou t o f 7 st ud ies w ith no su ch co m po ne nt re po rte d i nc re as ed F& V i nt ak e, es pe cia lly fru it, w he re as 4 ou t o f 8 st ud ies w ith a pa re nt inv olv em en t c om po ne nt vs 1 ou t o f 3 st ud ies wi th no su ch co m po ne nt re po rte d d ec re as ed fat in ta ke . T his le d t he au th or s t o c on clu de th at ov er all th er e w as in su ffi cie nt ev ide nc e t o de te rm ine th e i m pa ct of in vo lvi ng pa re nt s i n sch oo l/p re sch oo l n ut rit ion pr og ra m m es . T he au th or s a lso no te d t ha t t he m os t e ffe cti ve sc ho ol- ba se d p ro gr am m es ha d i nc or po ra te d r ole -m od els inc lud ing pe er s, te ac he rs an d h er oic fig ur es , re wa rd s, an d i nc re as ed ac ce ss to he alt hy fo od s. Bl eic h e t a l., 20 18 (8 0) 1.1 , 2 .2 Th is sy ste m at ic re vie w of 56 stu die s i n v ar iou s s et tin gs , o f wh ich 47 w er e c on du cte d i n sch oo ls, pr es ch oo ls or af te r sch oo l, f ou nd m os t s up po rt fo r e ffe cti ve ne ss of sc ho ol- ba se d i nt er ve nt ion s t ha t co m bin ed di et an d P A a nd inc lud ed a se co nd ar y h om e se tti ng . Th e r ev iew in clu de d 5 6 stu die s, RC Ts an d q ua si- ex pe rim en ta l s tu die s o r na tu ra l e xp er im en ts wi th a c on tro l g ro up , p ub lis he d be tw ee n 2 01 3 a nd 20 17 . In te rv en tio ns de sig ne d t o pr ev en t o r m an ag e w eig ht ga in th ro ug h d iet an d/ or PA (e .g. ed uc at ion al, en vir on m en ta l, p ar en ta l inv olv em en t). Ch ild re n a nd ad ole sce nt s ( ag ed 2– 19 ye ar s) in 56 st ud ies in m ult ipl e s et tin gs in clu din g sch oo ls or af te r s ch oo l ( 41 stu die s) an d p re sch oo ls (6 stu die s) in th e A m er ica s, Eu ro pe , C hin a, In dia , R ep ub lic of Ko re a a nd Au str ali a Of 24 sc ho ol- ba se d R CT st ud ies , 1 7 s ho we d sig nifi ca nt im pr ov em en t f or at le as t o ne ad ipo sit y- re lat ed ou tco m e ( BM I, B M I z -sc or e, BM I p er ce nt ile , b od y f at pe rce nt ag e, sk inf old th ick ne ss, w ais t c irc um fer en ce , o r p re va len ce of ov er we igh t o r o be sit y) . A ll o f t he sc ho ol- ba se d RC Ts th at re po rte d p os iti ve re su lts in clu de d a se co nd ar y h om e s et tin g a nd us ed co m bin ed die t a nd PA in te rv en tio ns , in clu din g i nt er ve nt ion ele m en ts su ch as en ha nc ed or le ng th en ed PA tim e, im ple m en ta tio n o f a nu tri tio n e du ca tio n cu rri cu lum , a nd im pr ov em en t i n p ar tic ipa nt s’ se lf- re gu lat ion an d e ffi ca cy . 102 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Bl ac k, M at vie nk o- Sik ar an d K ea rn ey , 20 17 (1 05 ) 2.2 , 4 .2, 4. 11 Th is sy ste m at ic re vie w of 39 sc ho ol- ba se d d iet an d PA in te rv en tio ns fo un d ins uffi cie nt ev ide nc e t o de te rm ine th e i m pa ct of pa re nt in vo lve m en t i n t he se int er ve nt ion s. Th e r ev iew in clu de d 3 9 stu die s, all of w hic h w er e RC Ts. Th e d at es of pu bli ca tio n ra ng ed fr om 19 96 to 20 14 , wi th a m ajo rit y p ub lis he d i n th e l as t 1 0 y ea rs. M os t i nt er ve nt ion s w er e m ult ipl e c om po ne nt nu tri tio n pr og ra m m es th at in clu de d co m bin at ion s o f c las s cu rri cu lum ac tiv iti es , s ch oo l fo od se rv ice m od ifi ca tio ns , ho m e a cti vit ies , e nh an ce d PE /P A a nd st ra te gie s t o en ga ge pa re nt s/f am ilie s. Th er e w er e a lso 4 stu die s o f sch oo l m ea l p ro gr am m es , 1 s ch oo l g ar de nin g s tu dy , 1 sch oo l fr uit pr og ra m m e s tu dy an d 2 sc ho ol ca nt ee n s tu die s. Th e m ajo rit y o f t he st ud ies (8 2% ) w er e u nd er ta ke n in sch oo ls (2 5 s tu die s) or pr es ch oo ls (6 st ud ies ) i n th e U nit ed St at es , E ur op e, Au str ali a a nd N ew Ze ala nd . Th e p ar tic ipa nt s i n t he pr im ar y sch oo l-b as ed pr og ra m m es we re pr ed om ina nt ly ag ed 8– 12 ye ar s. Th er e w er e s ub sta nt ial an d s ta tis tic all y sig nifi ca nt im pr ov em en ts in die ta ry in ta ke in on ly 31 % of ou tco m es as se sse d— all re lat ed to in cre as ed F& V o r d ec re as ed fa t i nt ak e. In pr es ch oo l/s ch oo l-b as ed st ud ies , 7 ou t o f 1 3 stu die s w ith a pa re nt in vo lve m en t c om po ne nt vs 4 ou t o f 7 st ud ies w ith no su ch co m po ne nt re po rte d i nc re as ed F& V i nt ak e, es pe cia lly fru it, w he re as 4 ou t o f 8 st ud ies w ith a pa re nt inv olv em en t c om po ne nt vs 1 ou t o f 3 st ud ies wi th no su ch co m po ne nt re po rte d d ec re as ed fat in ta ke . T his le d t he au th or s t o c on clu de th at ov er all th er e w as in su ffi cie nt ev ide nc e t o de te rm ine th e i m pa ct of in vo lvi ng pa re nt s i n sch oo l/p re sch oo l n ut rit ion pr og ra m m es . T he au th or s a lso no te d t ha t t he m os t e ffe cti ve sc ho ol- ba se d p ro gr am m es ha d i nc or po ra te d r ole -m od els inc lud ing pe er s, te ac he rs an d h er oic fig ur es , re wa rd s, an d i nc re as ed ac ce ss to he alt hy fo od s. Bl eic h e t a l., 20 18 (8 0) 1.1 , 2 .2 Th is sy ste m at ic re vie w of 56 stu die s i n v ar iou s s et tin gs , o f wh ich 47 w er e c on du cte d i n sch oo ls, pr es ch oo ls or af te r sch oo l, f ou nd m os t s up po rt fo r e ffe cti ve ne ss of sc ho ol- ba se d i nt er ve nt ion s t ha t co m bin ed di et an d P A a nd inc lud ed a se co nd ar y h om e se tti ng . Th e r ev iew in clu de d 5 6 stu die s, RC Ts an d q ua si- ex pe rim en ta l s tu die s o r na tu ra l e xp er im en ts wi th a c on tro l g ro up , p ub lis he d be tw ee n 2 01 3 a nd 20 17 . In te rv en tio ns de sig ne d t o pr ev en t o r m an ag e w eig ht ga in th ro ug h d iet an d/ or PA (e .g. ed uc at ion al, en vir on m en ta l, p ar en ta l inv olv em en t). Ch ild re n a nd ad ole sce nt s ( ag ed 2– 19 ye ar s) in 56 st ud ies in m ult ipl e s et tin gs in clu din g sch oo ls or af te r s ch oo l ( 41 stu die s) an d p re sch oo ls (6 stu die s) in th e A m er ica s, Eu ro pe , C hin a, In dia , R ep ub lic of Ko re a a nd Au str ali a Of 24 sc ho ol- ba se d R CT st ud ies , 1 7 s ho we d sig nifi ca nt im pr ov em en t f or at le as t o ne ad ipo sit y- re lat ed ou tco m e ( BM I, B M I z -sc or e, BM I p er ce nt ile , b od y f at pe rce nt ag e, sk inf old th ick ne ss, w ais t c irc um fer en ce , o r p re va len ce of ov er we igh t o r o be sit y) . A ll o f t he sc ho ol- ba se d RC Ts th at re po rte d p os iti ve re su lts in clu de d a se co nd ar y h om e s et tin g a nd us ed co m bin ed die t a nd PA in te rv en tio ns , in clu din g i nt er ve nt ion ele m en ts su ch as en ha nc ed or le ng th en ed PA tim e, im ple m en ta tio n o f a nu tri tio n e du ca tio n cu rri cu lum , a nd im pr ov em en t i n p ar tic ipa nt s’ se lf- re gu lat ion an d e ffi ca cy . AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Br am an te et al ., 20 19 (3 5) 1.1 , 4 .1, 4. 2 Th is sy ste m at ic re vie w of 33 na tu ra l e xp er im en ts fo r c hil dh oo d o be sit y pr ev en tio n a nd co nt ro l, 2 4 of w hic h w er e s ch oo l-b as ed , fo un d t ha t s ch oo l-b as ed po lic ies fo cu sin g o n b ot h th e f oo d a nd be ve ra ge an d P A e nv iro nm en ts (ve rsu s t ar ge tin g o nl y on e) co ns ist en tly sh ow ed im pr ov em en ts in BM I. Th e r ev iew in clu de d 3 3 na tu ra l e xp er im en t s tu die s, pu bli sh ed be tw ee n 2 00 9 an d 2 01 7. Of 24 sc ho ol- ba se d s tu die s, 14 fo cu se d o n t he fo od / be ve ra ge en vir on m en t, 6 o n th e P A/ bu ilt en vir on m en t a nd 4 o n m ult ipl e e nv iro nm en ts. Am on g t he 29 st ud ies co nd uc te d i n t he U nit ed St at es , 3 5% ev alu at ed lo ca l po lic ies , 3 1% st at e/ re gio na l po lic ies (e .g. co m pe tit ive fo od la ws an d s ch oo l- dis tri ct fo od po lic ies ), 24 % fed er al- lev el po lic ies (e .g. Ch ild N ut rit ion an d S pe cia l Su pp lem en ta l N ut rit ion Pr og ra m fo r W om en , In fan ts, an d C hil dr en Re au th or iza tio n Ac t), an d 1 0% no n- go ve rn m en ta l p oli cie s. Ch ild re n i n t he U nit ed St at es (2 9 s tu die s), Ca na da (2 st ud ies ) a nd Au str ali a (2 st ud ies ); 73 % of st ud ies we re co nd uc te d i n t he sc ho ol se tti ng (2 4 s tu die s) ac ro ss se ve ra l g ra de le ve ls (1 in clu de d ch ild ca re se tti ng , 1 2 s tu die s inc lud ed el em en ta ry sc ho ol, 13 inc lud ed m idd le sch oo l, a nd 11 inc lud ed hi gh sc ho ol) . Eig ht of 14 sc ho ol- ba se d s tu die s t ha t f oc us ed on th e f oo d a nd be ve ra ge en vir on m en t a ch iev ed fav ou ra ble BM I o ut co m es . A ll 4 st ud ies th at to ok pla ce in a sch oo l s et tin g a nd fo cu se d o n b ot h th e f oo d/ be ve ra ge an d P A/ bu ilt en vir on m en ts, ha d f av ou ra ble BM I o ut co m es . T wo of th es e stu die s t ha t a lso fo cu se d o n h ea lth y m es sa gin g ac hie ve d t he la rg es t i m pr ov em en t i n B M I. Fu rth er m or e, 3 o f 4 sc ho ol- ba se d s tu die s t ha t fo cu se d o n t he fo od an d b ev er ag e e nv iro nm en t sh ow ed re du ce d S SB co ns um pt ion . T he se 3 stu die s e va lua te d s ta te - o r s ch oo l-w ide po lic ies to de cre as e a cce ss to SS Bs . O ne of 3 sch oo l-b as ed stu die s t ha t f oc us ed on th e f oo d a nd be ve ra ge en vir on m en t s ho we d i nc re as ed F& V c on su m pt ion as re su lt of im ple m en tin g a co m pe tit ive fo od po lic y. Ad dit ion all y, th e e ffe cti ve ne ss on BM I wa s a sse sse d b y t he le ve l o f g ov er nm en t p oli cy . Am on g t he 26 st ud ies in th e U nit ed St at es ev alu at ing go ve rn m en t p oli cie s, 15 (5 8% ) re po rte d f av ou ra ble eff ec ts on BM I— 6 of 7 ( 85 % ) f ed er al po lic ies , 3 of 9 (3 3% ) s ta te / re gio na l p oli cie s, an d 6 of 10 (6 0% ) l oc al po lic ies . 103ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Br ow n a nd Su m m er be ll, 20 09 (8 1) 1.1 , 3 .2 Th is sy ste m at ic re vie w of 38 sc ho ol- ba se d o be sit y pr ev en tio n s tu die s f ou nd ins uffi cie nt ev ide nc e o n th e e ffe cti ve ne ss of di et ar y int er ve nt ion s a lon e, an d inc on sis te nt ev ide nc e f or PA in te rv en tio ns al on e o r co m bin ed in te rv en tio ns , th ou gh th ey m ay he lp pr ev en t o be sit y a nd ov er we igh t i n t he lo ng te rm . Th e r ev iew in clu de d 3 8 stu die s w hic h w er e e ith er RC Ts or co nt ro lle d c lin ica l tri als pu bli sh ed be tw ee n 19 93 an d 2 00 7. Of th e 3 8 s tu die s i nc lud ed , 3 inc lud ed di et in te rv en tio ns on ly (n ut rit ion ed uc at ion , sch oo l b re ak fas t, ch an ge in sch oo l lu nc he s), 15 in clu de d PA in te rv en tio ns on ly, an d 20 in clu de d b ot h d iet an d P A int er ve nt ion s. Ch ild re n a nd ad ole sce nt s a ge d 4– 18 ye ar s l ar ge ly in pr im ar y sch oo ls (2 3 s tu die s) an d se co nd ar y s ch oo ls (1 2 s tu die s) bu t a lso in ki nd er ga rte n o r pr es ch oo l ( 2 s tu die s) an d h igh sch oo l ( 1 s tu dy ) w or ldw ide wi th a m ajo rit y o f s tu die s co nd uc te d i n t he U nit ed St at es an d E ur op e. On e o f 3 (3 3% ) d iet st ud ies , 5 of 15 (3 3% ) P A stu die s a nd 9 of 20 (4 5% ) c om bin ed di et an d PA st ud ies de m on str ate d s ign ific an t d iff ere nc es be tw ee n i nt er ve nt ion an d c on tro l fo r B MI . T he rev iew fo un d i ns uffi cie nt ev ide nc e t o d raw co nc lus ion s a bo ut th e e ffe cti ve ne ss of die ta ry int er ve nt ion s o r t o c om pa re th eir effi ca cy to PA in ter ve nt ion s. T he y s ug ge st th at die ta ry int er ve nt ion s ( su ch as pr ov idi ng br ea kfa st in sch oo ls) m ay he lp pr ev en t p up ils fro m be co m ing ov er we igh t in th e s ho rt ter m . A lth ou gh inc on sis ten t, t he re is so m e s ug ge sti on th at PA in ter ve nt ion s m ay he lp pu pil s m ain ta in a he alt hy w eig ht in th e s ho rt ter m , p ar tic ula rly am on g y ou ng er ch ild ren an d g irl s. In stu die s wi th co m bin ed in ter ve nt ion s, th e fi nd ing s a re als o i nc on sis ten t, b ut th e a ut ho rs su gg es t t ha t co m bin ed di et an d P A i nt er ve nt ion s m ay he lp to pr ev en t c hil dr en fro m be co m ing ov er we igh t in th e lon g t erm . Br ow n e t a l., 20 16 (7 3) 1.1 Th is sy ste m at ic re vie w of 17 sc ho ol- ba se d c hil dh oo d ob es ity pr ev en tio n s tu die s fo un d t ha t m ult ifa ce te d pr og ra m m es w er e m or e l ike ly to im pr ov e B M I o ut co m es . Th e r ev iew in clu de d 1 7 stu die s ( RC Ts an d n on - ra nd om ize d c on tro lle d tri als w ith no -in te rv en tio n co nt ro ls) pu bli sh ed be tw ee n 20 07 an d 2 01 4. M os t s tu die s u se d a c om bin at ion of en vir on m en ta l, e du ca tio na l, PA an d p ar en ta l in vo lve m en t str at eg ies , w hil e f ou r u se d on ly ed uc at ion al str at eg ies . Sc ho ol- ba se d s tu die s i n pr im ar y s ch oo l-a ge d c hil dr en in Eu ro pe , C an ad a, Ch ile , N ew Ze ala nd an d t he U nit ed St at es . Pr og ra m m es la sti ng be tw ee n 6 –1 2 m on th s t ha t inv olv e m ult ipl e e nv iro nm en ta l, e du ca tio na l a nd ph ys ica l s tra te gie s a pp ea r t o b e m os t l ike ly to re su lt in BM I o r B M I z -sc or e i m pr ov em en t. Six ou t of 14 st ud ies re po rte d i m pr ov em en ts in BM I fo r th e t ot al int er ve nt ion gr ou p, wh er ea s 9 of th e 1 4 stu die s r ep or te d i m pr ov em en ts in at le as t o ne su bg ro up (e .g. gi rls , o lde r c hil dr en ). 104 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Br ow n e t a l., 20 19 (5 7) 1.1 , 3 .2 Th is Co ch ra ne sy ste m at ic re vie w an d m et a- an aly sis of 15 3 o be sit y p re ve nt ion stu die s, of w hic h 1 13 we re sc ho ol- ba se d, fo un d sig nifi ca nt eff ec ts on BM I a nd BM I z -sc or e f ro m co m bin ed die t/P A a nd PA al on e i n ch ild re n 6 –1 2 y ea rs an d o f P A alo ne in ch ild re n 1 3– 18 ye ar s in th e s ch oo l s et tin g, bu t n o eff ec t o n c hil dr en 0– 5 y ea rs in th e p re sch oo l o r c hil d c ar e se tti ng . Th e r ev iew in clu de d 1 53 stu die s ( 10 8 c lus te r-R CT s a nd 45 RC Ts) pu bli sh ed be tw ee n 19 93 an d 2 01 6. St ud ies in clu de d a co m bin at ion of di et an d P A int er ve nt ion s ( 93 st ud ies ), PA al on e ( 39 st ud ies ) a nd die t a lon e ( 21 st ud ies ). T he int er ve nt ion s u se d v ar iou s ap pr oa ch es (e .g. ed uc at ion al, en vir on m en ta l, p oli cy , p ar en t inv olv em en t). Ch ild re n a ge d 0 –1 8 y ea rs in m ult ipl e s et tin gs (9 1 s tu die s in pr im ar y o r s ec on da ry sc ho ol an d 2 2 i n p re sch oo l) m os tly fro m H IC (1 39 st ud ies ) a s w ell as up pe r- (1 3 s tu die s) an d low er m idd le- inc om e c ou nt rie s (1 st ud y) . In th e m et a- an aly sis of st ud ies in ch ild re n 0 –5 ye ar s i n c hil dc ar e o r p re sch oo l, n o s ign ifi ca nt eff ec ts we re re po rte d f or co m bin ed di et an d P A o r fo r P A a lon e. In ch ild re n 6 –1 2 y ea rs in th e s ch oo l se tti ng , s ign ifi ca nt eff ec t w as fo un d o n B M I z-s co re fr om co m bin ed di et an d P A i nt er ve nt ion s (-0 .04 ; 9 5% CI: -0 .08 , - 0.0 1; 15 st ud ies w ith 23 25 2 p ar tic ipa nt s) an d o n B M I fr om PA al on e (-0 .10 kg /m 2; 95 % CI: -0 .14 , - 0.0 6; 12 st ud ies wi th 15 92 9 p ar tic ipa nt s), w ith no si gn ifi ca nt eff ec t f ro m di et al on e i nt er ve nt ion s. In ch ild re n 13 –1 8 y ea rs, si gn ifi ca nt eff ec t w as fo un d o n B M I z-s co re fr om PA al on e ( -0 .20 ; 9 5% CI: -0 .30 , - 0.1 0; 1 s tu dy w ith 10 0 p ar tic ipa nt s) an d o n B M I ( -1 .53 kg /m 2; 95 % CI: -2 .67 , - 0.3 9; 4 s tu die s w ith 72 0 pa rti cip an ts) , w ith no si gn ifi ca nt eff ec t f ro m di et alo ne or di et an d P A c om bin ed . T he y a lso po ole d re su lts fr om st ud ies th at co m pa re d d iff er en t int er ve nt ion s a ga ins t e ac h o th er (e .g. di et al on e vs co m bin ed di et /P A) an d a ll s uc h a na lys es di d no t fi nd di ffe re nc es in im pa ct. Ca i e t a l 2 01 4 ( 61 ) 1.1 , 3 .2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 28 st ud ies , 23 of w hic h i n t he sc ho ol se tti ng , s ho we d a m od er at e bu t s ign ifi ca nt im pr ov em en t on bl oo d p re ssu re fr om ch ild ob es ity pr ev en tio n pr og ra m m es , w ith co m bin ed die t a nd PA in te rv en tio ns be ing th e m os t e ffe cti ve . Th e r ev iew in clu de d 23 st ud ies as se ssi ng 28 in te rv en tio ns (R CT s, qu as i-e xp er im en ta l s tu die s an d n at ur al ex pe rim en ts) , pu bli sh ed be tw ee n 1 98 5 an d 2 01 2. Ob es ity pr ev en tio n int er ve nt ion s. Sc ho ol- ba se d i nt er ve nt ion s w er e of a du ra tio n o f m ini m um 6 m on th s a nd ut iliz ed va rio us ap pr oa ch es , la rg ely ed uc at ion al bu t a lso aff ec tin g t he en vir on m en t, fo od se rv ice , o r in vo lvi ng pa re nt s. T we nt y- th re e o f t he 28 in te rv en tio ns in vo lve d a s ch oo l s et tin g, 18 us ed a co m bin at ion of di et an d P A int er ve nt ion s, 8 a pp lie d a PA - on ly int er ve nt ion an d 2 us ed a d iet -o nl y i nt er ve nt ion . Ch ild re n ( n= 18 92 5) w ith m ea n b as eli ne ag e r an ge 6.0 –1 3.9 ye ar s, lar ge ly in th e sch oo l s et tin g i n t he U nit ed St at es , E ur op e, Au str ali a a nd Ca na da . Th e m et a- an aly sis fo un d m od er at e b ut sig nifi ca nt av er ag e r ed uc tio ns in th e w eig ht ed m ea n d iff er en ce (W M D) in bl oo d p re ssu re, in bo th SB P ( W M D: -1 .64 m m Hg ; 9 5% CI : - 2.5 6, -0 .71 ; P = 0.0 01 ; 1 9 s tu die s) an d D BP (W M D: -1 .44 m m Hg ; 9 5% CI : - 2.2 8, -0 .60 ; P = 0.0 01 ; 1 8 stu die s). In th e s ub gr ou p a na lys es , s ign ifi ca nt re su lts w er e s ee n f or th e c om bin ed di et an d P A int er ve nt ion s i n b ot h S BP (W M D: -2 .11 m m Hg ; 95 % CI: -3 .19 , - 1.0 3) an d D BP (W M D: -1 .51 m m Hg ; 9 5% CI: -2 .55 , - 0.4 7) bu t n ot fo r t he int er ve nt ion s w ith PA al on e. 105ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ca i e t a l 2 01 4 ( 62 ) 1.1 , 3 .2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 20 int er ve nt ion s, 19 of w hic h we re sc ho ol- ba se d, sh ow ed th at ch ild ho od ob es ity pr ev en tio n p ro gr am m es sig nifi ca nt ly im pr ov ed LD L-C an d H DL -C , w ith co m bin ed die t a nd PA in te rv en tio ns be ing th e m os t e ffe cti ve . Th e r ev iew in clu de d 1 7 stu die s ( RC Ts, qu as i- ex pe rim en ta l s tu die s a nd na tu ra l e xp er im en ts) of 20 int er ve nt ion s p ub lis he d be tw ee n 1 98 5 a nd 20 12 . Th e 1 7 s tu die s a sse sse d 20 in te rv en tio ns (s om e stu die s h ad m or e t ha n o ne int er ve nt ion gr ou p) . S ch oo l- ba se d i nt er ve nt ion ne ed ed to ha ve a du ra tio n o f m ini m um 6 m on th s a nd ut iliz ed va rio us ap pr oa ch es , la rg ely ed uc at ion al bu t a lso aff ec tin g th e e nv iro nm en t o r in vo lvi ng pa re nt s. Ni ne te en of th e 20 in te rv en tio ns in vo lve d a s ch oo l s et tin g, 10 us ed a co m bin at ion of di et an d P A int er ve nt ion s, 8 a pp lie d a PA - on ly int er ve nt ion an d 2 us ed a d iet -o nl y i nt er ve nt ion . Ch ild re n ( n= 13 13 6) w ith m ea n b as eli ne ag e r an ge 6.0 –1 3.9 ye ar s, lar ge ly in th e sch oo l s et tin g i n t he U nit ed St at es , E ur op e, Au str ali a a nd Ca na da . Th e m et a- an aly sis fo un d a si gn ifi ca nt de sir ab le eff ec t o n t he w eig ht ed m ea n d iff er en ce (W M D) of LD L-C (W M D: -6 .06 m g/ dl; 95 % CI : - 11 .09 , -1 .02 ; P = 0.0 18 ) a nd H DL -C (W M D: 1. 87 m g/ dl; 95 % CI : 0 .39 , 3 .34 ; P = 0.0 13 ) a nd no n- sig nifi ca nt re du cti on s i n t ot al ch ole ste ro l a nd tr igl yc er ide . In th e s ub gr ou p a na lys es of di ffe re nt in te rv en tio n ty pe s, sig nifi ca nt re su lts w er e f ou nd fo r L DL -C fro m di et an d P A ( W M D: -8 .49 m g/ dl; 95 % CI : -1 2.7 7, -4 .21 ), die t o nl y ( W M D: -1 7.5 9m g/ dl; 95 % CI : - 25 .58 , - 9.6 0) an d P A o nl y ( W M D: -2 .67 m g/ dl; 95 % CI : - 4.3 0, -1 .03 ) i nt er ve nt ion s a nd on to ta l c ho les te ro l fr om di et -o nl y i nt er ve nt ion s (W M D: -1 7.2 1 m g/ dl; 95 % CI : - 25 .76 , - 8.6 6) . Ca lve rt, D em ps ey an d P ov ey , 2 01 9 (1 31 ) 3.1 , 4 .1 Th is sy ste m at ic re vie w of 29 sc ho ol- ba se d h ea lth y ea tin g i nt er ve nt ion stu die s i n a do les ce nt s fo un d t he m os t e ffe cti ve int er ve nt ion co m po ne nt s t o be pe er in vo lve m en t, us e o f ed uc at ion al m ed ia, en ha nc ed av ail ab ilit y o f h ea lth y f oo ds , an d t ail or ed co m pu te r-b as ed fee db ac k. Th e r ev iew in clu de d 2 9 stu die s ( 9 R CT s, 7 q ua si- ex pe rim en ta l d es ign , 3 co ho rt stu die s) pu bli sh ed be tw ee n 1 98 7 a nd 20 17 . In te rv en tio ns in clu de d he alt hy ea tin g l es so ns (2 0 stu die s), he alt hy ea tin g ac tiv iti es su ch as pr ac tic al ac tiv iti es or ro le- pla yin g ( 13 stu die s), pr ob lem so lvi ng wo rk sh ee ts (1 6 s tu die s) an d/ or a pr ac tic al les so n ( 11 stu die s), pe er in vo lve m en t (9 st ud ies ) a nd pa re nt al inv olv em en t ( 10 st ud ies ). Ad ole sce nt s a ge d 1 1– 16 ye ar s in th e s ch oo l s et tin g, lar ge ly in th e U nit ed St at es an d E ur op e, as w ell as in Au str ali a, Ca na da , Ch ina , Is ra el an d T un isi a. Ov er all , 2 4 o f t he 29 st ud ies w er e s uc ce ssf ul in pr om ot ing di et ar y b eh av iou r c ha ng e. Th e au th or s c on clu de d t ha t t he in te rv en tio ns ap pe ar ed m or e e ffe cti ve in th is ag e g ro up w he n th ey in vo lve d p ee rs (9 of 9 stu die s w ith th is co m po ne nt w er e s uc ce ssf ul) , u se d e du ca tio na l m ed ia to de liv er he alt h m es sa ge s ( 7 o f 7 st ud ies wi th th is co m po ne nt w er e s uc ce ssf ul) , in cre as ed av ail ab ilit y o f h ea lth y f oo ds in sc ho ol (6 of 6 stu die s w ith th is co m po ne nt ), an d i nc or po ra te d co m pu te r-b as ed in div idu ali ze d f ee db ac k w ith no rm at ive in fo rm at ion on ea tin g b eh av iou rs (4 of 5 s tu die s w ith th is co m po ne nt w er e s uc ce ssf ul) . 106 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ca i e t a l 2 01 4 ( 62 ) 1.1 , 3 .2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 20 int er ve nt ion s, 19 of w hic h we re sc ho ol- ba se d, sh ow ed th at ch ild ho od ob es ity pr ev en tio n p ro gr am m es sig nifi ca nt ly im pr ov ed LD L-C an d H DL -C , w ith co m bin ed die t a nd PA in te rv en tio ns be ing th e m os t e ffe cti ve . Th e r ev iew in clu de d 1 7 stu die s ( RC Ts, qu as i- ex pe rim en ta l s tu die s a nd na tu ra l e xp er im en ts) of 20 int er ve nt ion s p ub lis he d be tw ee n 1 98 5 a nd 20 12 . Th e 1 7 s tu die s a sse sse d 20 in te rv en tio ns (s om e stu die s h ad m or e t ha n o ne int er ve nt ion gr ou p) . S ch oo l- ba se d i nt er ve nt ion ne ed ed to ha ve a du ra tio n o f m ini m um 6 m on th s a nd ut iliz ed va rio us ap pr oa ch es , la rg ely ed uc at ion al bu t a lso aff ec tin g th e e nv iro nm en t o r in vo lvi ng pa re nt s. Ni ne te en of th e 20 in te rv en tio ns in vo lve d a s ch oo l s et tin g, 10 us ed a co m bin at ion of di et an d P A int er ve nt ion s, 8 a pp lie d a PA - on ly int er ve nt ion an d 2 us ed a d iet -o nl y i nt er ve nt ion . Ch ild re n ( n= 13 13 6) w ith m ea n b as eli ne ag e r an ge 6.0 –1 3.9 ye ar s, lar ge ly in th e sch oo l s et tin g i n t he U nit ed St at es , E ur op e, Au str ali a a nd Ca na da . Th e m et a- an aly sis fo un d a si gn ifi ca nt de sir ab le eff ec t o n t he w eig ht ed m ea n d iff er en ce (W M D) of LD L-C (W M D: -6 .06 m g/ dl; 95 % CI : - 11 .09 , -1 .02 ; P = 0.0 18 ) a nd H DL -C (W M D: 1. 87 m g/ dl; 95 % CI : 0 .39 , 3 .34 ; P = 0.0 13 ) a nd no n- sig nifi ca nt re du cti on s i n t ot al ch ole ste ro l a nd tr igl yc er ide . In th e s ub gr ou p a na lys es of di ffe re nt in te rv en tio n ty pe s, sig nifi ca nt re su lts w er e f ou nd fo r L DL -C fro m di et an d P A ( W M D: -8 .49 m g/ dl; 95 % CI : -1 2.7 7, -4 .21 ), die t o nl y ( W M D: -1 7.5 9m g/ dl; 95 % CI : - 25 .58 , - 9.6 0) an d P A o nl y ( W M D: -2 .67 m g/ dl; 95 % CI : - 4.3 0, -1 .03 ) i nt er ve nt ion s a nd on to ta l c ho les te ro l fr om di et -o nl y i nt er ve nt ion s (W M D: -1 7.2 1 m g/ dl; 95 % CI : - 25 .76 , - 8.6 6) . Ca lve rt, D em ps ey an d P ov ey , 2 01 9 (1 31 ) 3.1 , 4 .1 Th is sy ste m at ic re vie w of 29 sc ho ol- ba se d h ea lth y ea tin g i nt er ve nt ion stu die s i n a do les ce nt s fo un d t he m os t e ffe cti ve int er ve nt ion co m po ne nt s t o be pe er in vo lve m en t, us e o f ed uc at ion al m ed ia, en ha nc ed av ail ab ilit y o f h ea lth y f oo ds , an d t ail or ed co m pu te r-b as ed fee db ac k. Th e r ev iew in clu de d 2 9 stu die s ( 9 R CT s, 7 q ua si- ex pe rim en ta l d es ign , 3 co ho rt stu die s) pu bli sh ed be tw ee n 1 98 7 a nd 20 17 . In te rv en tio ns in clu de d he alt hy ea tin g l es so ns (2 0 stu die s), he alt hy ea tin g ac tiv iti es su ch as pr ac tic al ac tiv iti es or ro le- pla yin g ( 13 stu die s), pr ob lem so lvi ng wo rk sh ee ts (1 6 s tu die s) an d/ or a pr ac tic al les so n ( 11 stu die s), pe er in vo lve m en t (9 st ud ies ) a nd pa re nt al inv olv em en t ( 10 st ud ies ). Ad ole sce nt s a ge d 1 1– 16 ye ar s in th e s ch oo l s et tin g, lar ge ly in th e U nit ed St at es an d E ur op e, as w ell as in Au str ali a, Ca na da , Ch ina , Is ra el an d T un isi a. Ov er all , 2 4 o f t he 29 st ud ies w er e s uc ce ssf ul in pr om ot ing di et ar y b eh av iou r c ha ng e. Th e au th or s c on clu de d t ha t t he in te rv en tio ns ap pe ar ed m or e e ffe cti ve in th is ag e g ro up w he n th ey in vo lve d p ee rs (9 of 9 stu die s w ith th is co m po ne nt w er e s uc ce ssf ul) , u se d e du ca tio na l m ed ia to de liv er he alt h m es sa ge s ( 7 o f 7 st ud ies wi th th is co m po ne nt w er e s uc ce ssf ul) , in cre as ed av ail ab ilit y o f h ea lth y f oo ds in sc ho ol (6 of 6 stu die s w ith th is co m po ne nt ), an d i nc or po ra te d co m pu te r-b as ed in div idu ali ze d f ee db ac k w ith no rm at ive in fo rm at ion on ea tin g b eh av iou rs (4 of 5 s tu die s w ith th is co m po ne nt w er e s uc ce ssf ul) . AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ch riq ui, Pi ck el an d St or y, 20 14 (4 1) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 24 st ud ies on st at e a nd dis tri ct co m pe tit ive fo od an d be ve ra ge la ws an d p oli cie s fo un d p os iti ve ou tco m es fo r po lic ies fo cu se d o n i n- sch oo l av ail ab ilit y a nd co ns um pt ion , bu t m ixe d r es ult s f or he alt h ou tco m es (B M I, o be sit y a nd ov er we igh t). Th is re vie w inc lud ed 24 na tu ra l e xp er im en ts of re al law s o r p oli cie s p ub lis he d be tw ee n 2 00 6 a nd 20 12 . St ud y d es ign s w er e c ro ss- se cti on al (2 0 s tu die s), lon git ud ina l ( 3 s tu die s) or a co m bin at ion (1 st ud y) . In te rv en tio ns st ud ied fo cu se d on co m pe tit ive fo od an d/ or be ve ra ge st at e l aw s ( 14 stu die s), di str ict po lic ies (8 stu die s) or bo th (2 st ud ies ) (e .g. nu tri tio n s ta nd ar ds fo r sch oo l m ea ls, sw itc hin g f ro m wh ole to 1% or no n- fat m ilk , ba nn ing SS Bs , p ro hib iti on s o n th e u se of fo od as a re wa rd ). Eig ht ee n s tu die s f oc us ed on fo od an d b ev er ag e p oli cie s, 4 f oc us ed on be ve ra ge -o nl y po lic ies , a nd 2 on fo od -o nl y po lic ies . Ch ild re n a nd ad ole sce nt s i n ele m en ta ry, m idd le an d h igh sch oo l in th e U nit ed St at es , in th e s ch oo l s et tin g. Fif te en of th e 2 4 f oo d a nd be ve ra ge la w or po lic y i nt er ve nt ion s i nfl ue nc ed ou tco m es in th e e xp ec te d d ire cti on . T he re m ain ing 9 stu die s re po rte d m ixe d o r n on sig nifi ca nt re su lts , b ut 3 of th es e r ep re se nt ed ea rly po lic ies fr om 20 02 an d 20 04 in on e s ta te on ly. O f 4 st ud ies m ea su rin g BM I o r o be sit y, 2 s tu die s h ad ex pe cte d r es ult s an d 2 st ud ies ha d m ixe d r es ult s. Of 11 st ud ies m ea su rin g c on su m pt ion , p ur ch as ing or di et ar y int ak e, 6 s tu die s h ad ex pe cte d r es ult s, 4 h ad m ixe d r es ult s a nd 1 ha d n on -si gn ifi ca nt re su lts . Of 13 st ud ies m ea su rin g a va ila bil ity an d a cce ss, 11 ha d e xp ec te d r es ult s a nd 2 ha d m ixe d r es ult s. Co lle y e t a l., 20 19 (7 6) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 9 s ch oo l fo od pr og ra m m es in Ca na da fo un d t ha t m ult ico m po ne nt int er ve nt ion s w er e p os iti ve ly as so cia te d w ith ch ild re n's de ve lop m en t o f n ut rit ion kn ow led ge , d iet ar y be ha vio ur ch an ge s a nd int ak e o f h ea lth y f oo ds . Th e r ev iew in clu de d s tu die s of 9 sch oo l fo od pr og ra m m es , pu bli sh ed in 11 pa pe rs be tw ee n 2 00 5 a nd 20 15 . Al l s tu die s i nc lud ed sc ho ol fo od pr ov isi on , u su all y i n co m bin at ion w ith m ult ipl e ot he r in te rv en tio ns su ch as po lic y, ed uc at ion , fa m ily an d co m m un ity in vo lve m en t, 8 o f wh ich w er e m ult ico m po ne nt co m bin ing sc ho ol fo od w ith he alt hy ea tin g p oli cie s ( 5 stu die s), ac tiv e o r p as siv e nu tri tio n e du ca tio n ( 8 stu die s), an d/ or nu tri tio n in th e s ch oo l c ur ric ulu m (5 stu die s). Ch ild re n i n p rim ar y s ch oo ls in va rio us re gio ns in Ca na da , inc lud ing in dig en ou s co m m un iti es ; p ro gr am m es of te n t ar ge te d d isa dv an ta ge d ch ild re n. Th e m ult ico m po ne nt sc ho ol fo od pr og ra m m e int er ve nt ion s w er e p os iti ve ly as so cia te d w ith ch ild re n's de ve lop m en t o f n ut rit ion kn ow led ge in 4 o ut of 6 stu die s w hic h m ea su re d i t, in die ta ry be ha vio ur an d f oo d p re fer en ce s i n a ll 7 st ud ies m ea su rin g t he se , a nd in in ta ke of he alt hy fo od s in 6 o ut of 8 stu die s w hic h a sse sse d i nt ak es . 107ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W De Bo ur de au dh uij et al ., 2 01 1 ( 68 ) 1.1 , 3 .1 Th is sy ste m at ic re vie w of sc ho ol- ba se d o be sit y pr ev en tio n i nt er ve nt ion s su gg es ts th at in te rv en tio ns fo cu se d o n e du ca tio n alo ne ha ve le ss of an eff ec t on ob es ity pr ev en tio n th an m ult ico m po ne nt pr og ra m m es . Th e r ev iew in clu de d 1 1 stu die s r ep or te d i n 2 7 a rti cle s of va rio us st ud y d es ign s (o bs er va tio na l, e xp er im en ta l, ex tra po lat ed an d e xp er ien ce - ba se d s ou rce s) pu bli sh ed be tw ee n 1 99 1 a nd 20 07 . In te rv en tio ns ai m ed at pr ev en tio n o f o be sit y a nd ob es ity -re lat ed di se as es , ta rg et ing di et ar y a nd PA be ha vio ur th ro ug h ed uc at ion by te ac he rs or he alt h p ro fes sio na ls alo ne in 5 s tu die s ( e.g . c las sro om les so ns , b oa rd ga m e) or a c om bin at ion of su ch ed uc at ion an d e nv iro nm en ta l m ea su re s i n 6 st ud ies (e .g. cla ssr oo m le sso ns , m od ify ing sch oo l m ea ls an d t uc k s ho ps , inc re as ing PE ho ur s o r a cti ve les so n t im e, pr ov idi ng pla yg ro un d e qu ipm en t a nd ac tiv ity , p ar en t o r c om m un ity m ee tin gs ) Pr im ar y s ch oo l c hil dr en ag ed 6– 12 ye ar s ( 6 s tu die s) an d ad ole sce nt s a ge d 1 3– 18 ye ar s (5 st ud ies ) i n E ur op e, in th e sch oo l-s et tin g. In pr im ar y s ch oo l c hil dr en , m ult ico m po ne nt int er ve nt ion s s ho we d m or e f av ou ra ble re su lts on va rio us ou tco m es (a nt hr op om et ry, kn ow led ge , be ha vio ur ) t ha n t ho se on ly us ing ed uc at ion . Ho we ve r, o nl y 1 of 4 m ult ico m po ne nt st ud ies sh ow ed an eff ec t o n a nt hr op om et ry. In ad ole sce nt s, 2 m ult ico m po ne nt pr og ra m m es ha d p os iti ve eff ec t o n a nt hr op om et ric ob es ity m ea su re s. T he au th or s s ug ge st th at in te rv en tio ns wi th an ed uc at ion co m po ne nt al on e h av e les s o f a n e ffe ct on ob es ity pr ev en tio n t ha n m ult ico m po ne nt pr og ra m m es co m bin g ed uc at ion w ith en vir on m en ta l in te rv en tio ns su ch as PA an d t he fo od en vir on m en t. de Sa an d L oc k, 20 08 (7 7) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 30 st ud ies of sc ho ol- ba se d int er ve nt ion s t o p ro m ot e F& V c on su m pt ion co nc lud ed th at su ch in te rv en tio ns ca n inc re as e b ot h i nt ak e a nd kn ow led ge , a s 7 0% of st ud ies inc re as ed F& V i nt ak e. Th e r ev iew in clu de d 3 0 stu die s r ep or te d i n 3 4 a rti cle s pu bli sh ed be tw ee n 1 99 8 a nd 20 07 . T he st ud ies in clu de d 1 8 RT Cs an d 1 2 n on -ra nd om ize d co nt ro lle d t ria ls, al l w ith a fo llo w- up ti m e b ey on d 3 m on th s. Sc ho ol- ba se d i nt er ve nt ion s en co ur ag ing F& V co ns um pt ion (f re e o r su bs idi se d F &V , c las sro om ac tiv iti es , s ch oo l-w ide int er ve nt ion s a nd sc ho ol po lic y, te ac he r/p ee r lea de r/p ar en t/c om m un ity inv olv em en t a nd /o r m od ifi ca tio n t o t he sc ho ol fo od se rv ice ); T we nt y- th re e s tu die s h ad m ult ipl e co m po ne nt s. Ch ild re n a nd ad ole sce nt s a ge d 5– 18 ye ar s, pa re nt s, an d o th er co m m un ity m em be rs in th e Un ite d S ta te s, Eu ro pe an d N ew Ze ala nd in th e s ch oo l s et tin g. Sc ho ol- ba se d i nt er ve nt ion s e nc ou ra gin g F& V c on su m pt ion w er e f ou nd to be eff ec tiv e at in cre as ing bo th in ta ke an d k no wl ed ge . Tw en ty -tw o o f t he 30 st ud ies (> 70 % ) i nc re as ed F& V i nt ak e ( fro m + 0.1 4 s er vin gs pe r d ay to + 0.9 9 s er vin gs pe r d ay ), wi th no ne de cre as ing int ak e. Fiv e o ut of 7 stu die s r ep or te d s ta tis tic all y sig nifi ca nt re su lts on nu tri tio n k no wl ed ge or pr efe re nc e, wh er ea s 1 of 7 stu die s h ad po sit ive im pa ct on F& V i nt ak e a nd w eig ht -re lat ed ou tco m es . T wo of th e n at ion al pr og ra m m es pr ov idi ng fr ee or su bs idi ze d F &V ha d b ee n ev alu at ed an d s ho we d s ign ifi ca nt in cre as e d ur ing th e s ch em e, wi th on e o f t he m al so sh ow ing su sta ine d i nc re as es af te r t he sc he m e fi nis he d. 108 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W De Bo ur de au dh uij et al ., 2 01 1 ( 68 ) 1.1 , 3 .1 Th is sy ste m at ic re vie w of sc ho ol- ba se d o be sit y pr ev en tio n i nt er ve nt ion s su gg es ts th at in te rv en tio ns fo cu se d o n e du ca tio n alo ne ha ve le ss of an eff ec t on ob es ity pr ev en tio n th an m ult ico m po ne nt pr og ra m m es . Th e r ev iew in clu de d 1 1 stu die s r ep or te d i n 2 7 a rti cle s of va rio us st ud y d es ign s (o bs er va tio na l, e xp er im en ta l, ex tra po lat ed an d e xp er ien ce - ba se d s ou rce s) pu bli sh ed be tw ee n 1 99 1 a nd 20 07 . In te rv en tio ns ai m ed at pr ev en tio n o f o be sit y a nd ob es ity -re lat ed di se as es , ta rg et ing di et ar y a nd PA be ha vio ur th ro ug h ed uc at ion by te ac he rs or he alt h p ro fes sio na ls alo ne in 5 s tu die s ( e.g . c las sro om les so ns , b oa rd ga m e) or a c om bin at ion of su ch ed uc at ion an d e nv iro nm en ta l m ea su re s i n 6 st ud ies (e .g. cla ssr oo m le sso ns , m od ify ing sch oo l m ea ls an d t uc k s ho ps , inc re as ing PE ho ur s o r a cti ve les so n t im e, pr ov idi ng pla yg ro un d e qu ipm en t a nd ac tiv ity , p ar en t o r c om m un ity m ee tin gs ) Pr im ar y s ch oo l c hil dr en ag ed 6– 12 ye ar s ( 6 s tu die s) an d ad ole sce nt s a ge d 1 3– 18 ye ar s (5 st ud ies ) i n E ur op e, in th e sch oo l-s et tin g. In pr im ar y s ch oo l c hil dr en , m ult ico m po ne nt int er ve nt ion s s ho we d m or e f av ou ra ble re su lts on va rio us ou tco m es (a nt hr op om et ry, kn ow led ge , be ha vio ur ) t ha n t ho se on ly us ing ed uc at ion . Ho we ve r, o nl y 1 of 4 m ult ico m po ne nt st ud ies sh ow ed an eff ec t o n a nt hr op om et ry. In ad ole sce nt s, 2 m ult ico m po ne nt pr og ra m m es ha d p os iti ve eff ec t o n a nt hr op om et ric ob es ity m ea su re s. T he au th or s s ug ge st th at in te rv en tio ns wi th an ed uc at ion co m po ne nt al on e h av e les s o f a n e ffe ct on ob es ity pr ev en tio n t ha n m ult ico m po ne nt pr og ra m m es co m bin g ed uc at ion w ith en vir on m en ta l in te rv en tio ns su ch as PA an d t he fo od en vir on m en t. de Sa an d L oc k, 20 08 (7 7) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 30 st ud ies of sc ho ol- ba se d int er ve nt ion s t o p ro m ot e F& V c on su m pt ion co nc lud ed th at su ch in te rv en tio ns ca n inc re as e b ot h i nt ak e a nd kn ow led ge , a s 7 0% of st ud ies inc re as ed F& V i nt ak e. Th e r ev iew in clu de d 3 0 stu die s r ep or te d i n 3 4 a rti cle s pu bli sh ed be tw ee n 1 99 8 a nd 20 07 . T he st ud ies in clu de d 1 8 RT Cs an d 1 2 n on -ra nd om ize d co nt ro lle d t ria ls, al l w ith a fo llo w- up ti m e b ey on d 3 m on th s. Sc ho ol- ba se d i nt er ve nt ion s en co ur ag ing F& V co ns um pt ion (f re e o r su bs idi se d F &V , c las sro om ac tiv iti es , s ch oo l-w ide int er ve nt ion s a nd sc ho ol po lic y, te ac he r/p ee r lea de r/p ar en t/c om m un ity inv olv em en t a nd /o r m od ifi ca tio n t o t he sc ho ol fo od se rv ice ); T we nt y- th re e s tu die s h ad m ult ipl e co m po ne nt s. Ch ild re n a nd ad ole sce nt s a ge d 5– 18 ye ar s, pa re nt s, an d o th er co m m un ity m em be rs in th e Un ite d S ta te s, Eu ro pe an d N ew Ze ala nd in th e s ch oo l s et tin g. Sc ho ol- ba se d i nt er ve nt ion s e nc ou ra gin g F& V c on su m pt ion w er e f ou nd to be eff ec tiv e at in cre as ing bo th in ta ke an d k no wl ed ge . Tw en ty -tw o o f t he 30 st ud ies (> 70 % ) i nc re as ed F& V i nt ak e ( fro m + 0.1 4 s er vin gs pe r d ay to + 0.9 9 s er vin gs pe r d ay ), wi th no ne de cre as ing int ak e. Fiv e o ut of 7 stu die s r ep or te d s ta tis tic all y sig nifi ca nt re su lts on nu tri tio n k no wl ed ge or pr efe re nc e, wh er ea s 1 of 7 stu die s h ad po sit ive im pa ct on F& V i nt ak e a nd w eig ht -re lat ed ou tco m es . T wo of th e n at ion al pr og ra m m es pr ov idi ng fr ee or su bs idi ze d F &V ha d b ee n ev alu at ed an d s ho we d s ign ifi ca nt in cre as e d ur ing th e s ch em e, wi th on e o f t he m al so sh ow ing su sta ine d i nc re as es af te r t he sc he m e fi nis he d. AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W De lga do -N og ue ra et al ., 2 01 1 ( 11 6) 3.1 , 4 .1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 19 st ud ies on pr im ar y sch oo l in te rv en tio ns to pr om ot e F &V co ns um pt ion co nc lud ed th at co m pu te r- ba se d i nt er ve nt ion s w er e eff ec tiv e i n i nc re as ing F& V c on su m pt ion wh er ea s m ult ico m po ne nt int er ve nt ion s a nd fr ee or su bs idi ze d F &V in te rv en tio ns we re no t e ffe cti ve . Th e r ev iew in clu de s 1 9 stu die s ( 13 RC Ts an d 6 co nt ro lle d c lin ica l t ria ls) pu bli sh ed be tw ee n 1 99 8 an d 2 00 9. Int er ve nt ion s t o p ro m ot e F& V c on su m pt ion th ro ug h m ult ico m po ne nt int er ve nt ion s: sch oo l ed uc ati on al/ cu rri cu lum co m po ne nt s, sch oo l en vir on m en t m od ific ati on , tea ch er or pa ren t in vo lve m en t (1 1 s tu die s), fre e/ su bs idi ze d F& V ( 3 s tu die s) an d b oa rd ga m es or co m pu ter -b as ed int er ve nt ion s ( 3 s tu die s) co m pa red ag ain st co nt ro l w ith no in ter ve nt ion . Tw o s tu die s inv olv ed ot he r c om pa ris on s (fr ee F& V d ist rib ut ion vs m ult ico m po ne nt pr og ram m e an d t ea ch er led vs nu tri tio nis t led in ter ve nt ion ). Ch ild re n a ge d 5 –1 2 y ea rs in pr im ar y s ch oo ls in th e U nit ed St at es , E ur op e a nd N ew Ze ala nd . Po ole d r es ult s o f 2 RC Ts of co m pu te r-b as ed int er ve nt ion s i n 6 06 pr im ar y s ch oo l c hil dr en sh ow ed eff ec tiv en es s i n i m pr ov ing co ns um pt ion of F& V ( SM D: 0. 33 ; 9 5% CI : 0 .16 , 0 .50 ; P= 0.0 00 1) . N o s ign ifi ca nt di ffe re nc es w er e f ou nd in po ole d a na lys is of 7 RC Ts of m ult ico m po ne nt int er ve nt ion s ( n= 48 00 ) o r p oo lin g r es ult s of 2 RC Ts ev alu at ing fr ee or su bs idi ze d F &V int er ve nt ion s ( n= 15 36 ). T he au th or s c on clu de th at co m pu te r-b as ed in te rv en tio ns co uld be co ns ide re d i n s ch oo ls, gi ve n t ha t t he y a re eff ec tiv e a nd ch ea pe r t ha n o th er al te rn at ive s. De -R eg il, Je ffe rd s an d P eñ a- Ro sa s, 20 17 (1 49 ) 4.1 Th is Co ch ra ne re vie w an d m et a- an aly sis of 13 st ud ies of po int -o f-u se fo rti fic at ion of fo od s w ith iro n- co nt ain ing M NP s, 8 o f w hic h w er e co nd uc te d i n s ch oo ls, fo un d re du ce d a na em ia an d i ro n de fic ien cy in pr es ch oo l- an d sch oo l-a ge ch ild re n. Th e r ev iew in clu de d 1 3 stu die s ( 5 R CT s, 6 r an do m ize d clu ste r t ria ls, 2 pr e- an d p os t- te st de sig n) . Iro n- co nt ain ing M NP s f or po int -o f-u se fo rti fic at ion of fo od s. Ch ild re n a ge d 2 –1 2 ( n= 58 10 ) in La tin Am er ica , A fri ca an d As ia. Ei gh t o f t he 13 st ud ies to ok pl ac e i n a sc ho ol se tti ng . Co m pa red to no in ter ve nt ion or pl ac eb o, ch ild ren rec eiv ing iro n- co nt ain ing M NP fo r p oin t-o f-u se for tifi ca tio n o f fo od s h ad lo we r r isk of an ae m ia (p rev ale nc e r ati o ( PR ): 0 .66 ; 9 5% CI : 0 .49 , 0 .88 ; 1 0 tri als ; 2 44 8 c hil dr en ; m od era te -q ua lit y e vid en ce ) an d i ro n d efi cie nc y ( PR : 0 .35 ; 9 5% CI : 0 .27 , 0 .47 ; 5 tri als ; 1 36 4 c hil dr en ; m od era te -q ua lit y e vid en ce ) as w ell as hi gh er ha em og lob in (M D: 3. 37 g/ L; 95 % CI : 0 .94 , 5 .80 ; 1 1 t ria ls; 27 46 ch ild ren ; low -q ua lit y e vid en ce ). O nly on e t ria l w ith 11 5 ch ild ren re po rte d o n a ll- ca us e m or ta lit y ( ze ro ca se s; low -q ua lit y e vid en ce ). T he re wa s n o e ffe ct on di arr ho ea (R R: 0.9 7; 95 % CI : 0 .53 , 1 .78 ; 2 tr ial s; 36 6 c hil dr en ; lo w- qu ali ty ev ide nc e) . T he au th or s co nc lud e t ha t p oin t-o f-u se fo rti fic ati on of fo od s wi th iro n- co nt ain ing M NP s r ed uc es an ae m ia an d i ro n d efi cie nc y i n p res ch oo l- an d s ch oo l-a ge ch ild ren , b ut in for m ati on on m or ta lit y, m or bid ity , de ve lop m en ta l o ut co m es an d a dv ers e e ffe cts is lim ite d. 109ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W do Am ar al e M elo et al ., 2 01 7 ( 12 9) 3.1 Th is sy ste m at ic re vie w of 11 nu tri tio n i nt er ve nt ion s us ing in fo rm at ion an d co m m un ica tio n t ec hn olo gy co uld no t d et er m ine th e m os t eff ec tiv e t yp e o f in te rv en tio n du e t o h et er og en eit y o f stu die s b ut su gg es ts th at lon g- te rm in te rv en tio ns wi th fr eq ue nt ex po su re to te ch no log ica l re so ur ce s a nd aim ed at a sin gle he alt h be ha vio ur ch an ge m ay im pr ov e n ut rit ion be ha vio ur s. Th e r ev iew in clu de d 1 1 stu die s ( 2 RC Ts, 5 clu ste r RC Ts, 4 qu as i-e xp er im en ta l) pu bli sh ed be tw ee n 2 00 7 an d 2 01 5. Nu tri tio na l in te rv en tio ns us ing in fo rm at ion an d co m m un ica tio n t ec hn olo gie s (co m pu te r g am es an d pr og ra m s, te xt m es sa ge s a nd int er ac tiv e C D- RO M s) Ad ole sce nt s a ge d 9 –1 7 y ea rs in th e s ch oo l s et tin g i n H IC. Fiv e o f 1 1 s tu die s h ad po sit ive eff ec ts on di ve rse die ta ry ou tco m es (e .g. in cre as ed in ta ke of F& V, br ow n b re ad an d/ or de cre as ed in ta ke of SS Bs , jun k f oo d) th at w er e s ta tis tic all y d iff er en t f ro m ba se lin e o r c om pa ris on gr ou p. Ta rg et ing he alt hy ea tin g a sso cia te d w ith PA se em s t o b e h av e a gr ea te r im pa ct in ad ole sce nt s' he alt h, es pe cia lly on w eig ht . H alf of th e s tu die s w hic h c om bin ed die t a nd PA pr es en te d p os iti ve ou tco m es in bo th va ria ble s. Al l in te rv en tio ns us ing ga m es we re eff ec tiv e; th is ca n b e a ttr ibu te d t o t he en te rta ini ng w ay in w hic h t he se in te rv en tio ns pr om ot e e du ca tiv e l ea rn ing . H et er og en eit y o f th e s tu die s m ak es it ha rd to st at e w ha t t yp e of in te rv en tio n i s m or e e ffe cti ve ; h ow ev er, th e au th or s s ug ge st th at lo ng -te rm in te rv en tio ns wi th fr eq ue nt ex po su re to te ch no log ica l re so ur ce s t ha t a lso ha ve a th eo re tic al co m po ne nt aim ed at a sin gle he alt h b eh av iou r c ha ng e c an po te nt ial ly im pr ov e n ut rit ion be ha vio ur s. 110 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Do ak et al . 2 00 6 (8 3) 1.1 , 2 .2, 3. 2 Th is sy ste m at ic re vie w of 25 lar ge ly sch oo l-b as ed ob es ity pr ev en tio n s tu die s f oc us ing on di et an d P A f ou nd th at a m ajo rit y o f s tu die s w er e eff ec tiv e i n i m pr ov ing ou tco m es re lat ed to he igh t, we igh t o r s kin fo lds ; h ow ev er, no n- eff ec tiv e i nt er ve nt ion s m or e o fte n i nc lud ed co m po ne nt s s uc h a s f am ily or pa re nt in vo lve m en t. Th e r ev iew in clu de d 2 5 stu die s o f v ar iou s d es ign s (R CT s, co nt ro lle d t ria ls, sm all -sc ale co nt ro lle d t ria ls, his to ric al co nt ro lle d t ria ls, qu as i-e xp er im en ta l d es ign s) pu bli sh ed be tw ee n 1 97 7 an d 2 00 3. On e s tu dy w as un pu bli sh ed . Ob es ity pr ev en tio n int er ve nt ion s t ar ge tin g d iet an d/ or PA (e .g. ed uc at ion , co un se llin g, cu rri cu lum , fo od se rv ice ), so m et im es su pp lem en te d b y a dd iti on al co m po ne nt s ( e.g . fa m ily or co m m un ity in vo lve m en t, ex tra cu rri cu lar ac tiv ity ). Ch ild re n a ge d 4 –1 6 y ea rs, lar ge ly in th e s ch oo l s et tin g, lar ge ly in th e U nit ed St at es an d Eu ro pe bu t a lso in Au str ali a an d R us sia . Th e a ut ho rs co m pa red fe atu res of 17 in ter ve nt ion s eff ec tiv e i n i m pr ov ing ou tco m es re lat ed to he igh t, we igh t a nd /o r s kin fol ds ve rsu s 8 no n- eff ec tiv e int er ve nt ion s a nd co m pa red th es e a ga ins t c rit eri a for ev alu ati ng ch ild ho od ob es ity in ter ve nt ion s, dr aw ing on ex ist ing m od els fo r c au sa l p ath wa ys to th e c on dit ion of ob es ity ris k i n c hil dr en . T he y f ou nd th at a s lig ht ly sm all er pr op or tio n o f t he eff ec tiv e int er ve nt ion s v s. th e n on -e ffe cti ve in ter ve nt ion s int er ve ne d o n b ot h d iet an d a cti vit y ( 82 % of stu die s v s. 88 % of st ud ies ), i nc lud ed ac tiv iti es ou tsi de sc ho ol as pa rt of th e i nt er ve nt ion (6 7% vs . 88 % ), t arg ete d t he ph ys ica l e nv iro nm en t ( 24 % vs . 2 5% ), r eq uir ed ac tiv e p ar tic ipa tio n o f c hil dr en (8 2% vs . 1 00 % ) a nd ac tiv ely in vo lve d p are nt s (4 7% vs . 5 7% ) o r t he br oa de r c om m un ity (6 % vs . 14 % ), a nd ha d a m ea n l on ge r d ur ati on (1 33 w ee ks vs . 6 1 w ee ks ). H ow ev er, m os t o f t he se di ffe ren ce s are sm all an d n on e o f t he se co m pa ris on s a re sta tis tic all y s ign ific an t. O n t he ot he r h an d, th e eff ec tiv e i nt er ve nt ion s m or e o fte n i nt er ve ne d o n tel ev isi on vi ew ing (2 0% vs . 0 ) a nd pu t in pl ac e a s str uc tu re th at co uld be co nt inu ed (e .g. in ter ve nt ion ac ce pt ed by sc ho ol co m m un ity , fi na nc ial ly su sta ina ble ) ( 94 % vs . 8 8% ). Dr ies se n e t a l., 20 14 (3 8) 1.1 , 4 .1 Th is sy ste m at ic re vie w of sch oo l fo od en vir on m en t po lic ies an d i nt er ve nt ion s fo un d t ha t m os t s tu die s re po rte d f av ou ra ble ou tco m es , p ar tic ula rly fo r co ns um pt ion or av ail ab ilit y o f he alt hy fo od s a nd be ve ra ge s, bu t r ep or te d l es s c on clu siv e re su lts on BM I. Th e r ev iew in clu de d 1 8 stu die s p ub lis he d b et we en 20 01 an d 2 01 3. M os t s tu die s we re na tu ra l e xp er im en ts, re po rti ng on eff ec ts of st at e or na tio na l p oli cy . Ch an ge s t o t he sc ho ol fo od en vir on m en t i nc lud ed ch an ge s i n f oo d o r b ev er ag e av ail ab ilit y i n c an te en s, sn ac k b ar s a nd ve nd ing m ac hin es . T he ch an ge s w er e im ple m en te d w ith or w ith ou t a p oli cy di re cti ve , a nd 11 of th e s tu die s w er e n at ur al ex pe rim en ts of th e e ffe ct of st at e o r n at ion al po lic y ch an ge s t ha t i m pa cte d t he sch oo l fo od en vir on m en t. Ch ild re n a ge d 5 –1 8 y ea rs in pr im ar y a nd se co nd ar y s ch oo l se tti ng in th e U nit ed Ki ng do m an d U nit ed St at es . Th e r ev iew co ns ide red ou tco m es re lat ed to bo dy we igh t, f oo d o r b ev era ge co ns um pt ion an d pu rch as ing . O ve ral l, 1 7 o f t he 18 st ud ies re po rte d a s tat ist ica lly si gn ific an t in cre as e i n h ea lth y e ati ng be ha vio ur s a nd /o r d ec rea se in BM I o f c hil dr en . O f th e 1 6 s tu die s m ea su rin g p ur ch as ing be ha vio ur or di et ar y o ut co m es , 1 5 r ep or ted a sig nifi ca nt int er ve nt ion eff ec t ( 8 o f 9 po lic y s tu die s, 7 o f 7 sci en tifi c s tu die s). Tw o o f 4 st ud ies (a ll o n p oli cy ) m ea su rin g B MI re po rte d s ign ific an t im pa ct. Th e au th or s c on clu de th at hig h- lev el po lic y c ha ng es im pa cti ng th e s ch oo l fo od en vir on m en t a re po ssi ble , h av e p ot en tia l a s o be sit y r ed uc tio n str ate gie s, an d c an re ac h m an y c hil dr en . 111ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Du dle y, Co tto n an d P er alt a, 20 15 (9 2) 2.2 , 3 .1, 4. 2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 49 st ud ies fo un d t ha t m os t o f t he te ac hin g s tra te gie s l ed to po sit ive ch an ge s i n h ea lth y ea tin g b eh av iou rs in pr im ar y sch oo l c hil dr en , e sp ec ial ly en ha nc ed cu rri cu la, cr os s- cu rri cu la an d e xp er ien tia l lea rn ing ap pr oa ch es . Th e r ev iew in clu de d 4 9 stu die s ( 1 R CT , 1 3 q ua si- ex pe rim en ta l s tu die s, an d 35 cl us te r-c on tro lle d t ria ls) pu bli sh ed be tw ee n 1 97 4 an d 2 01 1. Te ac hin g s tra te gy int er ve nt ion s i nc lud ed : 1) En ha nc ed cu rri cu lum ap pr oa ch es (i .e. sp ec ial ty nu tri tio n e du ca tio n pr og ra m m es be yo nd ex ist ing he alt h c ur ric ula ) ( 29 stu die s); 2) cr os s-c ur ric ula r ap pr oa ch es (i .e. nu tri tio n ed uc at ion pr og ra m m es de liv er ed ac ro ss tw o o r m or e tra dit ion al sch oo l s ub jec ts) (1 1 s tu die s); 3) pa re nt al inv olv em en t ( 10 st ud ies ); 4) ex pe rim en ta l le ar nin g ap pr oa ch es (e .g. sc ho ol ga rd en , c oo kin g d em os ) (1 0 s tu die s); 5) co nt ing en t re inf or ce m en t a pp ro ac he s (i. e. re wa rd s o r in ce nt ive s giv en to st ud en ts in re sp on se to de sir ed be ha vio ur s) (7 st ud ies ); 6) lit er ar y ab str ac tio n a pp ro ac he s (i. e. lit er at ur e r ea d b y/ to ch ild re n w he re by a ch ar ac te r pr om ot es /e xe m pli fie s po sit ive be ha vio ur s) (3 stu die s); 7) ga m es -b as ed ap pr oa ch es (2 st ud ies ); an d 8) w eb -b as ed ap pr oa ch es (2 stu die s). Ch ild re n a ge d 5 –1 2 y ea rs (n = 38 00 1) in pr im ar y s ch oo l se tti ng s, pr im ar ily in th e Un ite d S ta te s a nd Eu ro pe , b ut als o C an ad a, Au str ali a, Ne w Ze ala nd , a nd Tr ini da d a nd To ba go . M os t o f t he te ac hin g s tra te gie s l ed to po sit ive ch an ge s i n e at ing be ha vio ur s, bu t i n t he po oli ng of re su lts fr om st ud ies in vo lvi ng 20 23 4 ch ild re n, fe we r s tra te gie s h ad m ea n e ffe ct siz es M d> 0.4 , w hic h w as co ns ide re d a s t he cu t o ff fo r g oo d i nv es tm en ts in th is re vie w ac co rd ing to ‘H at tie ’s Z on e o f D es ire d E ffe cts ’ o bs er ve d i n ed uc at ion st ud ies . C ur ric ulu m -b as ed ap pr oa ch es we re m os t p op ula r, b ut on ly co ns ide re d g oo d inv es tm en ts in su bs eq ue nt m et a- an aly se s wh en im ple m en te d a lon gs ide ot he r t ea ch ing str at eg ies (M d= 0.1 2 f or fo od co ns um pt ion / en er gy in ta ke , M d= 0.4 5 f or F& V c on su m pt ion ). En ha nc ed cu rri cu lum ap pr oa ch es w er e e ffe cti ve at in cre as ing kn ow led ge (M d= 0.7 5) bu t n ot at re du cin g s ug ar in ta ke fr om be ve ra ge s w ith ad de d su ga r o r f ru it jui ce s ( M d= 0.2 8) . C ro ss- cu rri cu lar ap pr oa ch es w er e e ffe cti ve to im pr ov e F &V co ns um pt ion (M d= 0.6 3) as w ell as su ga r in ta ke fro m be ve ra ge s ( M d= 0.4 2) , w he re as co nt ing en t re inf or ce m en t a pp ro ac he s w er e e ffe cti ve to inc re as e F &V co ns um pt ion (M d= 1.3 4) bu t th is wa s b as ed on on ly 2 s tu die s. Ex pe rie nt ial lea rn ing st ra te gie s w er e a sso cia te d w ith th e lar ge st po ole d m ea n e ffe ct siz es fo r a ll 3 ou tco m e ar ea s ( M d= 1.3 1 f or fo od co ns um pt ion /e ne rg y int ak e, M d= 0.6 8 f or F& V c on su m pt ion , M d= 1.3 5 pr efe re nc e a nd in cre as ed nu tri tio na l k no wl ed ge ou tco m es ). Re du cin g s ug ar co ns um pt ion an d pr efe re nc e w er e m os t i nfl ue nc ed by cr os s- cu rri cu lar ap pr oa ch es em be dd ed in in te rv en tio ns . Pa re nt al inv olv em en t a lso ge ne ra lly le d t o go od re su lts in in div idu al stu die s ( 91 % po sit ive ou tco m es ), alt ho ug h t he po ole d m ea n e ffe ct siz e (M d= 0.3 9) w as ju st be low th e c ut -o ff. 112 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Eje m ot -N wa dia ro et al ., 2 01 5 ( 17 4) 4.7 Th is sy ste m at ic re vie w an d m et a- an aly sis of 22 st ud ies , 12 of w hic h w er e c on du cte d in sch oo ls or ch ild da y- ca re ce nt re s, fo un d t ha t h an d wa sh ing pr om ot ion m ay re du ce di ar rh oe a e pis od es in ch ild da y- ca re ce nt re s a nd sch oo ls by ab ou t 3 0% an d pe rh ap s b y s im ila r r at es in low er in co m e c ou nt rie s. Th e r ev iew in clu de d 2 2 R CT s pu bli sh ed be tw ee n 1 99 7 an d 2 01 3. Ha nd w as hin g p ro m ot ion (e du ca tio n a cti vit ies , w ith or wi th ou t p ro vis ion of so ap ). Tw elv e o f t he 22 st ud ies w er e co nd uc te d i n c hil d d ay -c ar e ce nt re s o r s ch oo ls in m ain ly HI C i nv olv ing ch ild re n a ge d un de r 3 ye ar s a nd up to 10 ye ar s ( n= 54 00 6) . O th er stu die s w er e c on du cte d in co m m un ity se tti ng s o r ho sp ita ls in low - a nd m idd le- inc om e c ou nt rie s. Ha nd w as hin g p ro m ot ion at ch ild da y- ca re fac ilit ies or sc ho ols pr ev en ts ar ou nd ab ou t 3 0% of dia rrh oe a e pis od es in H IC (R R: 0. 70 ; 9 5% C: I 0 .58 , 0.8 5; nin e t ria ls; 46 64 pa rti cip an ts; hi gh qu ali ty ev ide nc e) . T he au th or s e sti m at ed th at ha nd wa sh ing m ay pr ev en t a si m ila r p ro po rti on in lo w- an d m idd le- inc om e c ou nt rie s b ut on ly tw o t ria ls fro m ur ba n E gy pt an d K en ya w er e i nc lud ed in th e re vie w (R R: 0. 66 ; 9 5% CI : 0 .43 , 0 .99 ; 2 tr ial s; 45 38 0 p ar tic ipa nt s; low qu ali ty ev ide nc e) . Ev an s e t a l., 20 12 (5 9) 1.1 , 4 .1 Th is sy ste m at ic re vie w an d m et a- an aly sis of int er ve nt ion s t o i nc re as e F& V i nt ak e i n c hil dr en ag ed 5– 12 ye ar s f ou nd th at sch oo l-b as ed in te rv en tio ns m od er at ely im pr ov e f ru it int ak e b ut ha ve m ini m al im pa ct on ve ge ta ble in ta ke . Th e r ev iew in clu de d 2 7 stu die s w hic h c on sis te d o f ra nd om ize d a nd no n- ra nd om ize d c on tro lle d t ria ls pu bli sh ed be tw ee n 1 99 6 an d 2 00 8. Sc ho ol- ba se d i nt er ve nt ion s to in cre as e F &V in ta ke , fo cu sin g s ole ly on F& V o r as pa rt of w ide r o bje cti ve to pr om ot e h ea lth y d iet s. Sp ec ifi c i nt er ve nt ion s inc lud ed nu tri tio n e du ca tio n (a s p ar t o f s ch oo l c ur ric ulu m or th ro ug h m es sa ge s), inv olv em en t o f s ch oo l co m m un ity or pa re nt s (p ro jec ts fo r p ar en ts- ch ild re n, ne ws let te rs, te ac he r-s tu de nt co m m un ica tio ns ), pr ov isi on of F& V o r h ea lth y f oo ds , fo od pr ep ar at ion an d/ or ta sti ng du rin g s ch oo l. T he m ajo rit y we re m ult ico m po ne nt pr og ra m m es ; w he re as sin gle -c om po ne nt pr og ra m m es w er e m ain ly fre e o r s ub sid ize d f ru it- dis tri bu tio n s ch em es . Ch ild re n a ge d 5 –1 2 y ea rs (n = 26 36 1) in pr im ar y s ch oo ls in th e U nit ed Ki ng do m , Ne th er lan ds , t he U nit ed St at es an d N ew Ze ala nd . M et a- an aly sis on 21 of th e 2 7 s tu die s f ou nd th at F& V i nt ak e i nc re as ed by 0. 25 po rti on s/d ay (9 5% CI : 0 .06 , 0 .43 ), m ain ly du e t o i nc re as es in fru it co ns um pt ion . S ep ar at e a na lys es fo r fru it co ns um pt ion sh ow ed an in cre as e o f 0 .24 po rti on s/d ay (9 5% CI : 0 .05 , 0 .43 ) a nd fo r ve ge ta ble s a no n- sig nifi ca nt in cre as e o f 0 .07 po rti on s/d ay (9 5% CI : - 0.0 3, 0.1 6) ; h ow ev er, he te ro ge ne ity w as m od er at e t o h igh . T he po ole d es tim at es fo r s ing le- co m po ne nt pr og ra m m es we re sm all er, al th ou gh he te ro ge ne ity w as hig he r t ha n f or al l s tu die s c om bin ed . T he au th or s co nc lud ed th at sc ho ol- ba se d i nt er ve nt ion s m od er at ely im pr ov e f ru it int ak e b ut ha ve m ini m al im pa ct on ve ge ta ble in ta ke . T he y a lso no te d t ha t m ult ico m po ne nt pr og ra m m es te nd ed to re su lt in lar ge r im pr ov em en ts wh ile si ng le co m po ne nt in te rv en tio ns , in clu din g f re e a nd su bs idi ze d F &V di str ibu tio n s ch em es , t en de d t o be le ss eff ec tiv e, alt ho ug h t he re w er e t oo fe w stu die s i nc lud ed to m ak e fi rm co nc lus ion s. 113ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Fe ng et al ., 2 01 7 (5 6) 1.1 , 3 .2, 5. 1, 5.2 Th is m et a- an aly sis of 76 stu die s o f s ch oo l-b as ed ob es ity in te rv en tio ns in Ch ina co nc lud es th at co m pr eh en siv e s ch oo l-b as ed int er ve nt ion s i nv olv ing PA an d h ea lth ed uc at ion ar e m or e e ffe cti ve th an th e P A on ly int er ve nt ion s. Th e r ev iew in clu de d 3 0 ra nd om ize d a nd 46 no n- ra nd om ize d c on tro lle d t ria ls pu bli sh ed be tw ee n 1 99 7 an d 2 01 5. In te rv en tio ns to pr ev en t (3 6 s tu die s) or tr ea t ( 40 stu die s) ob es ity (b ut no t t o tre at ob es ity co m pli ca tio ns su ch as ty pe 2 dia be te s, hy pe rte ns ion ). T he m os t co m m on in te rv en tio ns w er e PA an d h ea lth ed uc at ion , ot he r in te rv en tio ns in clu de d die ta ry im pr ov em en t (e .g. di re ctl y m od ify ing sch oo l lu nc he s), w eig ht m an ag em en t ( i.e . w eig ht m on ito rin g a nd gu ida nc e am on g o ve rw eig ht an d o be se ch ild re n) , s ch oo l p oli cie s, ps yc ho log ica l c ou ns ell ing , ph ys ica l in fra str uc tu re su pp or t ( e.g . u pg ra din g sp or t f ac ilit ies ). Fo rty -se ve n stu die s u se d c om pr eh en siv e int er ve nt ion s w ith m ult ipl e co m po ne nt s. Ch ild re n a ge d 6 –1 9 y ea rs (n = 72 62 0) in pr im ar y a nd se co nd ar y s ch oo ls in Ch ina . M or e t re at m en t s tu die s w er e e ffe cti ve in at lea st on e a nt hr op om et ric ou tco m e ( e.g . B M I, we igh t s ta tu s, wa ist or hi p c irc um fer en ce , s kin fo ld th ick ne ss, bo dy fa t p er ce nt ag e) be tw ee n th e i nt er ve nt ion an d c on tro l g ro up s ( P< 0.0 5) co m pa re d w ith pr ev en tio n s tu die s ( 85 .0% vs . 5 8.3 % ). T he co m pr eh en siv e i nt er ve nt ion s inv olv ing PA an d h ea lth ed uc at ion w er e m or e eff ec tiv e t ha n t he PA on ly int er ve nt ion s f or tre at m en t a nd pr ev en tio n s tu die s. Am on g t he pr ev en tio n s tu die s t ar ge tin g a ll c hil dr en , 1 9 o f 2 7 m ult ico m po ne nt in te rv en tio ns (7 0.4 % ) v s 2 ou t of 9 sin gle co m po ne nt s ( 22 .2% ) w er e e ffe cti ve . Am on g t he m ult ico m po ne nt pr ev en tio n s tu die s, th e m os t e ffe cti ve st ud ies w er e t ho se co m bin ing PA an d h ea lth ed uc at ion (1 5 o ut of 19 , 7 8.9 % ). Am on g t ho se co m bin ing ot he r in te rv en tio ns , a ll th e e ffe cti ve st ud ies (4 ou t o f 8 , 5 0% ) c om bin ed he alt h e du ca tio n w ith w eig ht m an ag em en t. Th e m et a- an aly se s s ho we d c om pr eh en siv e int er ve nt ion s i nv olv ing PA an d h ea lth ed uc at ion ha d a la rg er eff ec t o n t he ch an ge of BM I t ha n PA on ly int er ve nt ion s i n b ot h t re at m en t s tu die s [(S M D: -1 .80 ; 9 5% CI : - 2.1 5, -1 .44 ) v s. (S M D: -0 .91 ; 9 5% CI : - 1.1 5, -0 .67 )] an d p re ve nt ion stu die s [ (S M D: -0 .19 ; 9 5% CI : - 0.2 7, -0 .11 ) v s. (S M D: + 0.0 5; 95 % CI : - 0.0 4, + 0.1 5) ]. 114 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Fre ric hs et al ., 20 15 (1 24 ) 3.1 , 4 .1, 4. 2, 4.4 , 4.5 , 4 .6 Th is sy ste m at ic re vie w of 10 2 stu die s o f s ch oo l a rch ite ctu re an d d es ign fo un d e vid en ce th at a ra ng e o f p hy sic al fac to rs co nt rib ut e t o c re at ing su pp or tiv e, he alt hy -e at ing sch oo l e nv iro nm en ts. Th e r ev iew in clu de d 1 02 stu die s p ub lis he d b et we en 19 96 an d 2 01 4. Th irt y- nin e stu die s c on tri bu te d t o t he qu ali ta tiv e fi nd ing s, 59 co nt rib ut ed to qu an tit at ive fin din gs an d 4 us ed m ixe d m et ho ds an d c on tri bu te d to bo th . In te rv en tio ns th at ha d a p hy sic al en vir on m en t co m po ne nt co nt rib ut ing to int er ve nt ion im pa ct. Ch ild re n i n p rim ar y a nd se co nd ar y s ch oo ls glo ba lly , b ut m os t s tu die s w er e c on du cte d in th e U nit ed St at es . An aly sis of 31 st ud ies of fo od se rv ice in sc ho ols fou nd ne ga tiv e e ffe cts on fo od ch oic es fro m sh or t sch oo l m ea l ti m es , lo ng ca fet eri a s er vic e l ine s o r ea sy ac ce ss an d a pp ea l o f u nh ea lth y s na ck s a nd co m pe tit ive fo od s. An aly sis of 23 st ud ies of w ate r ac ce ss an d v en din g m ac hin es fo un d n eg ati ve eff ec ts fro m lim ite d a cce ss to an d p oo r q ua lit y of wa ter or fro m pr es en ce of le ss he alt hy ite m s in ve nd ing m ac hin es . S tu die s o n t he eff ec t o f diff ere nt fo od s i n v en din g m ac hin es ha d m ixe d res ult s; 3 o f 7 st ud ies ob se rve d a sso cia tio ns be tw ee n i nc rea se d a cce ss to le ss he alt hy ite m s a nd po or er die ta ry be ha vio ur s. Six of 9 stu die s o n w ate r ac ce ss fou nd im pr ov em en ts in wa ter in ta ke , b ut no t in te rm s o f re du ce d S SB co ns um pt ion . Tw elv e of 16 st ud ies fo un d t ha t in cre as ed ac ce ssi bil ity of he alt hy ite m s o r d ec rea se d a cce ssi bil ity of les s h ea lth y i tem s i m pr ov ed di et ar y b eh av iou rs m ea su red by sa le, se lf- rep or ted da ta or pl ate wa ste . A na lys is of 21 st ud ies of sc ho ol ga rd en on -si te foo d p ro du cti on fo un d t ha t t he se co uld ea sil y b e i nt eg rat ed in to cu rri cu la an d h av e be ne fit s b ey on d i m pr ov ed di et ar y b eh av iou r; ho we ve r, b arr ier s s uc h a s i nt en siv e t im e a nd res ou rce re qu ire m en ts we re ide nt ifie d. An aly sis of 14 st ud ies on sc ho ol tea ch ing ki tch en s f ou nd th at th es e w ere po pu lar am on g s ta ff a nd st ud en ts, an d co uld ea sil y b e i nt eg rat ed w ith ot he r a ca de m ic to pic s. Fiv e o f 7 st ud ies fo un d s ign ific an t in cre as es in th e c on su m pt ion of , fo r e xa m ple , v eg et ab les . An aly sis of 15 st ud ies of ed uc ati on al sig na ge , wa yfi nd ing an d m ark et ing fo un d t ha t t he se ap pr oa ch es ar e a cce pt ab le str ate gie s t o s tu de nt s, pa ren ts an d t ea ch ers fo r a dd res sin g h ea lth y e ati ng in sch oo ls, th ou gh re su lts w ere m ixe d a nd of ten no t s us ta ine d. An aly sis of st ud ies of th e d ini ng or kit ch en do m ain fo un d t ha t o ve rcr ow din g i n t he ca fet eri a c on tri bu ted to st ud en ts’ ne ga tiv e s oc ial ex pe rie nc es w he rea s l ac k o f a pp ro pr iat e k itc he n fac ilit ies an d e qu ipm en t h ind ere d s ch oo ls’ ab ilit y to pr ov ide m ea ls wi th hi gh nu tri tio na l q ua lit y a nd ap pe al. 115ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Fri ed ric h, Sc hu ch an d W ag ne r, 2 01 2 (5 3) 1.1 , 3 .1, 3. 2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 23 RC T stu die s f ou nd th at co m bin ed PA an d n ut rit ion ed uc at ion int er ve nt ion s i n t he sc ho ol se tti ng w er e m or e e ffe cti ve in im pr ov ing BM I t ha n e ith er PA or nu tri tio n e du ca tio n i n iso lat ion . Th e r ev iew in clu de d 2 3 R CT s pu bli sh ed be tw ee n 1 99 9 an d 2 01 0. Fiv e s tu die s i nc lud ed on ly PA in te rv en tio ns , 2 in clu de d on ly nu tri tio n e du ca tio n a nd 16 in clu de d b ot h P A a nd nu tri tio n e du ca tio n. Al l o f t he stu die s e nc ou ra ge d h ea lth y die ts th ro ug h p re se nt at ion s an d d ida cti c m at er ial s; 7 stu die s a lso ha d i nt er ve nt ion s wi th sc ho ol m ea ls an d ca fet er ias . Ch ild re n a nd ad ole sce nt s ag ed 7– 17 ye ar s i n t he sch oo l s et tin g ( n= 17 69 3) , pr im ar ily in th e U nit ed St at es an d E ur op e, bu t a lso Ch ina , Au str ali a, In dia an d I sra el. Re su lts of th e m et a- an aly sis sh ow ed th at th e s tu die s w ith co m bin ed PA an d n ut rit ion ed uc at ion re su lte d i n s ign ifi ca nt im pr ov em en ts in BM I ( SM D: -0 .37 ; 9 5% CI: -0 .63 , - 0.1 2; 16 stu die s; 99 97 pa rti cip an ts) , w he re as ne ith er th e 5 i so lat ed PA in te rv en tio ns no r t he 2 iso lat ed nu tri tio n e du ca tio n i nt er ve nt ion s r es ult ed in sig nifi ca nt im pr ov em en ts in BM I. Gi ru m an d W as ie, 20 18 (1 84 ) 5.3 Th is sy ste m at ic re vie w an d m et a- an aly sis of 8 sch oo l-b as ed de wo rm ing pr og ra m m es fo un d a n o ve ra ll im pr ov em en t in ha em og lob in af te r d ew or m ing , w ith th e g re at es t i m pa ct on an ae m ia ra te s i n s tu die s t ha t co m bin ed de wo rm ing w ith hy gie ne pr om ot ion an d/ or co -a dm ini str at ion of iro n a nd re tin ol. Th e r ev iew in clu de d 8 st ud ies (5 cr os s-s ec tio na l a nd 3 RC Ts) pu bli sh ed be tw ee n 1 99 8 an d 2 01 5. Sc ho ol ba se d d ew or m ing pr og ra m m es w he re ch ild re n re ce ive d a t l ea st on e o r m or e r ou nd of de wo rm ing six m on th s a pa rt an d w er e fo llo we d f or at le as t a ye ar. Ch ild re n ( n= 1 0 05 23 9) in pr es ch oo l, p rim ar y a nd se co nd ar y s ch oo l s et tin g i n In dia , N ige ria , t he U nit ed Re pu bli c o f T an za nia , T ha ila nd an d V iet N am . M et a- an aly sis of 5 stu die s f ou nd an ov er all im pr ov em en t i n h ae m og lob in af te r d ew or m ing (1 .62 g/ dl; 95 % CI: 1. 01 g/ dl, 2. 25 g/ dl) . Th e p re va len ce of an ae m ia wa s m ar ke dly ch an ge d i n i nd ivi du al stu die s t ha t c om bin ed de wo rm ing w ith hy gie ne pr om ot ion an d/ or co -a dm ini str at ion of iro n a nd re tin ol. 116 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Go rd on , D yn an an d S ieg el, 20 18 (1 36 ) 3.1 , 4 .1, 4. 2 Th is sy ste m at ic re vie w of 48 sc ho ol ca fet er ia stu die s fo un d t ha t i nt er ve nt ion s t ha t ch an ge th e e nv iro nm en t to fa vo ur fa st an d i nt uit ive th ink ing fo r s ele cti ng he alt hie r o pt ion s w er e b ot h m or e c om m on an d m or e eff ec tiv e t ha n i nt er ve nt ion s th at fa vo ur sl ow an d co gn iti ve ly de m an din g c ho ice th ro ug h e du ca tin g c hil dr en . Th e r ev iew in clu de d 4 8 stu die s ( 17 RC Ts an d 3 1 c ro ss- ov er in te rv en tio ns ) p ub lis he d be tw ee n 2 01 2 a nd 20 17 . Va rio us ca fet er ia int er ve nt ion s, wh ich we re gr ou pe d b y t he be ha vio ur al ec on om ics th eo ry of Ka hn em an int o “ sy ste m 1” (f as t a nd int uit ive th ink ing ), “ sy ste m 2” (s low an d c og nit ive ly de m an din g) or bo th . “ Sy ste m 1” in te rv en tio ns in clu de ch an ge s i n p ric e, co lou r gr ou pin g, aff ec tiv e c ue s ( i.e . em ot ico ns ), pla ce m en t o f F& V, ex pr es s l ine s, cre at ive na m es an d a ttr ac tiv e l ab els , ad ve rti sin g a nd ot he r en vir on m en ta l c ha ng es su ch as ex po su re, co nv en ien ce , inc en tiv es , la be llin g, m es sa gin g ( e.g . b an ne rs, an no un ce m en ts) , a nd re fer en ce s t o s oc ial no rm s. “S ys te m 2” in te rv en tio ns inc lud e c las sro om nu tri tio na l ed uc at ion , n ut rit ion al inf or m at ion po sti ng an d ed uc at ion al pr og ra m m es ou tsi de th e c las sro om . Ch ild re n a nd ad ole sce nt s i n ele m en ta ry, m idd le an d h igh sch oo l. G eo gr ap hic al loc at ion no t i nd ica te d. De fin ing su cce ss as a 30 % im pr ov em en t i n a de sir ed ou tco m e ( i.e . c ha ng e i n c on su m pt ion an d/ or se lec tio n o f f ru its , v eg et ab les , s ug ar y dr ink s, pla in m ilk an d w ho le gr ain s) or sta tis tic all y s ign ifi ca nt re du cti on in BM I, t he au th or s f ou nd th at 24 of 27 (8 9% ) o f “ sy ste m 1” stu die s w er e s uc ce ssf ul vs . 8 ou t o f 1 3 ( 62 % ) o f m ixe d s tu die s a nd 4 ou t o f 8 (5 0% ) o f “ sy ste m 2” in te rv en tio ns . “ Sy ste m 1” st ud ies w er e sig nifi ca nt ly m or e e ffe cti ve th an “s ys te m 2” stu die s ( P = 0. 03 3) , b ut no si gn ifi ca nt di ffe re nc e in eff ec tiv en es s w as ob se rv ed be tw ee n a ny of th e ot he r in te rv en tio n g ro up s w he n c om pa re d. Ni ne ou t o f 1 1 s tu die s u sin g i nc en tiv es w er e s uc ce ssf ul as w er e a ll 5 st ud ies us ing em ot ico ns , w he re as 1 ou t o f 3 st ud ies us ing m es sa gin g w as eff ec tiv e. Ho we ve r, t he au th or s n ot e t ha t a s t he la tte r w as a s ing le co m po ne nt st ud y, th is de m on str at es th at m es sa gin g a lon e m ay si gn ifi ca nt ly inc re as e co ns um pt ion of ve ge ta ble s. 117ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Go ri e t a l., 20 17 (4 9) 1.1 , 2 .2, 3 .2 Th is re vie w an d m et a- an aly sis of 72 st ud ies in ov er we igh t o r o be se pr im ar y sch oo l a ge ch ild re n, 47 of wh ich w er e s ch oo l-b as ed , fo un d l ar ge st eff ec t o n B M I z-s co re fr om co m bin ed sch oo l-b as ed di et an d P A int er ve nt ion s w ith a fam ily co m po ne nt . Th e r ev iew , u pd at ing a p re vio us re vie w an d m et a- an aly sis by W at er s et al . ( 20 11 ), inc lud ed 72 stu die s ( RC T a nd no n- ra nd om ize d c on tro lle d t ria ls) of in te rv en tio ns la sti ng ≥ 12 we ek s, pu bli sh ed be tw ee n 19 93 an d 2 01 6. In te rv en tio ns fo cu se d o n die t ( 8 s tu die s), PA (2 2 stu die s) or bo th (4 2 s tu die s), co m pa re d w ith st an da rd ca re. Di et ar y i nt er ve nt ion s u su all y co ns ist ed of pr es cri bin g a d ail y d iet fo llo wi ng re co gn ize d g uid eli ne s, m on ito re d t hr ou gh a de ta ile d nu tri tio n d iar y w ith fe ed ba ck pr ov ide d b y s pe cifi ca lly tra ine d o pe ra to rs. Th e ap pr oa ch es w er e t yp ica lly ed uc at ion al, bu t s om e a lso ch an ge d t he en vir on m en t an d w er e c om ple m en te d b y co un se llin g b y s ch oo l h ea lth nu rse s o r in vo lve d p ar en ts. Ov er we igh t a nd ob es e c hil dr en ag ed 2– 18 ye ar s, in sch oo l an d/ or fa m ily se tti ng (4 7 stu die s i nc lud ed a sch oo l co m po ne nt ). St ud ies w er e wo rld wi de , t ho ug h m os tly fro m H IC. Th e u pd at e a dd ed 35 ne w stu die s t o t he pr ev iou s r ev iew by W at er s e t a l. ( 20 11 ) o f 3 7 stu die s. Am on g 4 7 r es ult s c on du cte d i n s ch oo l or co m bin ed sc ho ol/ fam ily se tti ng , s ign ifi ca nt re su lts w er e f ou nd fo r c om bin ed di et /P A i n ch ild re n a ge d 6 –1 2 y ea rs (m ea n B M I z -sc or e: -0 .13 ; 9 5% CI : - 0.1 9, -0 .06 ; 1 2 s tu die s) or PA alo ne (m ea n B M I z -sc or e: -0 .11 ; 9 5% CI : - 0.1 6, -0 .06 ; 1 4 s tu die s) in th e s ch oo l s et tin g, th ou gh th e b es t r es ult s w er e a ch iev ed fo r c om bin ed di et / PA in ch ild re n a ge d 6 –1 2 y ea rs in a c om bin ed sch oo l a nd fa m ily se tti ng (m ea n B M I z -sc or e: -0 .15 ; 9 5% CI : - 0.2 1, -0 .03 ; 6 st ud ies ). No sig nifi ca nt re su lts w er e f ou nd fo r o th er su bg ro up an aly se s f or th is or ot he r a ge gr ou ps (2 –5 ye ar s, 13 –1 8 y ea rs) fo r a ny in te rv en tio n o r s et tin g co m bin at ion . Gu er ra et al ., 2 01 4 (6 6) 1.1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 38 st ud ies on sc ho ol- ba se d c om bin ed PA an d n ut rit ion al ed uc at ion int er ve nt ion s d id no t s ho w sig nifi ca nt re du cti on s i n B M I. Th is re vie w inc lud ed 38 RC Ts pu bli sh ed be tw ee n 1 98 8 an d 2 01 2. Co m bin ed nu tri tio n ed uc at ion an d P A int er ve nt ion s. Ch ild re n a ge d 6 –1 7 y ea rs (n = 28 87 0) in sc ho ols la rg ely in th e U nit ed St at es an d Eu ro pe , b ut al so in Au str ali a, Isr ae l, C hin a, In dia an d N ew Ze ala nd Th e m ain an aly sis sh ow ed an in sig nifi ca nt im pr ov em en t i n B M I b et we en in te rv en tio n an d c on tro l g ro up s ( SM D: -0 .03 ; 9 5% CI : - 0.0 9, 0.0 4) fo r c om bin ed PA an d n ut rit ion al ed uc at ion int er ve nt ion s. T he re se ar ch , w hic h w as do ne as pa rt of a lar ge r p ro jec t a na lys ing eff ec tiv en es s of PA al on e ( Gu er ra et al ., 2 01 3) or nu tri tio n ed uc at ion al on e ( Sil ve ira et al ., 2 01 3) , s ho we d th at on ly nu tri tio n e du ca tio n a lon e w as eff ec tiv e; ho we ve r, t his ha d b ee n d riv en by on e s tu dy . 118 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ha m el an d Ro bb ins , 2 01 3 (1 28 ) 3.1 Th is sy ste m at ic re vie w of 15 st ud ies co nc lud es th at co m pu te r-b as ed an d we b- ba se d i nt er ve nt ion s c an im pr ov e e at ing be ha vio ur s an d d iet -re lat ed ph ys ica l ou tco m es am on g c hil dr en an d a do les ce nt s. Th e r ev iew in clu de d 1 5 RC Ts or qu as i-e xp er im en ta l stu die s. 5 c om pu te r-b as ed an d 1 0 we b- ba se d i nt er ve nt ion s to pr om ot e h ea lth y e at ing am on g c hil dr en an d ad ole sce nt s. Ch ild re n a nd ad ole sce nt s ag ed 6– 18 ye ar s, m os tly in th e s ch oo l s et tin g ( 10 of 15 stu die s) an d l ar ge ly in th e Un ite d S ta te s b ut al so in Eu ro pe (B elg ium , F ra nc e, th e Ne th er lan ds ). Al th ou gh 11 of th e 1 5 i nt er ve nt ion s r es ult ed in sta tis tic all y s ign ifi ca nt po sit ive ch an ge s in ea tin g b eh av iou r o r d iet -re lat ed ph ys ica l ou tco m es (e .g. BM I, w eig ht , % bo dy fa t a nd wa ist ci rcu m fer en ce ), stu die s t ha t i nc lud ed po st- int er ve nt ion fo llo w- up , r an gin g f ro m 3– 18 m on th s, sh ow ed th at ch an ge s w er e n ot m ain ta ine d. Co nd uc tin g t he in te rv en tio n a t sch oo l o r u sin g i nd ivi du all y t ail or ed fe ed ba ck en ha nc ed su cce ss. Th e a ut ho rs su gg es t t ha t t he clo se ly su pe rv ise d e nv iro nm en t o f a cl as sro om , alo ng w ith th e p ot en tia l fo r s oc ial su pp or t f ro m te ac he rs or fe llo w cla ssm at es , m ay en ha nc e th e e ffe ct of a sch oo l-b as ed in te rv en tio n o ve r a ho m e- ba se d a pp ro ac h t ha t m ay be co m ple te d alo ne . He rsc h e t a l., 20 14 (1 41 ) 3.3 Th is sy ste m at ic re vie w of 8 stu die s o n ( lar ge ly sch oo ls- ba se d) co ok ing cl as se s f or ch ild re n s ho we d b en efi cia l re su lts in fo od pr ep ar at ion sk ills , F &V co ns um pt ion , F& V p re fer en ce an d b loo d pr es su re, as w ell as BM I am on g o ve rw eig ht or ob es e ch ild re n a ge d 5 –1 2 y ea rs. Th e r ev iew in clu de d 8 RC T an d q ua si- ex pe rim en ta l stu die s p ub lis he d b et we en 20 03 an d 2 01 4. Co ok ing ed uc at ion pr og ra m m es fo r c hil dr en w ith fo od pr ep ar at ion le sso ns , a s pa rt of cu rri cu lum (6 st ud ies ), af te r s ch oo l p ro gr am m e ( 1 stu dy ) o r in th e e ve nin g ( 1 stu dy ). T hr ee of th e s tu die s en ga ge d p ar en ts, ei th er th ro ug h s ep ar at e l es so ns or a n ew sle tte r t ha t w as se nt ho m e. Ch ild re n a ge d 5 –1 2 y ea rs in th e p rim ar y s ch oo l s et tin g i n th e U nit ed St at es Sig nifi ca nt re su lts w er e o bs er ve d i n 2 of 2 stu die s m ea su rin g c hil dr en ’s w illi ng ne ss to tr y ne w fo od s, 4 o ut of 4 stu die s m ea su rin g f oo d pr ep ar at ion sk ills an d c oo kin g c on fid en ce , 5 ou t of 5 stu die s m ea su rin g c on su m pt ion of da ily F& V se rv ing s ( 4 s ho we d a si gn ifi ca nt in cre as e i n F &V an d 1 sh ow ed a sig nifi ca nt in cre as e i n f ru it on ly) , 2 o ut of 2 stu die s a na lys ing th e p re fer en ce fo r F& V, an d i n 1 st ud y m ea su rin g b loo d p re ssu re. No ns ign ifi ca nt ch an ge s i n B M I a m on g a ll c hil dr en we re ob se rv ed in 2 ou t o f 2 st ud ies m ea su rin g th is; ho we ve r, o ne of th em fo un d a si gn ifi ca nt de cre as e i n B M I a m on g o ve rw eig ht an d o be se pa rti cip an ts. 119ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Hi ng le et al ., 2 01 0 (9 4) 2.2 Th is sy ste m at ic re vie w of 24 st ud ies , 1 6 o f w hic h we re sc ho ol- ba se d, fo un d th at di re ct m et ho ds to en ga ge pa re nt s i n d iet ar y int er ve nt ion s, su ch as pa re nt at te nd an ce of ed uc at ion al or co un se llin g s es sio ns , we re m or e l ike ly to re su lt in po sit ive ch an ge s i n d iet ar y ou tco m es co m pa re d t o ind ire ct m et ho ds of pa re nt al inv olv em en t. Th e r ev iew in clu de d 2 4 R CT stu die s p ub lis he d b et we en 19 88 an d 2 00 8. Ob es ity pr ev en tio n, ch ro nic di se as e p re ve nt ion or he alt h p ro m ot ion int er ve nt ion s i nv olv ing a pa re nt co m po ne nt , e ith er dir ec tly (e .g. pr es en ce in nu tri tio n e du ca tio n s es sio ns , pa rti cip at ion in fa m ily be ha vio ur or co un se llin g o r pa re nt tr ain ing se ssi on s) or ind ire ctl y ( e.g . n ew sle tte r or pr ov isi on of in fo rm at ion th at di d n ot re qu ire re sp on se , inv ita tio ns to pa rti cip at e i n nu tri tio n a cti vit ies ). Ch ild re n a nd ad ole sce nt s ag ed 2– 18 ye ar s p rim ar ily in th e U nit ed St at es , b ut al so Au str ali a, Be lgi um , a nd Fr an ce , m os tly in th e s ch oo l s et tin g (1 6 s tu die s) bu t a lso in th e co m m un ity (8 st ud ies ). Al l 5 st ud ies us ing di re ct m et ho ds ac hie ve d a t lea st so m e p os iti ve di et ar y c ha ng e ( e.g . s at ur at e fat ty ac id, so diu m , e ne rg y a nd F& V i nt ak e) (2 po sit ive an d 3 m ixe d o ut co m es ), wh ile on ly 64 % of th e 1 9 s tu die s u sin g i nd ire ct m et ho ds ac hie ve d p os iti ve ch an ge s i n d iet ar y o r a dip os ity ou tco m es (7 po sit ive , 7 m ixe d a nd 5 no eff ec t). Am on g t ho se us ing in dir ec t m et ho ds , s tu die s re qu iri ng th at ch ild re n e ng ag e t he ir p ar en t i n an ac tiv ity w er e m os t l ike ly to re po rt po sit ive or m ixe d r es ult s. Lim ite d c on clu sio ns m ay be dr aw n r eg ar din g t he be st m et ho d t o i nv olv e pa re nt s i n c ha ng ing ch ild di et to pr ev en t o be sit y an d i m pr ov e h ea lth . In dir ec t m et ho ds re m ain th e m os t c om m on ly us ed st ra te gie s t o e ng ag e pa re nt s; ho we ve r, d ire ct m et ho ds of en ga ge m en t sh ow m or e p ro m ise an d t he re fo re w ar ra nt fu rth er re se ar ch . Ho we rto n e t a l., 20 07 (6 0) 1.1 , 3 .1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 7 stu die s fo un d t ha t s ch oo l-b as ed nu tri tio n i nt er ve nt ion s pr od uc ed a m od er at e inc re as e i n F &V co ns um pt ion am on g c hil dr en . Th e r ev iew in clu de d 7 stu die s ( 2 c ro ss- se cti on al qu as i-e xp er im en ta l s tu die s, 2 p re - a nd po st- te st de sig n, a r an do m ize d p air ed de sig n, a r an do m ize d fi eld tr ial an d a ra nd om ize d c om m un ity tr ial ) pu bli sh ed be tw ee n 1 99 8 an d 2 00 0. Sc ho ol- sp ec ifi c i nt er ve nt ion s to in cre as e F &V co ns um pt ion am on g c hil dr en . E ac h stu dy im ple m en te d a t lea st 2 s tra te gie s t o d eli ve r th e i nt er ve nt ion s, wi th ea ch st ud y i nc lud ing a cla ssr oo m -b as ed co m po ne nt . Ad dit ion all y, 6 o f t he st ud ies inc lud ed a fam ily co m po ne nt , 4 s tu die s i nc lud ed a fo od - se rv ice co m po ne nt , 2 st ud ies inc lud ed a co m m un ity co m po ne nt , a nd 1 stu dy inc lud ed a m ed ia ca m pa ign . Ch ild re n a ge d 7 –1 2 y ea rs (n = 81 56 ) a t p rim ar y s ch oo ls, m os tly in th e U nit ed St at es . Sc ho ol- ba se d n ut rit ion in te rv en tio ns ap pe ar to pr od uc e a m od er at e i nc re as e i n F &V co ns um pt ion am on g p rim ar y s ch oo l c hil dr en . In th e a na lys is of stu dy -le ve l e ffe ct siz es , t he co m bin ed es tim at e of th e n et di ffe re nc es in F& V i nt ak e w as 0. 38 se rv ing s ( 95 % CI : 0 .31 , 0 .44 ) a nd th e r ela tiv e ch an ge w as a 21 % (1 7% to 25 % ) i nc re as e i n F &V int ak e. F& V c on su m pt ion in cre as ed by ab ou t h alf of a se rv ing at th e i nd ivi du al lev el. In th e p oo led da ta , t he in te rv en tio n g ro up s c on su m ed 0. 45 m or e F &V se rv ing s t ha n t he co nt ro l g ro up s ( 95 % CI: 0. 33 , 0 .56 ). 120 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Hi ng le et al ., 2 01 0 (9 4) 2.2 Th is sy ste m at ic re vie w of 24 st ud ies , 1 6 o f w hic h we re sc ho ol- ba se d, fo un d th at di re ct m et ho ds to en ga ge pa re nt s i n d iet ar y int er ve nt ion s, su ch as pa re nt at te nd an ce of ed uc at ion al or co un se llin g s es sio ns , we re m or e l ike ly to re su lt in po sit ive ch an ge s i n d iet ar y ou tco m es co m pa re d t o ind ire ct m et ho ds of pa re nt al inv olv em en t. Th e r ev iew in clu de d 2 4 R CT stu die s p ub lis he d b et we en 19 88 an d 2 00 8. Ob es ity pr ev en tio n, ch ro nic di se as e p re ve nt ion or he alt h p ro m ot ion int er ve nt ion s i nv olv ing a pa re nt co m po ne nt , e ith er dir ec tly (e .g. pr es en ce in nu tri tio n e du ca tio n s es sio ns , pa rti cip at ion in fa m ily be ha vio ur or co un se llin g o r pa re nt tr ain ing se ssi on s) or ind ire ctl y ( e.g . n ew sle tte r or pr ov isi on of in fo rm at ion th at di d n ot re qu ire re sp on se , inv ita tio ns to pa rti cip at e i n nu tri tio n a cti vit ies ). Ch ild re n a nd ad ole sce nt s ag ed 2– 18 ye ar s p rim ar ily in th e U nit ed St at es , b ut al so Au str ali a, Be lgi um , a nd Fr an ce , m os tly in th e s ch oo l s et tin g (1 6 s tu die s) bu t a lso in th e co m m un ity (8 st ud ies ). Al l 5 st ud ies us ing di re ct m et ho ds ac hie ve d a t lea st so m e p os iti ve di et ar y c ha ng e ( e.g . s at ur at e fat ty ac id, so diu m , e ne rg y a nd F& V i nt ak e) (2 po sit ive an d 3 m ixe d o ut co m es ), wh ile on ly 64 % of th e 1 9 s tu die s u sin g i nd ire ct m et ho ds ac hie ve d p os iti ve ch an ge s i n d iet ar y o r a dip os ity ou tco m es (7 po sit ive , 7 m ixe d a nd 5 no eff ec t). Am on g t ho se us ing in dir ec t m et ho ds , s tu die s re qu iri ng th at ch ild re n e ng ag e t he ir p ar en t i n an ac tiv ity w er e m os t l ike ly to re po rt po sit ive or m ixe d r es ult s. Lim ite d c on clu sio ns m ay be dr aw n r eg ar din g t he be st m et ho d t o i nv olv e pa re nt s i n c ha ng ing ch ild di et to pr ev en t o be sit y an d i m pr ov e h ea lth . In dir ec t m et ho ds re m ain th e m os t c om m on ly us ed st ra te gie s t o e ng ag e pa re nt s; ho we ve r, d ire ct m et ho ds of en ga ge m en t sh ow m or e p ro m ise an d t he re fo re w ar ra nt fu rth er re se ar ch . Ho we rto n e t a l., 20 07 (6 0) 1.1 , 3 .1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 7 stu die s fo un d t ha t s ch oo l-b as ed nu tri tio n i nt er ve nt ion s pr od uc ed a m od er at e inc re as e i n F &V co ns um pt ion am on g c hil dr en . Th e r ev iew in clu de d 7 stu die s ( 2 c ro ss- se cti on al qu as i-e xp er im en ta l s tu die s, 2 p re - a nd po st- te st de sig n, a r an do m ize d p air ed de sig n, a r an do m ize d fi eld tr ial an d a ra nd om ize d c om m un ity tr ial ) pu bli sh ed be tw ee n 1 99 8 an d 2 00 0. Sc ho ol- sp ec ifi c i nt er ve nt ion s to in cre as e F &V co ns um pt ion am on g c hil dr en . E ac h stu dy im ple m en te d a t lea st 2 s tra te gie s t o d eli ve r th e i nt er ve nt ion s, wi th ea ch st ud y i nc lud ing a cla ssr oo m -b as ed co m po ne nt . Ad dit ion all y, 6 o f t he st ud ies inc lud ed a fam ily co m po ne nt , 4 s tu die s i nc lud ed a fo od - se rv ice co m po ne nt , 2 st ud ies inc lud ed a co m m un ity co m po ne nt , a nd 1 stu dy inc lud ed a m ed ia ca m pa ign . Ch ild re n a ge d 7 –1 2 y ea rs (n = 81 56 ) a t p rim ar y s ch oo ls, m os tly in th e U nit ed St at es . Sc ho ol- ba se d n ut rit ion in te rv en tio ns ap pe ar to pr od uc e a m od er at e i nc re as e i n F &V co ns um pt ion am on g p rim ar y s ch oo l c hil dr en . In th e a na lys is of stu dy -le ve l e ffe ct siz es , t he co m bin ed es tim at e of th e n et di ffe re nc es in F& V i nt ak e w as 0. 38 se rv ing s ( 95 % CI : 0 .31 , 0 .44 ) a nd th e r ela tiv e ch an ge w as a 21 % (1 7% to 25 % ) i nc re as e i n F &V int ak e. F& V c on su m pt ion in cre as ed by ab ou t h alf of a se rv ing at th e i nd ivi du al lev el. In th e p oo led da ta , t he in te rv en tio n g ro up s c on su m ed 0. 45 m or e F &V se rv ing s t ha n t he co nt ro l g ro up s ( 95 % CI: 0. 33 , 0 .56 ). AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Hu ng et al ., 2 01 5 (6 5) 1.1 Th is m et a- an aly sis of 27 sch oo l-b as ed ch ild ho od ob es ity pr ev en tio n pr og ra m m es di d n ot pr ov ide st ro ng ev ide nc e th at su ch pr og ra m m es ar e eff ec tiv e; m or eo ve r, s ing le co m po ne nt pr og ra m m es we re m or e e ffe cti ve th an m ult ico m po ne nt pr og ra m m es . Th e r ev iew in clu de d 2 7 sch oo l-b as ed ch ild ho od ob es ity pr ev en tio n pr og ra m m es of va rio us de sig ns (R CT s, CC Ts, int er ve nt ion w ith co nt ro l gr ou p, co ho rt) pu bli sh ed be tw ee n 1 98 8 a nd 20 13 . Sc ho ol- ba se d c hil dh oo d ob es ity pr ev en tio n pr og ra m m es (n ut rit ion , PA an d/ or pa re nt al inv olv em en t). Tw en ty -o ne of 27 st ud ies in clu de d a nu tri tio n c om po ne nt . Ch ild re n a nd ad ole sce nt s a ge d 6– 18 ye ar s ( n= 26 11 4) , in th e s ch oo l s et tin g m os tly in th e U nit ed St at es bu t a lso in Eu ro pe , A us tra lia an d C hil e. In a m et a- an aly sis of 54 SM D eff ec t s ize s b as ed on BM I o r s kin fo ld th ick ne ss ou tco m es , t he re wa s n ot st ro ng ev ide nc e t ha t s ch oo l-b as ed pr og ra m m es ha ve be en eff ec tiv e i n i m pr ov ing ch ild re n's BM I o r s kin fo ld th ick ne ss va lue s. Sig nifi ca nc e o f S M D wa s d efi ne d a s s m all (u p to 0. 20 ), m ed ium (u p t o 0 .50 ) o r l ar ge (0 .80 or gr ea te r). Th e w eig ht ed su m m ar y e ffe ct siz e f or all 54 eff ec t s ize s w as sm all (S M D: 0. 03 9; 95 % CI: -0 .01 3, 0.0 92 ; P = 0.1 45 ). A m et a- an aly sis of 18 eff ec t s ize s f ro m in te rv en tio ns th at co nt ain ed on ly on e p ro gr am m e c om po ne nt re su lte d i n l ar ge r y et sm all eff ec ts (S M D: 0. 11 2; 95 % CI : 0 .05 4, 0.1 7; P< 0.0 01 ). Co nv er se ly, a m et a- an aly sis of in te rv en tio ns w ith tw o o r m or e p ro gr am m e c om po ne nt s r es ult ed in a ve ry sm all eff ec t s ize of 0. 00 3 ( 95 % CI : - 0.0 64 , 0.0 70 , P = 0.9 35 ) w ith ex tre m e h et er og en eit y. Pr og ra m s t ha t c on ta ine d o nl y o ne m od er at or (P A o r n ut rit ion ) a nd an RC T d es ign pr od uc ed an eff ec t s ize of 0. 16 8 ( 95 % CI : 0 .08 5, 0.2 52 ) w ith no he te ro ge ne ity . 121ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ja im e a nd Lo ck , 20 09 (3 7) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 27 stu die s o f s ch oo l fo od an d nu tri tio n p oli cie s e xa m ini ng th e i m pa ct of nu tri tio n po lic y o n n ut rit ion ou tco m es fo un d t ha t s om e p oli cie s m ay im pr ov e t he fo od en vir on m en t a nd di et ar y int ak e b ut fo un d n o e vid en ce fo r im pa ct on BM I. Th e r ev iew in clu de d 2 7 ra nd om ize d a nd no n- ra nd om ize d c on tro lle d t ria ls, pu bli sh ed be tw ee n 1 99 1 an d 2 00 7. Sc ho ol fo od or nu tri tio n po lic ies (e .g. nu tri tio n gu ide lin es , re gu lat ion of fo od an d b ev er ag e a va ila bil ity , pr ice in te rv en tio n) . Ch ild re n a nd ad ole sce nt s a ge d 2– 18 ye ar s, fro m pr es ch oo ls (1 st ud y) , p rim ar y ( 7 s tu die s) an d s ec on da ry sc ho ols (1 0 stu die s) in th e U nit ed St at es an d E ur op e. M os t e vid en ce of eff ec tiv en es s w as fo un d f or th e i m pa ct of bo th nu tri tio n g uid eli ne s a nd pr ice in te rv en tio ns on in ta ke an d a va ila bil ity of fo od an d d rin ks , w ith le ss co nc lus ive re se ar ch on pr od uc t r eg ula tio n. So m e c ur re nt sc ho ol po lic ies ha ve be en eff ec tiv e i n i m pr ov ing th e fo od en vir on m en t a nd di et ar y i nt ak e i n s ch oo ls, bu t t he re is lit tle ev alu at ion of th eir im pa ct on BM I. F oo d a nd nu tri tio n p oli cie s h ad eff ec ts on m en u c om po sit ion . T hr ee ou t o f 4 st ud ies m ea su rin g f at co nt en t i n m en us fo un d t ha t t he int er ve nt ion sc ho ols ha d a si gn ifi ca nt de cre as e in to ta l a nd sa tu ra te d f at on th e s ch oo l m en us . Al l 4 st ud ies of in te rv en tio ns to in cre as e F &V off er ed sh ow ed th at gu ide lin es le d t o i nc re as ed fru it an d v eg et ab le av ail ab ilit y ( + 0.2 8 t o + 0.4 8 se rv ing s/d ay ). Al l t hr ee st ud ies m ea su rin g f at int ak e f ou nd si gn ifi ca nt de cre as es in to ta l fa t or sa tu ra te d f at (- 2.0 % to -1 0.9 % an d - 0.9 % to -5 .2% , re sp ec tiv ely ), an d b ot h s tu die s m ea su rin g fru it an d v eg et ab le int ak e s ho we d a po sit ive im pa ct (+ 0.3 0 t o + 0.3 7 s er vin gs /d ay ). Bo th stu die s t ha t m ea su re d t he im pa ct of re str ict ive nu tri tio n p oli cie s o n s ale s o f f oo d a nd be ve ra ge s su gg es te d a si gn ifi ca nt bu t l im ite d d ec re as e i n th e s ale s o f b an ne d f oo ds , s uc h a s c hip s a nd SS Bs , an d b ot h s tu die s t ha t m ea su re d t he im pa ct of re du cin g l ow -fa t f oo d p ric es sh ow ed si gn ifi ca nt inc re as es in lo w- fat sn ac ks an d F &V sa les Th e o ne stu dy th at ev alu at ed eff ec t o n B M I re po rte d n o diff er en ce be tw ee n g ro up s a t 1 ye ar of fo llo w- up . 122 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ja sp er, Le an d Ba rtr am , 2 01 2 (1 71 ) 4.6 , 4 .7 Th is sy ste m at ic re vie w of 47 w at er an d s an ita tio n int er ve nt ion s i n s ch oo ls fo un d t ha t w at er pr ov isi on an d a de qu at e s an ita tio n a nd wa te r f ac ilit ies ar e b en efi cia l to ch ild re n's he alt h. Th e r ev iew in clu de d 4 7 stu die s c om pr ise d o f 9 int er ve nt ion tr ial s, 4 R CT s, 1 o bs er va tio na l s tu dy , 1 pa rti cip at or y r es ea rch st ud y, 4 d es cri pt ive st ud ies , 2 6 cro ss- se cti on al stu die s a nd 2 ou tb re ak in ve sti ga tio ns . W at er an d s an ita tio n int er ve nt ion s: 11 ad dr es se d dr ink ing w at er, 7 ad dr es se d wa te r f or ha nd wa sh ing , 5 ad dr es se d p ro vid ing w at er fo r dr ink ing an d h an dw as hin g, 6 a dd re sse d s an ita tio n, 4 ad dr es se d s an ita tio n r ela te d to m en str ua tio n f ac ilit ies , an d 1 4 a dd re sse d p ro vid ing wa te r a nd sa nit at ion co m bin ed in sc ho ols . Ch ild re n i n s ch oo l s et tin g i n HI C a nd lo w- an d m idd le- inc om e c ou nt rie s, glo ba lly . Of re lev an ce to nu tri tio n, 20 st ud ies re po rte d inf ec tio us di se as e o ut co m es in clu din g d iar rh oe a. Th er e w as a re po rte d d ec re as e i n d iar rh oe al an d ga str oin te sti na l d ise as es w ith in cre as ed ac ce ss to ad eq ua te sa nit at ion an d h an dw as hin g f ac ilit ies an d s up pli es in sc ho ols . F ur th er m or e, 2 s tu die s re po rte d o ve rw eig ht re lat ed m ea su re s, on e o f wh ich re po rte d a 31 % re du cti on in th e r isk of ov er we igh t a sso cia te d w ith pr ov idi ng dr ink ing wa te r a nd ed uc at ion in sc ho ols . Je ns en et al ., 2 01 1 (1 58 ) 4.2 Th is sy ste m at ic re vie w of 30 sch oo l-b as ed st ud ies fo un d th at ec on om ic inc en tiv es m ay infl ue nc e s ch oo l c hil dr en ’s ch oic e o f f oo ds , s na ck s a nd be ve ra ge s Th e r ev iew in clu de d 3 0 stu die s o f b ot h e m pir ica l an d e xp er im en ta l d es ign inc lud ing RC Ts, cr os s se cti on al, si m ula tio n ex pe rim en ts, qu as i- ex pe rim en ta l, p ub lis he d be tw ee n 1 99 4 a nd 20 09 . Ec on om ic inc en tiv es fo r stu de nt s a nd th eir fa m ilie s us ing di re ct inc en tiv es (e .g. pr ice di ffe re nc es be tw ee n n ut rit iou s a nd no n- nu tri tio us fo od s, re wa rd s su ch as pe ns or ot he r s m all ite m s o r p ar tic ipa tio n i n a ra ffl e) an d/ or fo r s ch oo ls an d c om m un iti es us ing ind ire ct inc en tiv es (e .g. co m pe ns at ion fo r e ffo rts de vo te d t o p ro m ot ion of he alt hy nu tri tio n, re wa rd up on ac hie ve m en t o f s pe cifi c go als ). Sc ho ol- ag ed ch ild re n ( or so m et im es th eir pa re nt s) in pr im ar y o r s ec on da ry sc ho ols lar ge ly in th e U nit ed St at es as we ll a s i n E ur op e. In ge ne ra l, t he st ud ies su pp or te d t he hy po th es is th at bo th di re ct an d i nd ire ct ec on om ic inc en tiv es aff ec t f oo d a nd be ve ra ge ch oic es . Th e a ut ho rs co nc lud ed th at pr ice di ffe re nc es be tw ee n n ut rit iou s a nd no n- nu tri tio us fo od s we re eff ec tiv e f or al te rin g c on su m pt ion in sch oo l c afe te ria s o r f ro m ve nd ing m ac hin es an d fo r in cre as ing F& V c on su m pt ion at sc ho ols in co nt ex ts wh er e f oo ds or be ve ra ge s a re off er ed (fo r s ale ) a t t he sc ho ols an d s tu de nt s ( ty pic all y 10 –1 2 y ea rs or ol de r) br ing m on ey to sc ho ol to bu y s om e o f t he se ite m s. T he st ud ies on sm all re wa rd s a s i nc en tiv es w er e u su all y i m ple m en te d in co m bin at ion w ith ex te ns ive ed uc at ion al ac tiv iti es , t hu s t he au th or s w er e u na ble to dr aw an y c on clu sio ns on th e e ffe ct of th e i nc en tiv es . Co m pa re d t o t he se di re ct ec on om ic inc en tiv es , th e e vid en ce of in dir ec t i nc en tiv es w as so m ew ha t we ak er. Jo hn so n, W ee d an d T ou ge r- De ck er, 20 12 (7 1) 1.1 Th is sy ste m at ic re vie w of 7 sch oo l-b as ed m ult ico m po ne nt pr og ra m m es in m ino rit y ele m en ta ry sc ho ols in th e Un ite d S ta te s f ou nd so m e be ne fit s i n h ea lth be ha vio ur s an d a nt hr op om et ric m ea su re s. Th e r ev iew in clu de d 7 st ud ies (R CT s, no n- ra nd om ize d co nt ro lle d t ria ls, co ho rt stu die s) pu bli sh ed be tw ee n 19 96 an d 2 01 0. M ult ico m po ne nt sc ho ol- ba se d o ve rw eig ht an d ob es ity pr og ra m m es th at inc lud ed nu tri tio n e du ca tio n, op tim iza tio n o f t he sc ho ol fo od en vir on m en t a nd /o r P A. Ch ild re n i n e lem en ta ry sch oo ls wi th la rg e m ino rit y po pu lat ion s i n t he U nit ed St at es . Al l s tu die s r ep or te d s om e b en efi ts in he alt h be ha vio ur s a nd /o r a nt hr op om et ric m ea su re s, 5 o f wh ich re po rte d a po sit ive be ne fit on ov er we igh t or ad ipo sit y. Th e a ut ho rs no te d t ha t e ffe cti ve ne ss wa s g re at er w he n p ro gr am m e o bje cti ve s we re sp ec ifi c, im ple m en te d a cro ss th e s ch oo l en vir on m en ts, ex te nd ed in to th e c om m un ity an d we re cu ltu ra lly re lev an t. 123ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Jo m aa , M cD on ne ll an d P ro ba rt, 20 11 (1 56 ) 4.1 Th is re vie w fo un d p os iti ve eff ec ts of 13 sc ho ol fee din g stu die s o n e ne rg y i nt ak e, m icr on ut rie nt st at us , s ch oo l en ro llm en t a nd at te nd an ce of th e c hil dr en . Th e r ev iew in clu de d 1 3 stu die s ( RC Ts, in te rv en tio n co nt ro l s tu die s, cro sso ve r de sig n s tu die s, an d eff ec tiv en es s r ep or ts of ex ist ing sc ho ol fee din g pr og ra m m es ) p ub lis he d be tw ee n 1 99 6 a nd 20 10 . Sc ho ol fee din g p ro gr am m es (b re ak fas t, lun ch , s na ck s, pr ov isi on of fo rti fie d f oo ds ). Al l t yp es of sc ho ol fee din g we re in clu de d ( e.g . m ea ls ba se d o n s ta ple s, sn ac ks , fo rti fie d s na ck s a nd dr ink s). Pr im ar y s ch oo l a ge d c hil dr en in th e s ch oo l s et tin g i n de ve lop ing co un tri es . Th er e i s a re lat ive ly co ns ist en t p os iti ve eff ec t o f sch oo l fe ed ing pr og ra m m es on en er gy in ta ke an d m icr on ut rie nt st at us of sc ho ol- ag ed ch ild re n, an d a de cli ne in in fec tio ns an d m or bid iti es , pa rti cu lar ly fo r p ro gr am m es in w hic h m icr on ut rie nt fo rti fic at ion an d d ew or m ing ar e pr ov ide d t o p op ula tio ns w ith a hig h p re va len ce of m icr on ut rie nt de fic ien cie s a nd w or m in fec tio ns . M ixe d fi nd ing s w er e r ep or te d f ro m st ud ies in te rm s o f t he im pa ct of sc ho ol fee din g o n w eig ht , he igh t a nd BM I g ain s a m on g s ch oo l-a ge d ch ild re n. Lo ng -te rm st ud ies th at as se ss th e be ne fit s o f s ch oo l m ea ls, sn ac ks an d t ak e- ho m e ra tio ns on ch ild re n’s gr ow th an d h ou se ho ld fo od co ns um pt ion pa tte rn s a re st ill lac kin g. Jo ur da n e t a l., 20 16 (9 8) 2.2 , 3 .3 Th is sco pin g r ev iew of 11 stu die s o f m ult i-s et tin g N CD pr ev en tio n p ro gr am m es inv olv ing ch ild re n i n t he de sig n o r im ple m en ta tio n fo un d t ha t p os iti ve eff ec ts we re ob se rv ed at or ga niz at ion al lev el, in ch ild re n a nd in ad ult sta ke ho lde rs. Th e r ev iew in clu de d 1 1 stu die s p ub lis he d i n 1 2 pa pe rs be tw ee n 2 00 2 a nd 20 14 . D es ign s w er e m ain ly ca se st ud ies (s ing le or m ult ipl e) as w ell as on e R CT . Sc ho ol- an d c om m un ity - ba se d i nt er ve nt ion s wi th a ch ild in vo lve m en t co m po ne nt , w he re ch ild re n ar e c on su lte d a nd pr ov ide su gg es tio ns to ad ult sta ke ho lde rs re ga rd ing th e sch oo l p ro gr am m e o r w he re ch ild re n t ak e r es po ns ibi lit y fo r a nd in flu en ce de cis ion s. M os t s tu die s a lso in clu de d a c om m un ity /fa m ily co lla bo ra tio n c om po ne nt . Ch ild re n, ne ith er ag e n or ge og ra ph y s pe cifi ed . M os t int er ve nt ion s w er e d on e i n t he sch oo l s et tin g w ith co m m un ity inv olv em en t ( 8 i nt er ve nt ion s) an d a fe w we re m ain ly ba se d in th e c om m un ity w ith sc ho ol inv olv em en t ( 3 i nt er ve nt ion s). Po sit ive eff ec ts we re id en tifi ed at th e or ga niz at ion al lev el (e .g. po lic ies to en ha nc e F& V a va ila bil ity or st ru ctu ra l c ha ng es to cr ea te op po rtu nit ies fo r P A) , in ad ult st ak eh old er s ( e.g . he alt hie r e at ing ) a nd in ch ild re n ( e.g . h ea lth ier ea tin g, kn ow led ge , s elf -e ffi ca cy , le ad er sh ip) . Ho we ve r, t he nu m be r o f r ele va nt st ud ies w er e few an d h ad m et ho do log ica l c ha lle ng es . 124 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ka tz et al ., 2 00 8 (5 4) 1.1 , 2 .2, 3. 2 Th is sy ste m at ic re vie w an d m et a- an aly sis ex am ine d th e i nfl ue nc e o f s ch oo l- ba se d i nt er ve nt ion s o n bo dy w eig ht , fi nd ing th at co m bin at ion s o f n ut rit ion an d P A i nt er ve nt ion s a re eff ec tiv e a t a ch iev ing w eig ht re du cti on . Th e r ev iew in clu de d 1 9 stu die s ( 21 pa pe rs) , o f w hic h 14 w er e R CT s a nd 5 we re no n- ra nd om ize d c on tro lle d tri als , p ub lis he d b et we en 19 85 an d 2 00 4. Sc ho ol- ba se d i nt er ve nt ion s re lat ed to nu tri tio n, PA , re du cti on in te lev isi on vie wi ng or co m bin at ion s th er eo f, a im ed to pr ev en t un ne ce ssa ry w eig ht ga in or m an ag e w eig ht . In te rv en tio ns ut iliz ed m ult ipl e a pp ro ac he s inc lud ing in fo rm at ion se ssi on s, pr ac tic al ha nd s-o n sk ill bu ild ing , t ea ch er tra ini ng an d s ch oo l c afe te ria m od ifi ca tio ns . Ch ild re n a ge d 3 –1 8 y ea rs, m os tly in th e U nit ed St at es . M os t s tu die s t oo k p lac e i n ele m en ta ry sc ho ols , w ith a few in m idd le an d h igh sc ho ols an d 3 t ar ge te d h igh sc ho ols . Re su lts fr om th e m et a- an aly sis of 8 stu die s of co m bin ed nu tri tio n a nd PA in te rv en tio ns inv olv ing 10 75 2 c hil dr en sh ow ed si gn ifi ca nt re du cti on of bo dy w eig ht in ch ild re n ( SM D: -0 .29 ; 9 5% CI: -0 .45 , - 0.1 4; P= 0.0 00 2) . T he se int er ve nt ion s w er e e qu all y e ffe cti ve as th e s ing le nu tri tio n i nt er ve nt ion (S M D: -0 .39 ; 9 5% CI : -0 .56 , - 0.2 3) an d t he si ng le te lev isi on re du cti on int er ve nt ion (S M D: -0 .35 ; 9 5% CI : - 0.6 3, -0 .06 ), wh er ea s t he si ng le PA in te rv en tio n d id no t s ho w bo dy w eig ht re du cti on . T he au th or s c on clu de d th at a m ajo r c on tri bu tin g f ac to r t o t he su cce ss of co m bin ed nu tri tio n a nd PA in te rv en tio ns m ay ha ve be en th e n ut rit ion co m po ne nt . Po oli ng th e r es ult s o f t he 4 stu die s w ith a pa re nt /fa m ily co m po ne nt , t he co m bin ed di et an d P A i nt er ve nt ion im pr ov ed bo dy w eig ht in sch oo lch ild re n ( SM D: -0 .20 ; 9 5% CI -0 .41 , 0 .00 ; P= 0.0 5) . Kn ai et al ., 2 00 6 (7 9) 1.1 , 2 .2, 2. 3, 3.1 , 4.1 Th is sy ste m at ic re vie w of 15 st ud ies , 1 4 o f w hic h we re sc ho ol- ba se d, fo un d th at in te rv en tio ns ha d a s ign ifi ca nt eff ec t o f inc re as ing F& V i nt ak e i n ch ild re n a nd th e e vid en ce is str on ge st in fav ou r o f int er ve nt ion s u sin g m ult ipl e ap pr oa ch es . Th e r ev iew in clu de d 1 5 stu die s ( in 17 ar tic les ) o f wh ich 11 w er e R CT s a nd 4 w er e n on -ra nd om ize d co nt ro lle d t ria ls pu bli sh ed be tw ee n 1 99 7 a nd 20 04 . In te rv en tio ns to pr om ot e F &V co ns um pt ion (e .g. cl as sro om an d n on -c las sro om -b as ed ed uc at ion , c ha ng es to th e e nv iro nm en t, pa re nt inv olv em en t, tra ini ng fo r s ta ff) de liv er ed as sin gle or m ult i-a pp ro ac h int er ve nt ion s. Ch ild re n a nd ad ole sce nt s a ge d 5– 18 ye ar s, at sc ho ols in th e Un ite d S ta te s, Un ite d K ing do m an d I re lan d. Al l s tu die s e xc ep t on e w er e c on du cte d i n t he sch oo l s et tin g. Th e s tro ng es t e vid en ce to in cre as e F &V co ns um pt ion in ch ild re n w as fo un d f or m ult i- ap pr oa ch in te rv en tio ns . O f t he 15 st ud ies re vie we d, 10 ha d a si gn ifi ca nt po sit ive eff ec t on F& V c on su m pt ion , r an gin g f ro m + 0.3 to + 0.9 9 s er vin gs /d ay , p ar tic ula rly in pr im ar y sch oo ls. Si gn ifi ca nt im pr ov em en ts in F& V i nt ak e we re se en in 8 ou t o f 1 1 s tu die s w ith a pa re nt al inv olv em en t c om po ne nt an d 3 ou t o f 4 st ud ies wi th a co m m un ity in vo lve m en t c om po ne nt , 6 ou t o f 9 st ud ies w ith in cre as ed ex po su re to F& V in th e s ch oo l s et tin g ( e.g . c an te en s), 6 ou t o f 1 0 stu die s m od ify ing lu nc he s o r s na ck s p ro vid ed , 6 o ut of 9 stu die s i nt eg ra tin g F &V kn ow led ge or pr ep ar at ion in to th e c ur ric ulu m , 8 ou t o f 1 1 stu die s i nc lud ing a te ac he rs' tr ain ing co m po ne nt , 7 o ut of 9 inc lud ing a sch oo l fo od se rv ice st aff tra ini ng , a nd 1 ou t o f 2 st ud ies w ith a sch oo l nu tri tio n c ou nc il o r p oli cy . 125ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ko ng , L iu an d T ao , 20 16 (1 14 ) 3.1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 17 sc ho ol- ba se d n ut rit ion ed uc at ion int er ve nt ion s f or ob es ity in Ch ina fo un d s ign ifi ca nt im pa ct of in te rv en tio ns am on g s tu die s w ith hi gh er m et ho do log ica l q ua lit ies an d i n s tu die s w he re th e du ra tio n w as lo ng er th an 2 ye ar s. T he eff ec t o f n ut rit ion ed uc at ion on ob es ity w as sm all er th an th e e ffe cts of in te rv en tio ns ai m ed at pr ev en tin g m aln ut rit ion an d iro n- de fic ien cy an ae m ia. Th e r ev iew in clu de d 1 7 stu die s o n s ch oo l-b as ed nu tri tio n e du ca tio n, of wh ich 16 ha d i nt er ve nt ion an d c on tro l g ro up s w he re as on e u se d p re - a nd po st- te st de sig n w ith ou t a co nt ro l gr ou p. In ad dit ion , 9 stu die s w er e i nc lud ed in a co m pa ra tiv e m et a- an aly sis of nu tri tio n e du ca tio n t o ad dr es s u nd er nu tri tio n. Sc ho ol- ba se d n ut rit ion ed uc at ion in te rv en tio ns fo r ob es ity in clu din g n ut rit ion ed uc at ion , h ea lth ed uc at ion an d P A. Co m pr eh en siv e ap pr oa ch es w er e a im ed at fac ilit at ing di et ar y b eh av iou r ch an ge s a nd in cre as ing PA . Se ve ra l s tu die s e ng ag ed pa re nt s, te ac he rs or sc ho ol pe rso nn el in ad dit ion to th e ch ild re n. N ut rit ion ed uc at ion int er ve nt ion s a im ed at un de rn ut rit ion w er e i nc lud ed fo r t he co m pa ra tiv e a na lys is. Ch ild re n i n p rim ar y s ch oo ls in Ch ina . Ov er all , t he st ud ies sh ow ed no si gn ifi ca nt im pa ct of nu tri tio n e du ca tio n o n o be sit y ( OR : 0 .76 ; 95 % CI : 0 .55 , 1 .05 ; P = 0.0 9) . H ow ev er, w he n 3 stu die s w ith un ba lan ce d b as eli ne ch ar ac te ris tic s be tw ee n g ro up s a nd se lec tio n b ias in th e s tu dy su bje cts w er e e xc lud ed , t he im pa ct of nu tri tio n ed uc at ion on ob es ity in th e r em ain ing 14 stu die s i nv olv ing 17 27 7 c hil dr en w as si gn ifi ca nt (O R: 0. 73 ; 9 5% CI : 0 .55 , 0 .98 ; P = 0.0 3) . A n an aly sis st ra tifi ed ac co rd ing to th e d ur at ion of in te rv en tio n r ev ea led th at th e i nt er ve nt ion wa s e ffe cti ve on ly in th e 2 st ud ies in vo lvi ng 60 29 ch ild re n t ha t l as te d f or m or e t ha n 2 ye ar s (O R: 0. 49 ; 9 5% CI : 0 .42 , 0 .58 ; P < 0.0 00 01 ). T he au th or s c on clu de d t ha t s ch oo l-b as ed nu tri tio n ed uc at ion is a fea sib le an d c os t-e ffe cti ve st ra te gy th at he lps ch ild re n d ev elo p h ea lth y h ab its at ea rly ag e; ho we ve r, s uc h s tu die s u nd er ta ke n in Ch ina ha ve so m e i m po rta nt lim ita tio ns (i .e. ina de qu at e i nt er ve nt ion du ra tio n, in ap pr op ria te ra nd om iza tio n m et ho ds , s ele cti on bi as , un ba lan ce d b as eli ne ch ar ac te ris tic s b et we en co nt ro l a nd in te rv en tio n g ro up s, an d a bs en t sa m ple si ze ca lcu lat ion ) t ha t m igh t a ffe ct th e es tim at ed eff ec tiv en es s o f t he in te rv en tio n. By co m pa ris on , in te rv en tio ns ai m ed at pr ev en tin g ch ild ho od m aln ut rit ion ha d a po ole d O R o f 0.6 8 ( 95 % CI : 0 .64 , 0 .71 ; P < 0.0 00 1; 3 s tu die s) an d t ho se ai m ed at pr ev en tin g i ro n- de fic ien cy an ae m ia ha d a po ole d O R o f 0 .49 (9 5% CI : 0 .33 , 0.7 2; P< 0.0 01 ; 5 st ud ies ). 126 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Kr ish na sw am i e t al. , 2 01 2 ( 90 ) 2.2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 16 st ud ies fo un d a po sit ive re lat ion sh ip be tw ee n l ev els of co m m un ity en ga ge m en t a nd ou tco m e pe rfo rm an ce in sc ho ol- ba se d int er ve nt ion s r ela te d t o d iet , PA an d b od y w eig ht . Th e r ev iew in clu de d 16 st ud ies of va rio us de sig ns (R CT s, m at ch ed - co nt ro l e xp er im en ts an d ob se rv at ion al an d q ua si- ex pe rim en ta l) pu bli sh ed in pe er -re vie we d j ou rn als be tw ee n 2 00 0 a nd 20 11 . Sc ho ol- ba se d o be sit y pr ev en tio n i nt er ve nt ion s fo cu sin g o n d iet , P A a nd bo dy we igh t w ith di ffe re nt le ve l of pa rti cip at or y i nv olv em en t fro m co m m un iti es ac co rd ing to a co m pu te d " co m m un ity en ga ge m en t p er fo rm an ce " sco re ba se d o n c ap ac ity - bu ild ing co m po ne nt s a nd lev el/ de pt h o f p ar tn er inv olv em en t a cro ss fo ur re se ar ch st ag es (e .g. ta rg et ing lo ng -te rm im pa ct, re m ov ing en vir on m en ta l ba rri er s t o p ar tic ipa tio n or be ha vio ur, bu ild ing lea de rsh ip, pr om ot ing se lf- es te em an d e m po we rm en t, es ta bli sh ing ch an ne ls fo r fu nd ing or eq uip m en t, an d ad dr es sin g s ys te m -le ve l a nd po lic y c om po ne nt s). Sc ho ol- ag ed ch ild re n ( m ea n ag e 1 0.7 ye ar s), in vo lvi ng div er se gr ou ps (A fri ca n- Am er ica n, H isp an ic, N at ive Am er ica n, ru ra l, u rb an , low -in co m e) w ith at le as t 3 0% m ino rit y/ low -in co m e s ta tu s i n th e U nit ed St at es . Ob es ity pr ev en tio n i nt er ve nt ion s w ith gr ea te r sch oo l– co m m un ity pa rtn er sh ip ac hie ve d m or e we igh t-, di et -, an d P A- re lat ed ou tco m es in div er se sc ho ol- ag e p op ula tio ns co m pa re d t o int er ve nt ion s w ith le ss co m m un ity en ga ge m en t. Ov er all co m m un ity en ga ge m en t p er fo rm an ce bu t n ot sp ec ifi c c om m un ity en ga ge m en t ch ar ac te ris tic s ( e.g . re se ar ch st ag e, pa rtn er ty pe , o r n um be r o f in vo lve d p ar tn er s) wa s po sit ive ly co rre lat ed w ith ou tco m es (r = 0.6 6; 95 % CI : 0 .25 , 0 .88 ), su gg es tin g t ha t a gr ea te r de gr ee of co m m un ity en ga ge m en t i s l ink ed to gr ea te r a ch iev em en t o f in te rv en tio n o ut co m es . Th e 2 st ud ies w ith th e h igh es t c om m un ity en ga ge m en t p er fo rm an ce sc or es ac hie ve d t he hig he st ou tco m e p er fo rm an ce . In an an aly sis of th e s pe cifi c c om m un ity en ga ge m en t f ea tu re s, po sit ive co m m un ity en ga ge m en t-o ut co m e co rre lat ion s w er e p re se nt fo r in te rv en tio ns inv olv ing pa rtn er s i n n ee ds as se ssm en ts (r= 0.7 7; p= 0.0 4) bu t n ot fo r t ho se w ith ou t su ch in vo lve m en t. T he au th or s c on clu de d t ha t inv olv ing co m m un iti es in ne ed s a sse ssm en ts an d g oa l-s et tin g a pp ea re d t o p ro m ot e ou tco m es by in cre as ing co m m un ity su pp or t fo r t he in te rv en tio n, fa cil ita tin g s ub se qu en t im ple m en ta tio n a nd w ell -ta ilo re d d es ign . 127ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Kr ist jan sso n e t a l., 20 07 (1 48 ) 4.1 Th is Co ch ra ne re vie w an d m et a- an aly sis of 18 st ud ies on th e e ffe cti ve ne ss of sc ho ol fee din g p ro gr am m es fo un d th at sc ho ol m ea ls re su lte d in sm all im pr ov em en ts in we igh t, he igh t i n y ou ng er ch ild re n, at te nd an ce , m at h pe rfo rm an ce an d b eh av iou r, es pe cia lly in di sa dv an ta ge d ch ild re n. Th e r ev iew in clu de d 1 8 stu die s c om pr ise d o f 7 RC Ts, 9 CB A s tu die s, an d 2 i nt er ru pt ed ti m e s er ies . St ud ies w er e p ub lis he d be tw ee n 1 92 6 a nd 20 04 . Sc ho ol fee din g p ro gr am m es inc lud ing m ea ls (b re ak fas t o r lun ch ) o r s na ck s ( inc lud ing m ilk ), fo rti fie d o r n on - fo rti fie d, co m pa re d a ga ins t no m ea ls or sn ac ks pr ov ide d. Ch ild re n a ge d 5 –1 9 y ea rs at sc ho ols (m os tly pr im ar y sch oo ls) in H IC (9 st ud ies ) a nd LIC (9 st ud ies ). In LI C, ch ild re n w ho w er e f ed at sc ho ol ga ine d m or e w eig ht th an co nt ro ls ov er th e s tu dy du ra tio n [ 0.3 9 k g ( 95 % CI : 0 .11 , 0 .67 ; 1 46 2 pa rti cip an ts; 3 stu die s) ov er an av er ag e o f 1 9 m on th s i n t he RC T s tu die s a nd 0. 71 kg (9 5% CI : 0.4 8, 0.9 5) ov er 11 .3 m on th s i n t he CB A s tu die s]. Ch ild re n w ho w er e f ed at sc ho ol at te nd ed sc ho ol m or e f re qu en tly th an th os e i n c on tro l g ro up s. Fo r h eig ht , re su lts fr om LI C w er e m ixe d; in RC Ts, diff er en ce s i n g ain s w er e i m po rta nt on ly fo r yo un ge r c hil dr en (0 .40 cm ; 9 5% CI : 0 .03 , 0 .77 ; 12 26 pa rti cip an ts; 3 stu die s), bu t r es ult s f ro m th e C BA s w er e l ar ge an d s ign ifi ca nt (1 .43 cm ; 95 % CI : 0 .46 , 2 .41 ; 9 86 pa rti cip an ts; 6 stu die s). Re su lts fo r h eig ht fr om H IC we re m ixe d, bu t ge ne ra lly po sit ive ba se d o n o ne RC T a nd on e C BA stu dy . F or ed uc at ion al an d c og nit ive ou tco m es , ch ild re n w ho w er e f ed at sc ho ol ga ine d m or e th an co nt ro ls on m at h a ch iev em en t a nd on so m e s ho rt- te rm co gn iti ve ta sk s. Re su lts fr om HI C w er e m ixe d, bu t g en er all y p os iti ve . T he au th or s c on clu de th at sc ho ol m ea ls m ay ha ve sm all ph ys ica l a nd ps yc ho so cia l b en efi ts fo r dis ad va nt ag ed st ud en ts. Kr op sk i, K ec kle y an d J en se n, 20 08 (8 5) 1.1 ,3. 2 Th is sy ste m at ic re vie w of 14 sc ho ol- ba se d o be sit y pr ev en tio n p ro gr am m es of at le as t 6 m on th s d ur at ion fo un d t ha t s uc h i nt er ve nt ion s m ay be eff ec tiv e f or ce rta in po pu lat ion s. Th e r ev iew in clu de d ex pe rim en ta l o r q ua si- ex pe rim en ta l s tu dy de sig ns , inc lud ing RC Ts an d c oh or t co nc ur re nt ly co nt ro lle d t ria ls, pu bli sh ed be tw ee n 1 99 3 an d 2 00 5. Sc ho ol- ba se d o be sit y pr ev en tio n p ro gr am m es of a d ur at ion of 6 m on th s t o 6 ye ar s, fo cu sin g o n n ut rit ion (e .g. fo od se rv ice , n ut rit ion ed uc at ion ), PA (e .g. PE , fit ne ss) or bo th . Ch ild re n a ge d 4 –1 3 y ea rs at sch oo l in m ult ipl e c ou nt rie s, m os tly in th e U nit ed St at es bu t a lso in Ch ile , G er m an y, Th ail an d a nd th e U nit ed Kin gd om . Th e o ne nu tri tio n s tu dy re po rte d a re du cti on in ov er we igh t ( bu t s ign ifi ca nc e w as no t r ep or te d) . On e o ut of 2 PA st ud ies an d 5 of 11 co m bin ed stu die s r ep or te d s ign ifi ca nt eff ec ts in at le as t on e s ub gr ou p. Po oli ng of da ta fo r m et a- an aly sis wa s n ot po ssi ble du e t o i nc on sis te nt re po rti ng of sp ec ifi c o ut co m e m ea su re s a nd m ea su re s o f er ro r. T he au th or s c on clu de d t ha t s ch oo l-b as ed ob es ity pr ev en tio n p ro gr am m es m ay be eff ec tiv e fo r c er ta in po pu lat ion s. T he y n ot ed th at th e hig he st qu ali ty st ud y r ev iew ed fo un d a la rg e a nd sig nifi ca nt re du cti on in od ds ra tio fo r o ve rw eig ht in gir ls. 128 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Kr ist jan sso n e t a l., 20 07 (1 48 ) 4.1 Th is Co ch ra ne re vie w an d m et a- an aly sis of 18 st ud ies on th e e ffe cti ve ne ss of sc ho ol fee din g p ro gr am m es fo un d th at sc ho ol m ea ls re su lte d in sm all im pr ov em en ts in we igh t, he igh t i n y ou ng er ch ild re n, at te nd an ce , m at h pe rfo rm an ce an d b eh av iou r, es pe cia lly in di sa dv an ta ge d ch ild re n. Th e r ev iew in clu de d 1 8 stu die s c om pr ise d o f 7 RC Ts, 9 CB A s tu die s, an d 2 i nt er ru pt ed ti m e s er ies . St ud ies w er e p ub lis he d be tw ee n 1 92 6 a nd 20 04 . Sc ho ol fee din g p ro gr am m es inc lud ing m ea ls (b re ak fas t o r lun ch ) o r s na ck s ( inc lud ing m ilk ), fo rti fie d o r n on - fo rti fie d, co m pa re d a ga ins t no m ea ls or sn ac ks pr ov ide d. Ch ild re n a ge d 5 –1 9 y ea rs at sc ho ols (m os tly pr im ar y sch oo ls) in H IC (9 st ud ies ) a nd LIC (9 st ud ies ). In LI C, ch ild re n w ho w er e f ed at sc ho ol ga ine d m or e w eig ht th an co nt ro ls ov er th e s tu dy du ra tio n [ 0.3 9 k g ( 95 % CI : 0 .11 , 0 .67 ; 1 46 2 pa rti cip an ts; 3 stu die s) ov er an av er ag e o f 1 9 m on th s i n t he RC T s tu die s a nd 0. 71 kg (9 5% CI : 0.4 8, 0.9 5) ov er 11 .3 m on th s i n t he CB A s tu die s]. Ch ild re n w ho w er e f ed at sc ho ol at te nd ed sc ho ol m or e f re qu en tly th an th os e i n c on tro l g ro up s. Fo r h eig ht , re su lts fr om LI C w er e m ixe d; in RC Ts, diff er en ce s i n g ain s w er e i m po rta nt on ly fo r yo un ge r c hil dr en (0 .40 cm ; 9 5% CI : 0 .03 , 0 .77 ; 12 26 pa rti cip an ts; 3 stu die s), bu t r es ult s f ro m th e C BA s w er e l ar ge an d s ign ifi ca nt (1 .43 cm ; 95 % CI : 0 .46 , 2 .41 ; 9 86 pa rti cip an ts; 6 stu die s). Re su lts fo r h eig ht fr om H IC we re m ixe d, bu t ge ne ra lly po sit ive ba se d o n o ne RC T a nd on e C BA stu dy . F or ed uc at ion al an d c og nit ive ou tco m es , ch ild re n w ho w er e f ed at sc ho ol ga ine d m or e th an co nt ro ls on m at h a ch iev em en t a nd on so m e s ho rt- te rm co gn iti ve ta sk s. Re su lts fr om HI C w er e m ixe d, bu t g en er all y p os iti ve . T he au th or s c on clu de th at sc ho ol m ea ls m ay ha ve sm all ph ys ica l a nd ps yc ho so cia l b en efi ts fo r dis ad va nt ag ed st ud en ts. Kr op sk i, K ec kle y an d J en se n, 20 08 (8 5) 1.1 ,3. 2 Th is sy ste m at ic re vie w of 14 sc ho ol- ba se d o be sit y pr ev en tio n p ro gr am m es of at le as t 6 m on th s d ur at ion fo un d t ha t s uc h i nt er ve nt ion s m ay be eff ec tiv e f or ce rta in po pu lat ion s. Th e r ev iew in clu de d ex pe rim en ta l o r q ua si- ex pe rim en ta l s tu dy de sig ns , inc lud ing RC Ts an d c oh or t co nc ur re nt ly co nt ro lle d t ria ls, pu bli sh ed be tw ee n 1 99 3 an d 2 00 5. Sc ho ol- ba se d o be sit y pr ev en tio n p ro gr am m es of a d ur at ion of 6 m on th s t o 6 ye ar s, fo cu sin g o n n ut rit ion (e .g. fo od se rv ice , n ut rit ion ed uc at ion ), PA (e .g. PE , fit ne ss) or bo th . Ch ild re n a ge d 4 –1 3 y ea rs at sch oo l in m ult ipl e c ou nt rie s, m os tly in th e U nit ed St at es bu t a lso in Ch ile , G er m an y, Th ail an d a nd th e U nit ed Kin gd om . Th e o ne nu tri tio n s tu dy re po rte d a re du cti on in ov er we igh t ( bu t s ign ifi ca nc e w as no t r ep or te d) . On e o ut of 2 PA st ud ies an d 5 of 11 co m bin ed stu die s r ep or te d s ign ifi ca nt eff ec ts in at le as t on e s ub gr ou p. Po oli ng of da ta fo r m et a- an aly sis wa s n ot po ssi ble du e t o i nc on sis te nt re po rti ng of sp ec ifi c o ut co m e m ea su re s a nd m ea su re s o f er ro r. T he au th or s c on clu de d t ha t s ch oo l-b as ed ob es ity pr ev en tio n p ro gr am m es m ay be eff ec tiv e fo r c er ta in po pu lat ion s. T he y n ot ed th at th e hig he st qu ali ty st ud y r ev iew ed fo un d a la rg e a nd sig nifi ca nt re du cti on in od ds ra tio fo r o ve rw eig ht in gir ls. AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W La ng fo rd et al ., 20 14 (4 6) 1.1 , 2 .2, 3 .2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 67 st ud ies (o f w hic h 3 4 c on ce rn ed di et an d/ or PA ) p ro vid e e vid en ce fo r t he eff ec tiv en es s o f s om e int er ve nt ion s b as ed on th e W HO H PS fr am ew or k f or im pr ov ing ce rta in he alt h ou tco m es bu t n ot ot he rs. Th e r ev iew in clu de d 6 7 stu die s o f c lus te r R CT s t ha t ha d m ult ipl e s ch oo ls in ea ch gr ou p, wh er e c lus te rs we re at th e l ev el of sc ho ol, di str ict or ot he r g eo gr ap hic al ar ea . St ud ies w er e p ub lis he d be tw ee n 1 99 2 a nd 20 13 . In te rv en tio ns im pl em en te d th ro ug h t he th re e H PS fra m ew or k s tra te gi es : cu rri cu lu m , e nv iro nm en t an d f am ilie s/c om m un iti es (e .g. he alt h e du ca tio n pr om ot ed th ro ug h f or m al sc ho ol cu rri cu lu m , c ha ng es to th e s ch oo l's ph ys ica l an d/ or so cia l e nv iro nm en t, en ga ge m en t w ith fa m ilie s an d t he w id er co m m un ity ). Th e s pe cifi c i nt er ve nt ion s in clu de d d iet an d/ or PA in te rv en tio ns , a s w ell as ot he r h ea lth is su es . T he 34 stu di es fo cu sin g o n d iet an d/ or PA in clu de d d ive rse in te rv en tio n a pp ro ac he s (e .g. nu tri tio n e du ca tio n, F& V sc he m es , c ha ng es in sc ho ol ca fe te ria s, cu rri cu lar an d e xt ra cu rri cu lar PA ). Ch ild re n a ge d 5 –1 5 y ea rs in pr im ar y a nd se co nd ar y s ch oo ls wo rld wi de , la rg ely in H IC (th e U nit ed St at es , E ur op e, Au str ali a, Ne w Ze ala nd ), bu t als o i n U M IC (C hin a, M ex ico ), LM IC (In dia , E gy pt ) a nd a LIC (U nit ed Re pu bli c o f T an za nia ). Va rio us su bg ro up m et a- an aly se s b y i nt er ve nt ion ty pe w er e p er fo rm ed of th e 3 4 n ut rit ion an d/ or PA in te rv en tio ns us ing th e H PS ap pr oa ch . On ly PA in te rv en tio ns al on e ( 3 t ria ls) w er e ab le to re du ce BM I s ign ifi ca nt ly (M D: -0 .38 kg / m ²; 95 % CI : - 0.7 3, -0 .03 ; 1 43 0 p ar tic ipa nt s), wh er ea s 1 nu tri tio n o nl y s tu dy an d 9 co m bin ed nu tri tio n/ PA st ud ies di d n ot sh ow si gn ifi ca nt eff ec t o n B M I. T he re w as lit tle ev ide nc e o f an eff ec t o n B M I w he n a ge an d g en de r w er e ta ke n i nt o a cco un t u sin g B M I z -sc or e; on ly th e sin gle PA in te rv en tio n m ea su rin g t his ou tco m e sh ow ed a sig nifi ca nt eff ec t ( M D: -0 .47 ; 9 5% CI: -0 .69 , - 0.2 5; 19 6 p ar tic ipa nt s), w he re as th e re du cti on s s ee n f ro m 1 nu tri tio n o nl y s tu dy or fr om 7 co m bin ed nu tri tio n/ PA st ud ies w er e ins ign ifi ca nt . T he po ole d r es ult s f ro m 9 nu tri tio n stu die s d em on str at ed si gn ifi ca nt in cre as e i n F& V i nt ak e ( SM D: 0. 15 ; 9 5% CI : 0 .02 , 0 .29 ; 6 21 0 pa rti cip an ts) , b ut th e i nc re as e w as si gn ifi ca nt fo r t he 4 co m bin ed nu tri tio n/ PA st ud ies . T he re du cti on s i n f at in ta ke w he n p oo lin g r es ult s fro m 7 nu tri tio n o nl y s tu die s o r f ro m 10 co m bin ed nu tri tio n/ PA st ud ies w er e i ns ign ifi ca nt . Th e a ut ho rs co nc lud ed th at ho lis tic sc ho ol- ba se d int er ve nt ion s, lik e t ho se pr op os ed in th e H PS fra m ew or k, ca n b e e ffe cti ve at im pr ov ing a nu m be r o f h ea lth ou tco m es in st ud en ts. 129ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W La ng fo rd et al ., 20 15 (9 7) 2.2 , 2 .3 Th is sy ste m at ic re vie w of pr oc es se s o f im ple m en tin g 26 sc ho ol- ba se d d iet an d PA st ud ies al ign ed to th e W HO H PS fr am ew or k f ou nd th at w or kin g w ith te ac he rs, pr ov idi ng on go ing tr ain ing an d s up po rt an d t ail or ing of pr og ra m m es ac ce nt ua te d re su lts , w hil e f am ily inv olv em en t w as ch all en gin g. Th e r ev iew in clu de d 2 6 clu ste r R CT s w ith pr oc es s ev alu at ion da ta pu bli sh ed be tw ee n 1 99 6 a nd 20 13 . Th es e s tu die s w er e d ra wn fro m 34 st ud ies fo cu sin g on di et an d P A o ut of 67 stu die s i nc lud ed in a re vie w of th e e ffe cti ve ne ss of th e HP S f ra m ew or k p ub lis he d in 20 14 . Th e p ro gr am m e i nt er ve nt ion s fo cu se d o n d iet an d/ or PA an d w er e i m ple m en te d th ro ug h t he th re e H PS fra m ew or k s tra te gie s: cu rri cu lum , e nv iro nm en t a nd fam ilie s/c om m un iti es (e .g. he alt h e du ca tio n p ro m ot ed th ro ug h f or m al sch oo l cu rri cu lum , c ha ng es to th e sch oo l's ph ys ica l o r s oc ial en vir on m en t, en ga ge m en t wi th fa m ilie s a nd th e w ide r co m m un ity ). Ch ild re n a ge d 5 –1 5 y ea rs in pr im ar y a nd se co nd ar y s ch oo ls in th e U nit ed St at es , E ur op e an d A us tra lia . Sc ho ol- ba se d d iet an d/ or PA pr og ra m m es fo llo wi ng th e H PS fr am ew or k w er e s ho wn as eff ec tiv e a t i nc re as ing PA , fi tn es s a nd F& V int ak e i n s ch oo lch ild re n ( La ng fo rd et al ., 2 01 4) . Fu rth er pr oc es s e va lua tio ns re ve ale d h igh le ve ls of ac ce pt ab ilit y a m on g t ea ch er s a nd st ud en ts, bu t i m ple m en ta tio n fi de lit y v ar ied co ns ide ra bly ac ro ss tri als . In vo lvi ng fa m ilie s, de sp ite be ing a k ey pa rt of th e H PS ap pr oa ch , w as re po rte d as hi gh ly ch all en gin g d ue to fa cto rs su ch as lan gu ag e b ar rie rs, m isu nd er sta nd ing th e pu rp os e o f t he pr og ra m m e a nd la ck of ti m e o r op po rtu nit y. In te rv en tio n e lem en ts ide nt ifi ed as im po rta nt in clu de d: ta ilo rin g p ro gr am m es to ind ivi du al sch oo ls’ ne ed s; ali gn ing in te rv en tio ns wi th sc ho ols ’ co re ai m s; wo rk ing w ith te ac he rs to de ve lop pr og ra m m es an d i nc re as e o wn er sh ip; an d p ro vid ing on -g oin g t ra ini ng , s up po rt an d co m m un ica tio n. Th e e m ph as is on ac ad em ic su bje cts (a nd th e c or re sp on din g l ow va lue pl ac ed on he alt h i nit iat ive s), an d l ac k o f in sti tu tio na l su pp or t w er e c ite d a s b ar rie rs to im ple m en ta tio n, as w as la ck of sp ac e f or PA , a nd la ck of m ot iva tio n of sc ho ol ca nt ee ns to in tro du ce he alt hie r f oo ds th at m ay no t b e p op ula r a nd th er efo re no t fin an cia lly vi ab le. La ve lle , M ac ka y an d P ell , 2 01 2 (5 5) 1.1 , 3 .2 Th is m et a- an aly sis of 43 stu die s o n s ch oo l-b as ed int er ve nt ion s f ou nd sig nifi ca nt ev ide nc e t ha t th es e i nt er ve nt ion s c an sig nifi ca nt ly re du ce ch ild re n’s BM I, e sp ec ial ly if t he y inc lud e a ph ys ica l e xe rci se co m po ne nt . Th e r ev iew in clu de s 4 3 stu die s ( clu ste r R CT , e ffi ca cy tri al, qu as i-e xp er im en ta l, pil ot RC T, qu as i-R CT , p ilo t pr os pe cti ve st ud y) pu bli sh ed be tw ee n 1 99 1 a nd 20 10 . In te rv en tio ns w er e c om pr ise d of PA on ly in 11 st ud ies (2 6% ), ed uc at ion on ly in th re e ( 7% ) a nd co m bin at ion s of th es e a nd im pr ov ed nu tri tio n i n t he re m ain ing 29 (6 7% ). Am on g t he 43 stu die s, 34 in clu de d P A, 6 inc lud ed an en vir on m en ta l co m po ne nt (e .g. ch an ge s t o sch oo l m ea ls or in sta lla tio n of he alt hy ve nd ing m ac hin es ) an d 2 8 i nc lud ed ed uc at ion on nu tri tio n. Ch ild re n a ge d ≤ 18 ye ar s (n = 36 57 9) in pr e- , p rim ar y an d s ec on da ry sc ho ols in th e U nit ed St at es , A us tra lia , Ge rm an y, Sp ain , B ra zil , Fra nc e, Ne th er lan ds , In dia , Ca na da , S ing ap or e, Eg yp t, Be lgi um , Ir ela nd , C hin a, Ch ile , Sw itz er lan d, Gr ee ce , T ha ila nd an d U nit ed Ki ng do m . M et a- an aly sis sh ow ed a sig nifi ca nt re du cti on in BM I o ve ra ll f or al l s tu die s ( -0 .17 kg /m 2; 95 % CI : -0 .08 , - 0.2 6: P= 0.0 01 ), fo r P A u se d i n i so lat ion (-0 .13 kg /m 2; 95 % CI : - 0.2 2, -0 .04 : P = 0.0 01 ) a nd co m bin ed w ith im pr ov ed nu tri tio n ( -0 .17 kg /m 2; 95 % CI : - 0.2 9, -0 .06 ; P < 0. 00 1) . In te rv en tio ns ta rg et ed at ov er we igh t o r o be se ch ild re n ( -0 .35 kg /m 2; 95 % CI : - 0.5 8, -0 .12 ; P = 0.0 03 ) h ad a gr ea te r e ffe ct on BM I t ha n t ho se de liv er ed to all ch ild re n ( -0 .16 kg /m 2; 95 % CI : - 0.2 5, -0 .06 ; P= 0.0 02 ). 130 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Li et al ., 2 00 8 (1 52 ) 4.1 Th is sy ste m at ic re vie w of 22 sc ho ol- ba se d o be sit y pr ev en tio n o r r ed uc tio n int er ve nt ion s i n C hin a f ou nd th at th os e i nc lud ing di et ar y m od ifi ca tio ns im pr ov ed we igh t-r ela te d o ut co m es . Th e r ev iew in clu de d 2 2 stu die s ( 12 RC Ts, 6 we re no n- ra nd om ize d c on tro lle d t ria ls an d 4 w er e p re - a nd po st- te st co m pa ris on s) pu bli sh ed be tw ee n 1 99 0 a nd 20 06 . He alt h e du ca tio n, PA an d/ or di et ar y i nt er ve nt ion . Th irt ee n o f t he 22 st ud ies inc lud ed di et ar y m od ifi ca tio n int er ve nt ion (e .g. in cre as e F& V a nd le gu m es , re str ict SS Bs , p or tio n s ize co nt ro l) co m bin ed w ith PA an d/ or he alt h e du ca tio n. Ch ild re n a nd ad ole sce nt s a ge d 3– 19 at sc ho ols in Ch ina ; 1 7 o f th e 2 2 s tu die s w er e t ar ge te d a t ov er we igh t o r o be se ch ild re n. Ov er all , m os t s tu die s r ep or te d a be ne fic ial eff ec t of th e i nt er ve nt ion w ith on e o r m or e o f t he stu dy ou tco m es , b ut al l t he st ud ies ha d s er iou s or m od er at e m et ho do log ica l w ea kn es se s. T en of th e 1 3 s tu die s i nc lud ing di et ar y m od ifi ca tio n re po rte d o n w eig ht -re lat ed ou tco m es an d all of th es e s tu die s r ep or te d e ffe cti ve ne ss on ov er we igh t a nd ob es ity pr ev ale nc e o r B M I. Lis sa u, 20 07 (8 4) 1.1 Th is sy ste m at ic re vie w of 14 ob es ity pr ev en tio n stu die s i n t he sc ho ol ar en a fo un d t ha t m ult ico m po ne nt stu die s i nc lud ing a nu tri tio n co m po ne nt w er e l es s o fte n eff ec tiv e. Th e r ev iew in clu de d 1 4 stu die s ( RC Ts, cl us te r C CT s, no n- ra nd om ize d c on tro lle d tri als ) p ub lis he d b et we en 19 95 an d 2 00 5. Ob es ity pr ev en tio n int er ve nt ion s ( e.g . e du ca tio n, fo od se rv ice , e nv iro nm en t) inc lud ing co m po ne nt s s uc h as nu tri tio n/ die t a nd PA . Ch ild re n i n p rim ar y a nd se co nd ar y s ch oo ls, la rg ely in th e U nit ed St at es an d E ur op e, as w ell as Ch ile . Ov er all , h alf of th e s tu die s w er e s uc ce ssf ul an d ha d a n e ffe ct on ei th er ov er we igh t o r o be sit y. Ho we ve r, o nl y 2 of th e 6 m ult ico m po ne nt st ud ies th at in clu de d a nu tri tio n c om po ne nt w er e eff ec tiv e. Lu ck ne r, M os s a nd Ge ric ke , 2 01 2 ( 51 ) 1.1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 68 si ng le an d m ult ico m po ne nt ob es ity pr ev en tio n s tu die s, of w hic h 50 re lat ed to nu tri tio n of ch ild re n i n s ch oo ls, ide nt ifi ed m ult ico m po ne nt int er ve nt ion s i n s ch oo ls an d e nc ou ra gin g r ed uc ed te lev isi on vi ew ing as pr om isi ng st ra te gie s t o pr ev en t o be sit y. Th e r ev iew in clu de d 6 8 co nt ro lle d r an do m ize d o r qu as i-e xp er im en ta l s tu die s pu bli sh ed be tw ee n 1 98 2 an d 2 00 8. Sin gle an d m ult ico m po ne nt int er ve nt ion s r ela te d t o nu tri tio n ( ch an ge in at lea st on e d ail y m ea l), PA (re gu lar ex er cis e) , t ele vis ion (in te rv en tio ns ai m ed at re du cin g t ele vis ion vi ew ing ) an d e du ca tio n ( dis se m ina tio n of in fo rm at ion or te ac hin g on ei th er ge ne ra l h ea lth y be ha vio ur or sp ec ifi ca lly re lat ed to nu tri tio n, PA or se de nt ar y b eh av iou r). Ch ild ren (5 5 s tu die s) an d a du lts (1 3 s tu die s) in th e U nit ed Sta tes , E ur op e, Au str ali a, Ca na da an d N ew Ze ala nd . Of th e 5 5 s tu die s f oc us ing on ch ild ren , 5 0 s tu die s w ere sch oo l-b as ed , 3 st ud ies inc lud ed yo un ge r c hil dr en (a ge d < 6 y ea rs) , 3 4 i nc lud ed ch ild ren (a ge d 6 –1 2 y ea rs) an d 18 in clu de d a do les ce nt s ( ag ed 12 –1 8 y ea rs) . In ch ild re n, th e h igh es t o be sit y r ed uc tio ns w er e ac hie ve d t hr ou gh pr om ot ing re du ce d t ele vis ion vie wi ng (M D: -0 .27 kg /m 2; 95 % CI : - 0.4 , - 0.1 3; 8 s tu die s), se co nd la rg es t t hr ou gh co m bin ing PA w ith ed uc at ion (M D: -0 .19 kg /m 2; 95 % CI : -0 ..3 7, -0 .02 ; 2 st ud ies ) a nd th ird la rg es t t hr ou gh ed uc at ion al on e ( M D: -0 .15 kg /m 2; 95 % CI : -0 .24 , - 0.0 7; 15 st ud ies ). On ly PA ha d a n e ffe ct on pe rce nt ag e o f b od y f at . In ot he r in te rv en tio ns , he te ro ge ne ity w as to o h igh or sh ow ed no sta tis tic all y s ign ifi ca nt re du cti on in ou tco m es . 131ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Lu yb li, Sc hm ille n an d S ot os -P rie to , 20 19 (7 0) 1.1 , 2 .2 Th is sco pin g r ev iew of 15 pr es ch oo l in te rv en tio ns in low so cio ec on om ic se tti ng s in HI C o n a dip os ity - re lat ed ou tco m es am on g pr es ch oo ler s f ou nd th at bo th su cce ssf ul an d i nc on clu siv e stu die s w er e m ult ico m po ne nt — in clu din g n ut rit ion , P A an d p ar en t i nv olv em en t. Th e r ev iew in clu de d 1 5 stu die s, 14 RC Ts an d 1 qu as i-e xp er im en ta l s tu dy , pu bli sh ed be tw ee n 2 00 5 an d 2 01 6. Ob es ity re du cti on int er ve nt ion s ( e.g . e du ca tio n, pa re nt in vo lve m en t, en vir on m en ta l, e co no m ic) fo cu sin g o n n ut rit ion , P A a nd scr ee n t im e. Pr es ch oo ler s a ge d 2 –5 ye ar s in ru ra l, l ow so cio ec on om ic or un de rse rv ed se tti ng s, lar ge ly in th e U nit ed St at es , a s w ell as in Be lgi um , G er m an y, Fra nc e, Isr ae l, N ew Ze ala nd an d Sw itz er lan d. Al l 9 su cce ssf ul int er ve nt ion s i n t er m s o f re du cin g a dip os ity -re lat ed ou tco m es (B M I z-s co re, w ais t c irc um fer en ce , p er ce nt bo dy fat ) w er e m ult ico m po ne nt , w hic h i nc lud ed nu tri tio n ( e.g . e du ca tio na l n ut rit ion le sso ns ), PA , scr ee n t im e a nd pa re nt in vo lve m en t. Sim ila rly , th e 6 in co nc lus ive in te rv en tio ns th at di d n ot sig nifi ca nt ly re du ce an th ro po m et ric m ea su re s we re al so m ult ico m po ne nt , w hic h i nc lud ed nu tri tio n, PA , m ed ia us e, sle ep an d p ar en t inv olv em en t. T he au th or s c on clu de d t ha t o be sit y ou tco m es ca n b e t ar ge te d i n l ow so cio ec on om ic se tti ng s t hr ou gh sc ho ol int er ve nt ion s w ith a m ult ico m po ne nt ap pr oa ch (n ut rit ion an d P A) an d th e i nc lus ion of pa re nt s, bu t t ha t f ur th er re se ar ch is ne ed ed re ga rd ing ap pr op ria te in te rv en tio ns . M an sfi eld an d Sa va ian o, 20 17 (4 3) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 31 st ud ies of th e U nit ed St at es N at ion al Sc ho ol Lu nc h Pr og ra m nu tri tio n s ta nd ar ds un de r t hr ee di ffe re nt leg isl at ive er as co nc lud ed th at in cre as ing ac ce ss to he alt hy fo od s d ur ing sc ho ol lun ch im pr ov ed st ud en ts’ die ta ry in ta ke s. Th e r ev iew in clu de d 3 1 stu die s ( int er ve nt ion stu die s, lon git ud ina l ob se rv at ion st ud ies , o ne - tim e o bs er va tio n s tu die s) pu bli sh ed be tw ee n 2 00 7 an d 2 01 6. Sc ho ol lun ch ch an ge s a s re su lt of th re e s eq ue nt ial fed er al po lic ies : 1 ) sch oo l w ell ne ss po lic ies (im ple m en te d 2 00 6– 20 07 ); 2) th e H ea lth y, Hu ng er -F re e Kid s A ct (p as se d 2 01 0– 20 12 ); an d 3 ) t he H ea lth y, Hu ng er - Fre e K ids Ac t ( im ple m en te d 20 12 –p re se nt ). Ch ild re n a nd ad ole sce nt s a ge d 5– 18 ye ar s i n s ch oo ls in th e Un ite d S ta te s. Of th e 1 9 i nt er ve nt ion an d l on git ud ina l ob se rv at ion st ud ies , 1 4 r ep or te d i m pr ov ed fo od co ns um pt ion be ha vio ur s ( e.g . in cre as ed se lec tio n, in ta ke an d s ale s o f h ea lth y f oo ds , a nd de cre as ed pl at e w as te ). On ly 2 o f 1 2 o ne -ti m e ob se rv at ion st ud ies re po rte d f oo d c on su m pt ion be ha vio ur s m ee tin g t ar ge t n ut rit ion st an da rd s. 132 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W M ay ne , Au ch inc los s a nd M ich ae l, 2 01 5 (3 9) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 37 st ud ies ev alu at ing th e effi ca cy of po lic y a nd bu ilt en vir on m en t c ha ng es on ob es ity -re lat ed ou tco m es fo un d i m pr ov em en ts in pu rch as ing or se lf- re po rte d die t f ro m re gu lat ion s t ha t re qu ire d i m pr ov em en ts to th e f oo d e nv iro nm en t. Th e r ev iew in clu de d 37 st ud ies of na tu ra l o r qu as i-e xp er im en ta l d es ign , pu bli sh ed be tw ee n 2 00 5 an d 2 01 4. In te rv en tio ns in clu de d int ro du cti on of po lic ies (d efi ne d a s m un ici pa l or fe de ra l g ov er nm en t re gu lat ion s a nd la ws inc lud ing sc ho ol dis tri ct po lic ies ) o r c ha ng e t o t he bu ilt en vir on m en t t ha t co uld aff ec t P A, di et or ob es ity . T he 5 sch oo l-b as ed nu tri tio n p oli cie s c on ce rn ed re str ict ion s o n s ug ar y f oo ds an d b ev er ag es or hi gh er fat fo od s, an d i nc re as es in av ail ab ilit y o f m ilk or F& V. Ad ult s a nd ch ild re n i n a ny se tti ng , la rg ely in th e U nit ed St at es as w ell as in Au str ali a, th e U nit ed Ki ng do m , C an ad a, Ch ile , a nd N ew Ze ala nd . Ov er all , t he m ajo rit y o f s tu die s e va lua tin g re gu lat ion s t ha t r eq uir ed im pr ov em en ts to th e f oo d e nv iro nm en t f ou nd im pr ov em en ts in pu rch as ing or se lf- re po rte d d iet , w hil e s tu die s th at si m ply re qu ire d t he po sti ng of nu tri tio na l inf or m at ion fo un d l itt le eff ec t, wi th a few ex ce pt ion s. Of th e 5 sc ho ol- ba se d n ut rit ion - re lat ed po lic ies , 4 ha d r es ult s i n t he ex pe cte d dir ec tio ns (e .g. re du cin g e ne rg y d en sit y, low nu tri en t d en se fo od , S SB s, ca lor ies or fa t, or inc re as ing F& V) an d 1 ha d m ixe d r es ult s. On ly 1 of th e 5 sc ho ol po lic y s tu die s m ea su re d B M I a nd re po rte d m ixe d r es ult s. 133ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W M cIs aa c e t a l., 20 19 (8 6) 1.2 , 1 .3, 1. 4, 1.5 Th is sco pin g r ev iew of 59 st ud ies id en tifi ed fac ilit at or s a nd ba rri er s t o t he im ple m en ta tio n o f s ch oo l nu tri tio n p oli cie s a nd de riv ed a s et of op po rtu nit ies fo r ac tio n t o m ak e s uc h p oli cie s m or e s uc ce ssf ul. Th e r ev iew in clu de d 5 9 stu die s ( 24 qu ali ta tiv e, 18 qu an tit at ive , 9 m ixe d m et ho ds , 8 re vie w stu die s) pu bli sh ed be tw ee n 2 00 4 an d 2 01 7. Va rio us nu tri tio n s tra te gie s (n ut rit ion st an da rd s, we lln es s an d h ea lth pr om ot ion , he alt hy ca nt ee n a nd sn ac ks , nu tri tio n e du ca tio n) ac ro ss m ult ipl e j ur isd ict ion al lev els , inc lud ing na tio na l le ve l ( 19 stu die s), st at e o r p ro vin cia l lev el (3 0 s tu die s), di str ict , co m m un ity or lo ca l le ve l ( 5 stu die s) an d m ult ile ve l ( 2 stu die s); 3 stu die s d id no t re po rt lev el. Ch ild re n a nd ad ole sce nt s i n ele m en ta ry /p rim ar y s ch oo l ( 13 stu die s), ju nio r/m idd le sch oo l (5 st ud ies ), hig h/ se co nd ar y sch oo l ( 7 s tu die s), m ult ipl e sch oo l le ve ls, an d s ta ff or ot he r sta ke ho lde rs (3 2 s tu die s). M os t stu die s w er e c on du cte d i n t he Un ite d S ta te s a nd ot he r H IC. Ac ro ss th e fi na l s tu die s i de nt ifi ed , fa cto rs em er ge d a s f ac ilit at or s ( or ba rri er s i n t he ca se of la ck of su ch ) t o t he im ple m en ta tio n o f s ch oo l nu tri tio n p oli cie s, wi th sy ste m im pli ca tio ns re lat ed to fiv e a re as to su pp or t p oli cy ac tio n. Op po rtu nit ies fo r a cti on w er e, fir st, to pr ov ide m ac ro -le ve l s up po rt to en co ur ag e p oli cy im ple m en ta tio n. Th is wa s a na lys ed in 41 st ud ies in te rm s o f c lar ity of po lic y e xe cu tio n a nd po lic y lan gu ag e, tra ini ng su pp or t a nd re so ur ce s f or sch oo l s ta ff, m on ito rin g a nd en fo rce m en t o f t he po lic ies an d c om m un ica tio n b et we en sc ho ols , po lic y- m ak er s a nd ot he r s ta ke ho lde rs. Th e se co nd op po rtu nit y i de nt ifi ed w as to ad dr es s th e fi na nc ial im pli ca tio ns of he alt hy fo od ac ce ss, an d t his w as in clu de d i n 3 9 s tu die s i n t er m s o f ac ce ss to an d p ro cu re m en t o f f ru it an d v eg et ab les an d o th er nu tri tio us op tio ns th at of te n h av e hig he r c os ts bu t g en er at e m or e r ev en ue an d hig he r p ro fit . T his ca n l ea d t o t ra de -o ffs be tw ee n ke ep ing co sts lo w fo r s tu de nt s a nd fa m ilie s a nd ge ne ra tin g s uffi cie nt in co m e f ro m ca fet er ias an d c an te en s, wh ile al so el im ina tin g h ist or ic pr ac tic es in sc ho ols of us ing un he alt hy fo od s t o ra ise fu nd s. T hir d, ali gn ing nu tri tio n a nd co re sch oo l p rio rit ies w as id en tifi ed as an op po rtu nit y fo r a cti on ; t his w as ex am ine d i n 2 0 s tu die s i n te rm s o f c om pe tin g p rio rit ies an d d em an ds on , fo r e xa m ple , a ca de m ic as se ssm en t t ha t m ak e it diffi cu lt fo r s ch oo ls to im ple m en t n ut rit ion - re lat ed ch an ge s a nd th e t im e- co ns um ing na tu re of po lic y i m ple m en ta tio n a nd pa rtn er inv olv em en t. 134 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W M cIs aa c e t a l., 20 19 (8 6) 1.2 , 1 .3, 1. 4, 1.5 Th is sco pin g r ev iew of 59 st ud ies id en tifi ed fac ilit at or s a nd ba rri er s t o t he im ple m en ta tio n o f s ch oo l nu tri tio n p oli cie s a nd de riv ed a s et of op po rtu nit ies fo r ac tio n t o m ak e s uc h p oli cie s m or e s uc ce ssf ul. Th e r ev iew in clu de d 5 9 stu die s ( 24 qu ali ta tiv e, 18 qu an tit at ive , 9 m ixe d m et ho ds , 8 re vie w stu die s) pu bli sh ed be tw ee n 2 00 4 an d 2 01 7. Va rio us nu tri tio n s tra te gie s (n ut rit ion st an da rd s, we lln es s an d h ea lth pr om ot ion , he alt hy ca nt ee n a nd sn ac ks , nu tri tio n e du ca tio n) ac ro ss m ult ipl e j ur isd ict ion al lev els , inc lud ing na tio na l le ve l ( 19 stu die s), st at e o r p ro vin cia l lev el (3 0 s tu die s), di str ict , co m m un ity or lo ca l le ve l ( 5 stu die s) an d m ult ile ve l ( 2 stu die s); 3 stu die s d id no t re po rt lev el. Ch ild re n a nd ad ole sce nt s i n ele m en ta ry /p rim ar y s ch oo l ( 13 stu die s), ju nio r/m idd le sch oo l (5 st ud ies ), hig h/ se co nd ar y sch oo l ( 7 s tu die s), m ult ipl e sch oo l le ve ls, an d s ta ff or ot he r sta ke ho lde rs (3 2 s tu die s). M os t stu die s w er e c on du cte d i n t he Un ite d S ta te s a nd ot he r H IC. Ac ro ss th e fi na l s tu die s i de nt ifi ed , fa cto rs em er ge d a s f ac ilit at or s ( or ba rri er s i n t he ca se of la ck of su ch ) t o t he im ple m en ta tio n o f s ch oo l nu tri tio n p oli cie s, wi th sy ste m im pli ca tio ns re lat ed to fiv e a re as to su pp or t p oli cy ac tio n. Op po rtu nit ies fo r a cti on w er e, fir st, to pr ov ide m ac ro -le ve l s up po rt to en co ur ag e p oli cy im ple m en ta tio n. Th is wa s a na lys ed in 41 st ud ies in te rm s o f c lar ity of po lic y e xe cu tio n a nd po lic y lan gu ag e, tra ini ng su pp or t a nd re so ur ce s f or sch oo l s ta ff, m on ito rin g a nd en fo rce m en t o f t he po lic ies an d c om m un ica tio n b et we en sc ho ols , po lic y- m ak er s a nd ot he r s ta ke ho lde rs. Th e se co nd op po rtu nit y i de nt ifi ed w as to ad dr es s th e fi na nc ial im pli ca tio ns of he alt hy fo od ac ce ss, an d t his w as in clu de d i n 3 9 s tu die s i n t er m s o f ac ce ss to an d p ro cu re m en t o f f ru it an d v eg et ab les an d o th er nu tri tio us op tio ns th at of te n h av e hig he r c os ts bu t g en er at e m or e r ev en ue an d hig he r p ro fit . T his ca n l ea d t o t ra de -o ffs be tw ee n ke ep ing co sts lo w fo r s tu de nt s a nd fa m ilie s a nd ge ne ra tin g s uffi cie nt in co m e f ro m ca fet er ias an d c an te en s, wh ile al so el im ina tin g h ist or ic pr ac tic es in sc ho ols of us ing un he alt hy fo od s t o ra ise fu nd s. T hir d, ali gn ing nu tri tio n a nd co re sch oo l p rio rit ies w as id en tifi ed as an op po rtu nit y fo r a cti on ; t his w as ex am ine d i n 2 0 s tu die s i n te rm s o f c om pe tin g p rio rit ies an d d em an ds on , fo r e xa m ple , a ca de m ic as se ssm en t t ha t m ak e it diffi cu lt fo r s ch oo ls to im ple m en t n ut rit ion - re lat ed ch an ge s a nd th e t im e- co ns um ing na tu re of po lic y i m ple m en ta tio n a nd pa rtn er inv olv em en t. AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Fo ur th , d ev elo pin g a co m m on pu rp os e a nd re sp on sib ilit y a m on g s ta ke ho lde rs, w hic h wa s i nc lud ed in 45 st ud ies in te rm s o f s ch oo l cli m at e, va lue s, op ini on s a nd cu ltu ra l n or m s re lat ed to fo od w ith in sch oo ls, at ho m e o r in th e co m m un iti es . T he se cu ltu ra l n or m s i nc lud ed th e pe rce ive d s tu de nt pr efe re nc e f or le ss he alt hy fo od s, th e s ch oo l c ap ac ity fo r l ea de rsh ip an d co or din at ion an d e ng ag em en t o f p ar en ts. A fif th ar ea of op po rtu nit y w as re co gn iti on of sc ho ol an d co m m un ity ch ar ac te ris tic s, wh ich w as ex am ine d in 33 st ud ies in te rm s o f c on te xt ua lis ing th e po lic ies to sc ho ol loc at ion an d d em og ra ph ics , ad dr es sin g i ssu es of po ve rty an d f oo d i ns ec ur ity th at m ay le ad to in eq uit y, sch oo l in fra str uc tu re (su ch as ca fet er ias or ga rd en s), w ide r inf ra str uc tu re an d t he ac ce ss to co m pe tit ive fo od s fro m th e s ch oo l s ur ro un din gs or at ho m e. M ea d e t a l., 20 17 (6 3) 1.1 Th is Co ch ra ne re vie w an d m et a- an aly sis of 70 st ud ies of be ha vio ur -c ha ng ing int er ve nt ion s f oc us ing on di et an d P A f or th e t re at m en t o f ov er we igh t o r o be sit y f ou nd th at su ch in te rv en tio ns m ay be be ne fic ial in ac hie vin g sm all , s ho rt- te rm re du cti on s in BM I, B M I z -sc or e a nd we igh t. Th e r ev iew in clu de d 7 0 R CT s pu bli sh ed be tw ee n 1 98 4 an d 2 01 6. Be ha vio ur ch an ge int er ve nt ion s ( e.g . e du ca tio n, co un se llin g, he alt hy fo od ch oic es , e nv iro nm en ta l co nt ro l) fo cu sin g o n d iet (2 stu die s), PA (6 st ud ies ) o r bo th (6 4 s tu die s). Ov er we igh t, ob es e o r s ev er ely ob es e c hil dr en ag ed 6– 11 ye ar s i n t he U nit ed St at es , Eu ro pe , A us tra lia , N ew Ze ala nd , B ra zil , C an ad a, Ho ng Ko ng , J ap an , M ala ys ia an d M ex ico . S tu die s w er e co nd uc te d i n m ult ipl e s et tin gs (sc ho ol, co m m un ity , h om e, pr im ar y a nd se co nd ar y c ar e, re se ar ch cl ini cs) ; 4 ou t o f 7 0 stu die s w er e c on du cte d i n t he sch oo l s et tin g. Th e b eh av iou r c ha ng e i nt er ve nt ion s s ign ifi ca nt ly re du ce d B M I ( -0 .53 kg /m 2; 95 % CI : 0 .82 , - 0.2 4; 27 85 pa rti cip an ts; 24 tr ial s), BM I z -sc or e ( -0 .06 un its ; 9 5% CI : - 0.1 0, -0 .02 ; 4 01 9 p ar tic ipa nt s; 37 tr ial s) an d w eig ht (- 1.4 5 k g; 95 % CI : - 1.8 8. -1 .02 ; 1 77 4 p ar tic ipa nt s; 17 tr ial s), bu t q ua lit y o f ev ide nc e w as gr ad ed as lo w. N o d iff er en ce s w er e ob se rv ed in su bg ro up an aly se s o n i nt er ve nt ion ty pe or se tti ng ; h ow ev er, in su bg ro up an aly se s ba se d o n s et tin g, th e t wo sc ho ol ba se d s tu die s inc lud ed in th e m et a- an aly se s d id no t s ho w sig nifi ca nt eff ec t o n B M I ( -0 .57 kg /m 2; 95 % CI : -4 .94 , 3 .80 ; 2 1 p ar tic ipa nt s; 1 t ria l), BM I z -sc or e (-0 .01 ; 9 5% CI : - 0.1 7, 0.1 5; 76 pa rti cip an ts; 2 tri als ) o r w eig ht (- 1.2 0 k g; 95 % CI : - 4.2 0, 1.8 0; 55 pa rti cip an ts; 1 tri al) . 135ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W M eik lej oh n, Ry an an d P ale rm o, 20 16 (7 4) 1.1 , 2 .2, 3. 1 Th is sy ste m at ic re vie w of sc ho ol- ba se d m ult i- str at eg y n ut rit ion ed uc at ion int er ve nt ion s f ou nd th at pa re nt al inv olv em en t a nd fac ilit at ion of nu tri tio n ed uc at ion by cl as sro om or ho m e e co no m ics te ac he rs we re th e c om po ne nt s t ha t m os t o fte n i m pr ov ed BM I a nd die ta ry in ta ke . Th e r ev iew in clu de d 1 1 R CT stu die s r ep or te d i n 1 3 p ap er s pu bli sh ed be tw ee n 2 00 0 an d 2 01 4. Nu tri tio n e du ca tio n int er ve nt ion s t o a dd re ss co ns um pt ion of F& V, wa te r, fis h, SS B, fa t, fib re or su cro se we re im ple m en te d a s p ar t o f m ult ico m po ne nt st ra te gie s. Co m po ne nt s i nc lud ed pa re nt al inv olv em en t, fac ilit at ion of nu tri tio n ed uc at ion by cl as sro om or ho m e e co no m ics te ac he rs or ot he r s ch oo l s ta ff, th eo ry -b as ed in str uc tio na l str at eg ies , in co rp or at ing po lic y c ha ng es w ith in sch oo l s et tin gs (c an te en s, fo od su pp ly an d v en din g m ac hin es ) a nd co m bin ing nu tri tio n e du ca tio n w ith PA pr og ra m m es . Ch ild re n a nd ad ole sce nt s a ge d 10 –1 8 y ea rs in se co nd ar y sch oo ls in HI C, m os tly in Eu ro pe an d a fe w in Au str ali a an d U nit ed St at es Of th e 1 1 s tu die s, 10 re po rte d s ign ifi ca nt ch an ge s in an th ro po m et ric m ea su re s ( we igh t, BM I, B M I z-s co re, sk inf old s, wa ist ci rcu m fer en ce or pe rce nt bo dy fa t) or di et ar y i nt ak e ( F& V, sn ac k f oo ds , fat s, su ga r, S SB ). T he m ult i-s tra te gy nu tri tio n ed uc at ion in te rv en tio n c om po ne nt s t ha t m os t of te n i m pr ov ed an th ro po m et ric s o r d iet ar y co ns um pt ion in th is ag e g ro up w er e p ar en ta l inv olv em en t ( co m po ne nt in al l 4 st ud ies w ith be ne fic ial eff ec t o n B M I a nd 8 ou t o f 9 st ud ies wi th be ne fic ial eff ec t o n d iet ar y i nt ak e) an d fac ilit at ion of nu tri tio n e du ca tio n b y c las sro om or ho m e e co no m ics te ac he rs or ot he r s ch oo l s ta ff (co m po ne nt in 3 ou t o f 4 st ud ies w ith be ne fic ial eff ec t o n B M I a nd al l 9 st ud ies w ith be ne fic ial eff ec t o n d iet ar y i nt ak e) . O th er co nt rib ut ing fac to rs we re th eo ry -b as ed in str uc tio na l str at eg ies , in co rp or at ing po lic y c ha ng es w ith in sch oo l s et tin gs (c ha ng es in ca nt ee ns , fo od su pp ly an d v en din g m ac hin es ) a nd co m bin ing nu tri tio n ed uc at ion w ith PA pr og ra m m es . M ich a e t a l., 20 18 (3 6) 1.1 , 4 .1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 91 st ud ies on th e e ffe ct of sc ho ol fo od en vir on m en t p oli cie s f ou nd im pr ov ed ta rg et ed di et ar y be ha vio ur s i n c hil dr en bu t i nc on clu siv e e ffe ct on ad ipo sit y a nd m et ab oli c m ea su re s. Th e r ev iew in clu de d 9 1 sch oo l fo od en vir on m en t po lic y i nt er ve nt ion s i nc lud ing 39 ra nd om ize d a nd 52 qu as i-e xp er im en ta l s tu die s, pu bli sh ed be tw ee n 1 99 1 an d 2 01 6. Sc ho ol foo d e nv iro nm en t po lic y i nt er ve nt ion s f oc us ing on di rec t o r in dir ec t p rov isi on of he alt hf ul foo ds an d be ve rag es (4 0 s tu die s o f F &V sch em es , w ate r f ou nt ain s, inc rea se d a va ila bil ity of he alt hy fo od s a t v en din g m ac hin es , e tc. ), c om pe tit ive foo d a nd be ve rag e n ut rit ion al qu ali ty sta nd ard s ( 29 st ud ies of res tri cti ve po lic ies on SS Bs an d u nh ea lth y s na ck s, or ba se d o n n ut rie nt s, ca lor ies or po rti on si ze s) an d/ or sch oo l m ea l n ut rit ion qu ali ty sta nd ard s ( 39 st ud ies of sc ho ol lun ch or br ea kfa st sta nd ard s). Ch ild re n a nd ad ole sce nt s i n pr es ch oo l ( 1 s tu dy ), pr im ar y (4 7 s tu die s) or se co nd ar y ( 27 ) sch oo l a nd on e a t m ult ipl e lev els , m os tly in th e U nit ed St at es , C an ad a, Ne w Ze ala nd an d E ur op e, bu t a lso Re pu bli c of Ko re a a nd Ir an . Di rec t o r in dir ec t p rov isi on po lic ies , w hic h l arg ely ta rg ete d F &V , in cre as ed co ns um pt ion of fr uit s (0 .27 se rv ing s/d ; 9 5% CI : 0 .17 , 0 .36 ; 1 5 e sti m ate s), F& V ( 0.2 8 s er vin gs /d ; 9 5% CI : 0 .17 , 0 .40 ; 1 6 es tim ate s) an d v eg et ab les (0 .04 se rv ing s/d ; 9 5% CI: 0. 01 , 0 .08 ; 1 1 e sti m ate s). Ot he r m et a- an aly se s fou nd no eff ec t o n t ot al ca lor ies in ta ke , w ate r co ns um pt ion , o r a ny ad ipo sit y a nd m et ab oli c m ea su res fro m th e p rov isi on po lic ies . C om pe tit ive foo d a nd be ve rag e s ta nd ard s r ed uc ed to ta l co ns um pt ion of SS Bs (- 0.1 8 s er vin gs /d ; 9 5% CI: -0 .31 , - 0.0 5; 2 e sti m ate s) an d u nh ea lth y sn ac ks (- 0.1 7 s er vin gs /d ; 9 5% CI : - 0.2 2, -0 .13 ; 2 es tim ate s) as w ell as in -sc ho ol co ns um pt ion of un he alt hy sn ac ks (- 0.0 5 s er vin gs /d ; 9 5% CI : - 0.0 8, -0 .02 ; 9 es tim ate s). Ot he r m et a- an aly se s f ou nd no eff ec t o n i n- sch oo l S SB co ns um pt ion , to ta l c alo rie s int ak e, or ov er we igh t/o be sit y f ro m co m pe tit ive foo d a nd be ve rag e s ta nd ard s. 136 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Sc ho ol m ea l s ta nd ar ds (m ain ly lun ch ) i nc re as ed fru it int ak e ( 0.7 6 s er vin gs /d ; 9 5% CI : 0 .37 , 1 .16 ; 2 es tim at es ), re du ce d i nt ak e o f t ot al fat (- 1.4 9% E; 95 % CI : - 2.4 2, -0 .57 ; 6 es tim at es ) a nd sa tu ra te d fat (- 0.9 3% E; 95 % CI : - 1.1 5, -0 .70 ; 4 es tim at es ), re du ce d s od ium in ta ke (- 17 0 m g/ d; 95 % CI: -2 42 , -9 8; 4 e sti m at es ) a nd re du ce d B M I p er ce nt ile (-1 .01 ; 9 5% CI : - 1.6 2, -0 .39 , 2 es tim at es ). Po sit ive eff ec ts we re al so se en on in cre as ed m ilk co ns um pt ion an d c ar bo hy dr at e i nt ak e i n s tu die s of po lic ies ta rg et ing th es e e lem en ts. O th er m et a- an aly se s f ou nd no eff ec t o n t ot al ca lor ies , die ta ry fib re, w ho le gr ain s, or ot he r a dip os ity an d m et ab oli c m ea su re s. M ur im i e t a l., 20 18 (1 02 ) 2.2 , 2 .3, 3. 1 Th is sy ste m at ic re vie w of 41 sch oo l- or pr es ch oo l-b as ed nu tri tio n e du ca tio n s tu die s ide nt ifi ed th e c ha ra cte ris tic s of su cce ssf ul int er ve nt ion s a s th os e w ith a m ult ico m po ne nt ap pr oa ch th at w er e a ge ap pr op ria te an d o f a de qu at e du ra tio n, th at ac tiv ely en ga ge d p ar en ts, an d th at en su re d fi de lit y a nd pr op er al ign m en t b et we en th e s ta te d o bje cti ve s, th e int er ve nt ion an d t he de sir ed ou tco m es . Th e r ev iew in clu de d 4 1 stu die s ( clu ste r R CT s, RC Ts, qu as i-e xp er im en ta l, pr e- an d p os t-t es t d es ign , lon git ud ina l, s tu dy w ith int er ve nt ion an d c on tro l, ra nd om ize d, m ult iye ar int er ve nt ion ) p ub lis he d be tw ee n 2 00 9 a nd 20 16 . Nu tri tio n e du ca tio n int er ve nt ion s, wi th or w ith ou t a dd iti on al pr og ra m m e c om po ne nt s. Ch ild re n a nd ad ole sce nt s i n pr es ch oo l ( 7s tu die s), pr im ar y sch oo l ( 26 st ud ies ) a nd se co nd ar y s ch oo l ( 8 s tu die s) wo rld wi de , w ith ab ou t a qu ar te r o f t he st ud ies fr om th e Un ite d S ta te s. Of th e 4 1 s tu die s, 19 w er e s uc ce ssf ul in m ee tin g th eir st at ed pr im ar y r es ea rch ob jec tiv es in te rm s o f n ut rit ion -re lat ed ou tco m es in clu din g we igh t-r ela te d o ut co m es , b ioc he m ica l m ar ke rs (H DL , d iet ar y i nt ak e o f c alo rie s a nd va rio us m ac ro - a nd m icr on ut rie nt s), di et ar y c on su m pt ion or pr efe re nc e ( of F& V, wh ole gr ain s o r u nh ea lth y be ve ra ge s o r s na ck s), di et ar y d ive rsi ty, ch ild re n’s nu tri tio n k no wl ed ge an d f oo d a va ila bil ity in ca fet er ias . N ine te en st ud ies on ly pa rti all y m et an d 3 di d n ot m ee t t he ir s ta te d o bje cti ve s o n nu tri tio n- re lat ed ou tco m es . C om pr eh en siv e, m ult ico m po ne nt an d m ult ile ve l a pp ro ac h nu tri tio n i nt er ve nt ion s w er e m or e l ike ly to su cce ed . L ike wi se , in te rv en tio ns im ple m en te d fo r m or e t ha n 6 m on th s w ith fr eq ue nt ex po su re s we re al so m or e s uc ce ssf ul. Ac tiv e p ar en ta l inv olv em en t ( e.g . fa ce -to -fa ce m ee tin gs , pa rti cip at ion in nu tri tio n o r c oo kin g c las se s, he alt h f air s o r t as tin g s es sio ns ) w as as so cia te d wi th hi gh er su cce ss th an pa ssi ve (e .g. pr ov isi on of m es sa ge s, we bs ite s, ho m ew or k) or no pa re nt inv olv em en t. 137ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Pa re nt in vo lve m en t w as pa rti cu lar ly im po rta nt in th e p re sch oo l s et tin g. Ag e- ap pr op ria te an d ex pe rie nt ial ac tiv iti es w er e a lso as so cia te d wi th su cce ssf ul stu die s ( e.g . h an ds -o n lea rn ing , c oo kin g o r t as tin g s es sio n, pl ay an d ga m es on to pic s i m po rta nt to th e a ge gr ou p) . Fu rth er m or e, stu die s t ha t h ad hi gh fid eli ty (i. e. th at in te rv en tio n a cti vit ies ar e e xe cu te d as pl an ne d i n t he m et ho ds ) a nd th at tr ain ed ex ist ing te ac he rs or en ga ge d n ut rit ion an d P A ex pe rts in th e i m ple m en ta tio n w er e m or e l ike ly to su cce ed . F ina lly , s uc ce ssf ul nu tri tio n e du ca tio n int er ve nt ion s h ad cl ea r a lig nm en t b et we en th e sta te d o bje cti ve s, th e d es ire d o ut co m e a nd th e im ple m en te d a cti vit ies . M uz aff ar, M et ca lfe an d Fie se , 2 01 8 ( 12 5) 3.1 , 3 .3 Th is na rra tiv e r ev iew of 6 sch oo l-b as ed ha nd s-o n cu lin ar y s kil l in te rv en tio ns ide nt ifi ed to o f ew st ud ies to dr aw co nc lus ive fin din gs bu t no te d s ub sta nt ial qu ali ta tiv e eff ec t s ize in in div idu al stu die s o n d iet ar y b eh av iou r an d o th er ou tco m es . Th e r ev iew in clu de d 1 RC T an d 5 qu as i-e xp er im en ta l stu die s p ub lis he d b et we en 19 98 an d 2 01 3. Sc ho ol- ba se d h an ds -o n cu lin ar y s kil l in te rv en tio ns , co m bin ed w ith nu tri tio n ed uc at ion le sso ns (2 st ud ies ), pa re nt an d c om m un ity co m po ne nt s ( 3 s tu die s), ga rd en ing cl as se s ( 2 stu die s), ta sti ng se ssi on s ( 2 stu die s), sc ho ol lun ch ro om co m po ne nt s ( 1 s tu dy ), tri ps to a far m er ’s m ar ke t ( 1 s tu dy ) or vi sit s t o a re sta ur an t ( 1 stu dy ). Ch ild re n a ge d 5 –1 2 y ea rs in sch oo ls in th e U nit ed St at es , Au str ali a a nd th e U nit ed Kin gd om . Al l 6 st ud ies w er e s uc ce ssf ul in m ee tin g t he ir ob jec tiv es . Q ua nt ita tiv e a na lys is av ail ab le fro m 3 stu die s i nd ica te d i m pr ov em en t i n f oo d pr efe re nc es , c oo kin g s kil ls, co ok ing se lf- effi ca cy , co ok ing be ha vio ur al int en tio ns , fo od -p re pa ra tio n fre qu en cy , k no wl ed ge , h ea lth y d iet ar y i nt ak e, BM I a nd bl oo d p re ssu re. Q ua lit at ive ev alu at ion s av ail ab le fro m al l 6 st ud ies in dic at ed su bs ta nt ial eff ec t i n t er m s o f p ro gr am m e a pp ea l a nd im pr ov em en t in co ok ing sk ills an d h ea lth y e at ing . 138 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Na th an et al ., 20 19 (9 1) 2.2 , 4 .1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 10 sc ho ol- or ce nt re -b as ed st ud ies fo un d so m e e vid en ce th at lu nc hb ox int er ve nt ion s a re eff ec tiv e in im pr ov ing ch ild re n’s lun ch bo xe s, pa rti cu lar ly th e inc lus ion of ve ge ta ble s. Th e r ev iew in clu de d 1 0 stu die s ( 8 R CT s a nd 2 qu as i-e xp er im en ta l d es ign s) pu bli sh ed in 13 pa pe rs be tw ee n 2 00 8 a nd 20 16 . In te rv en tio ns to in cre as e pa re nt kn ow led ge vi a de liv er y m od es su ch as pa m ph let s, ne ws let te rs, po ste rs or pa re nt wo rk sh op s. Al l 1 0 t ria ls we re m ult ico m po ne nt int er ve nt ion s; 8 a lso in clu de d an ed uc at ion co m po ne nt fo r c hil dr en , 4 pr ov ide d ph ys ica l re so ur ce s ( e.g . lun ch pa ck s, co nt ain er s) an d 2 p ro vid ed in ce nt ive s f or ch ild re n t o t as te ve ge ta ble s an d f ru it. Tw o c en tre -b as ed stu die s i nc or po ra te d t he de ve lop m en t o f a po lic y a nd th e c om m un ica tio n o f t his to pa re nt s. Se ve n s tu die s we re st an d- alo ne lu nc hb ox int er ve nt ion s a nd 3 we re pa rt of la rg er ch ild ob es ity pr ev en tio n p ro gr am m es . Ch ild re n a ge d 3 –1 4 y ea rs in ch ild ca re ce nt re s ( 4 s tu die s) an d s ch oo ls (6 st ud ies ) i n th e U nit ed St at es , t he U nit ed Kin gd om , A us tra lia , M ex ico an d I sra el. M et a- an aly sis of 4 stu die s ( 2 i n s ch oo ls an d 2 in ce nt re s) fo un d a m od er at e i nc re as e i n p ro vis ion of ve ge ta ble s ( SM D: 0. 40 ; 9 5% CI : 0 .16 , 0 .64 ; P= 0.0 01 ) b ut no t f ru it. Su b- an aly se s b y s et tin g re ve ale d t ha t a si gn ifi ca nt in cre as e r em ain ed in ce nt re -b as ed st ud ies , b ut no t i n s ch oo l-b as ed stu die s. Of th e o th er 4 stu die s m ea su rin g th is ou tco m e b ut no t i nc lud ed in th e m et a- an aly sis , 2 ce nt re -b as ed an d 1 sc ho ol- ba se d stu die s r ep or te d s ign ifi ca nt eff ec ts wh er ea s on e s ch oo l-b as ed st ud y r ep or te d n o s ign ifi ca nt eff ec t. Of th e 4 st ud ies m ea su rin g i nc lus ion of di scr et ion ar y f oo ds (i .e. en er gy -d en se , nu tri en t-p oo r f oo d i te m s s uc h a s s na ck s a nd co nf ec tio na ry ), 1 c en tre -b as ed st ud y r ep or te d po sit ive eff ec ts, 1 ce nt re -b as ed st ud y r ep or te d no eff ec t a nd 2 sch oo l-b as ed st ud ies ha d m ixe d re su lts . O f t he 3 stu die s m ea su rin g i nc lus ion of SS Bs , 1 ce nt re -b as ed st ud y r ep or te d p os iti ve eff ec ts wh er ea s 2 sc ho ol- ba se d s tu die s r ep or te d no n- sig nifi ca nt eff ec t. Of 3 stu die s m ea su rin g t he co nt en ts of ch ild re n’s lu nc hb ox es , 1 ce nt re -b as ed stu dy re po rte d n o e ffe ct, w hil e 2 sc ho ol- ba se d stu die s r ep or te d s ign ifi ca nt eff ec ts. O f 3 st ud ies m ea su rin g c hil dr en ’s d iet ar y i nt ak e o f F &V , 2 sch oo l-b as ed st ud y r ep or te d a si gn ifi ca nt ly hig he r m ea n c on su m pt ion of fr uit , v eg et ab les an d/ or ju ice , w he re as 1 ot he r s ch oo l-b as ed stu die s h ad m ixe d r es ult s. On e s ch oo l-b as ed stu dy m ea su rin g c on su m pt ion of fo od s h igh in fat or su ga rs fo un d n o i nt er ve nt ion eff ec t. Tw o m ult ico m po ne nt st ud ies co nd uc te d i n a ce nt re an d s ch oo l re sp ec tiv ely , re po rte d e ffe cts on w eig ht re lat ed ou tco m es . W hil e a ll i nc lud ed stu die s i n t he re vie w inc or po ra te d a pa re nt co m po ne nt , m an y s im ply re lie d o n p as siv e inf or m at ion di sse m ina tio n s tra te gie s. 139ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ne kit sin g e t a l., 20 18 (1 15 ) 3.1 , 4 .1, 4. 2 Th is m et a- an aly sis of 30 stu die s a im ing to in cre as e ve ge ta ble in ta ke in ch ild re n ag ed 2– 5 y ea rs, 24 of w hic h we re co nd uc te d i n p re sch oo l or ch ild ca re se tti ng s, fo un d th e l ar ge st eff ec t f ro m re pe at ed ta ste ex po su re. Th e r ev iew in clu de d 3 0 stu die s o f v ar iou s d es ign s ( 4 RC T, 8 c lus te r R CT , 6 cr os so ve r, 6 b et we en -su bje cts , 3 wi th in- su bje cts , a nd 3 pr e- an d p os t-t es t d es ign ) pu bli sh ed be tw ee n 2 01 1 an d 2 01 5. In te rv en tio ns to in cre as e ve ge ta ble co ns um pt ion in pr es ch oo l a ge ch ild re n th ro ug h e du ca tio na l int er ve nt ion s, re pe at ed ta ste ex po su re, pa iri ng , c ha ng ed fo od -se rv ice s, ex pli cit re wa rd , m od ell ing , c ho ice , v ar iet y an d v isu al pr es en ta tio n. Ch ild re n a ge d 2 –5 ye ar s, lar ge ly in th e p re sch oo l o r c hil d ca re se tti ng , la rg ely in th e Un ite d S ta te s a nd Eu ro pe , a s we ll a s i n A us tra lia , M ex ico an d Th ail an d. In th e o ve ra ll m et a- an aly sis of H ed ge ’s g , a sm all -m od er at e e ffe ct (g = 0.4 0) of in te rv en tio n wa s o bs er ve d f ro m po oli ng re su lts of 30 st ud ies an d a sl igh tly hi gh er eff ec t ( g= 0.4 2; 95 % CI : 0.3 3, 0.5 1; P< 0.0 01 ) f ro m po oli ng re su lts of all 44 in te rv en tio n a rm s. In su bg ro up an aly se s of in te rv en tio n s tra te gie s, th e s tu die s u sin g ta ste ex po su re ha d a si gn ifi ca nt ly hig he r im pa ct on in ta ke (g = 0.7 9; 95 % CI : 0 .53 , 1 .05 ) th an fo od se rv ice (g = 0.3 0; 95 % CI : 0 .10 , 0 .50 ) or ed uc at ion al (g = 0.3 0; 95 % CI : 0 .15 , 0 .46 ) str at eg ies al on e, es pe cia lly w he n c ou ple d w ith re wa rd an d m od ell ing in st ud ies (g = 1.0 8; 95 % CI : 0 .50 , 1 .66 ) w ith la rg es t e ffe ct siz e f or un fam ilia r o r d isl ike d v eg et ab les . H et er og en eit y re m ain ed hi gh in bo th th e m ain an d t he su bg ro up an aly se s. Ni eb yls ki et al ., 20 14 (1 51 ) 4.1 Th is sy ste m at ic re vie w of 34 stu die s o f f oo d p ro cu re m en t po lic ies , 1 9 o f w hic h w er e im ple m en te d i n t he sc ho ol se tti ng , fo un d t ha t n ea rly all of th es e d em on str at ed po sit ive ch an ge s i n h ea lth y fo od pu rch as ing pa tte rn s. Th e r ev iew in clu de d 3 4 stu die s. St ud ies w er e inc lud ed if th ey w er e R CT s, pr os pe cti ve an d r et ro sp ec tiv e no n- ra nd om ize d f oo d pr oc ur em en t i nt er ve nt ion s. He alt hy fo od pr oc ur em en t po lic ies su ch as re gu lat ion s, sta nd ar ds an d g uid eli ne s inc re as ing th e a va ila bil ity of he alt hy an d/ or de cre as ing th e a va ila bil ity of un he alt hy fo od s a nd be ve ra ge s s er ve d or so ld in sch oo ls. Al l p op ula tio ns . O ut of 34 stu die s, 19 to ok pl ac e i n a sch oo l s et tin g, wh ile ot he rs to ok pl ac e a t w or ks ite s, ho sp ita ls, co rre cti on al fac ilit ies , g ov er nm en t f ac ilit ies , et c. an d w er e p rim ar ily in th e Un ite d S ta te s, Un ite d K ing do m an d C an ad a. Al l o f t he fo od pr oc ur em en t i nt er ve nt ion s i n sch oo l s et tin gs de m on str at ed in cre as es in he alt hy fo od pu rch as ing pa tte rn s. M an y o f th e s ch oo l in te rv en tio ns th at al so in clu de d an ed uc at ion co m po ne nt w er e e ffe cti ve at inc re as ing th e i nt ak e o f h ea lth y f oo ds an d de cre as ing th e i nt ak e o f f oo ds hi gh in fa t, so diu m an d s ug ar. Tw o s tu die s t ha t a sse sse d h ea lth ou tco m es fo un d a re du cti on in bl oo d p re ssu re an d B M I. I n t he se st ud ies , p ro cu re m en t o f fo od in vo lve d p ro vid ing gr ea te r q ua nt iti es an d low er ing th e p ric e o f h ea lth y f oo ds in ca fet er ias an d v en din g m ac hin es . 140 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ni em eie r, H ek tn er an d E ng er, 20 12 (8 7) 2.2 Th is sy ste m at ic re vie w an d m et a- an aly sis in m ult ipl e se tti ng s s ug ge sts th at int er ve nt ion s t ha t r eq uir e pa re nt pa rti cip at ion m or e eff ec tiv ely re du ce BM I in ch ild re n a nd ad ole sce nt s. Th e r ev iew in clu de d 3 6 RC T s tu die s o f a to ta l o f 4 2 int er ve nt ion s, pu bli sh ed be tw ee n 2 00 5 a nd 20 10 . Nu tri tio n e du ca tio n ( cla ssr oo m lec tu res or in ter ac tiv e wo rks ho ps ), P A e du ca tio n (d isc us sio ns an d e du ca tio na l m ate ria ls) , P A s es sio ns (p lay , sk ills de ve lop m en t), be ha vio ur ed uc ati on or th era py (s es sio ns or in div idu al co un se llin g t o fos ter se lf- co nt ro l a nd se lf- effi ca cy to im pr ov e d iet ar y, PA or se de nt ar y a cti vit y ha bit s) or a co m bin ati on of th es e a cti vit ies . S ev en tee n int er ve nt ion s w ere pr ev en tiv e int er ve nt ion s, wh ile 25 w ere tre atm en t in ter ve nt ion s. Pa ren t p ar tic ipa tio n ( e.g . les so ns , c om ple m en ta ry se ssi on s, ha nd bo ok s, co un se llin g, m ee tin gs ) wa s r eq uir ed in 23 st ud ies , op tio na l in 9 stu die s a nd no t inc lud ed in 10 st ud ies . Ch ild re n a nd ad ole sce nt s ag ed 2– 19 ye ar s ( n= 74 55 ), glo ba lly . T he pr ev en tiv e s tu die s co ns ist ed of > 90 % of to ta l sa m ple , a nd th e m ajo rit y o f th es e s tu die s w er e s ch oo l- ba se d. In te rv en tio ns th at re qu ire d p ar en t p ar tic ipa tio n ha d g re at er su cce ss th an in te rv en tio ns w ith no pa re nt pa rti cip at ion (w eig ht ed av er ag e diff er en ce s i n e ffe ct siz es 0. 30 ; P = 0.0 27 ) an d w ith no or op tio na l p ar en t p ar tic ipa tio n (P = 0.0 16 ). T his su gg es ts th at w eig ht ‐re lat ed he alt h i nt er ve nt ion s t ha t r eq uir e p ar en t pa rti cip at ion m or e e ffe cti ve ly re du ce (o r in cre as e les s) th e B M I o f c hil dr en an d a do les ce nt s. M or eo ve r, s ign ifi ca nt di ffe re nc es ex ist ed am on g lev els of pa re nt pa rti cip at ion (P < 0.0 5) an d int er ve nt ion du ra tio n ( P= 0.0 06 ), an d t he lin ea r c om bin at ion of pa re nt pa rti cip at ion an d int er ve nt ion du ra tio n s ign ifi ca nt ly pr ed ict ed int er ve nt ion eff ec tiv en es s ( P= 0.0 01 ). Ni xo n e t a l., 20 12 (9 9) 2.2 , 3 .3, 4. 11 Th is sy ste m at ic re vie w of 12 ob es ity p re ve nt io n st ud ie s a m on g ch ild re n ag ed 4– 6 y ea rs in th e sc ho ol se tti ng h ig hl ig ht ed th at p ar en ta l i nv ol ve m en t re su lts in a hi gh er ra te of su cc es s. Th e r ev iew in clu de d 1 2 ind ivi du al stu die s p ub lis he d be tw ee n 1 99 8 a nd 20 10 . Th re e w er e n on -ra nd om ize d co nt ro lle d t ria ls an d 9 w er e RC Ts (o f w hic h 5 w er e c lus te r- RC Ts) . Al l in te rv en tio ns w er e a im ed at pr ev en tin g o be sit y b y inc re as ing PA an d h ea lth y ea tin g a nd re du cin g se de nt ar y b eh av iou r inc lud ing ed uc at ion , P A, m od ell ing be ha vio ur by te ac he rs an d p ar en ts an d en vir on m en ta l c ha ng es — w ith di ffe re nt le ve ls of pa re nt al inv olv em en t. Ch ild re n a ge d 4 –6 ye ar s i n pr es ch oo ls an d s ch oo ls in Eu ro pe , N or th Am er ica , A sia an d A us tra lia . Pa re nt al inv olv em en t w as hi gh in 3 ou t o f 4 stu die s r ep or tin g f av ou ra ble ou tco m es in w eig ht sta tu s a nd in al l 5 st ud ies re po rti ng fa vo ur ab le die ta ry be ha vio ur ou tco m es . In te rv en tio ns w ith th e h igh es t s uc ce ss ha d a hi gh le ve l o f p ar en ta l inv olv em en t a nd th e f oll ow ing : 1 ) D ev elo pin g sk ills an d b eh av iou ra l c ap ab ilit y; 2) D ev elo pin g se lf- effi ca cy ; 3 ) E du ca tin g p ar en ts an d c hil dr en (in cl as sro om -b as ed an d/ or pr ac tic al se ssi on s) ab ou t t he be ne fit s o f h ea lth fu l d iet ar y a nd PA be ha vio ur s; an d 4 ) M od ell ing he alt hf ul ea tin g an d P A b y p ar en ts an d/ or te ac he rs. Th e a ut ho rs als o n ot ed ch ild re n’s (a nd pa re nt s') pe rce ive d co m pe te nc e i n m ak ing he alt hy ch an ge s i n t he ir die t a nd PA le ve ls as a ke y e lem en t f or su cce ss. 141ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Nø rn be rg et al ., 20 16 (1 53 ) 4.1 Th is sy ste m at ic re vie w of 12 stu die s f ou nd in co nc lus ive ev ide nc e o n t he eff ec t o f ch oic e a rch ite ctu ra l n ud ge int er ve nt ion s p ro m ot ing ve ge ta ble co ns um pt ion am on g a do les ce nt s; ho we ve r, t he in te rv en tio ns th at se em ed m os t l ike ly to su cce ed in in cre as ing ve ge ta ble co ns um pt ion w er e th os e w he re th e v ar iet y off er ed w as in cre as ed . Th e r ev iew in clu de d 1 2 stu die s o f d iff er en t d es ign (R CT , e xp er im en ta l, q ua si- ex pe rim en ta l, p re - a nd po st- te st de sig n, co ho rt) pu bli sh ed be tw ee n 2 00 1 an d 2 01 3. Ch oic e a rch ite ctu ra l n ud gin g int er ve nt ion s i nfl ue nc ing ch oic e w ith ou t l im iti ng ch oic e s et s o r m ak ing alt er na tiv es co stl ier . T he sp ec ifi c i nt er ve nt ion s t o infl ue nc e a do les ce nt ’s ve ge ta ble in ta ke in clu de d 1) di str ibu tio n o f f re e ve ge ta ble s; 2) m od ifi ca tio ns to se rv ing st yle ; a nd 3) ch an gin g t he ph ys ica l en vir on m en t. Ad ole sce nt s a ge d 1 1– 19 ye ar s in th e s ch oo l s et tin g i n t he Un ite d S ta te s, Ca na da an d De nm ar k. Th e r ev iew fo un d i nc on clu siv e e vid en ce fr om th e 12 st ud ies on th e e ffe ct of ch oic e a rch ite ctu ra l nu dg e i nt er ve nt ion s o n p ro m ot ing ve ge ta ble int ak e a m on g a do les ce nt s i n s ch oo ls. Th e a ut ho rs co nc lud ed th at pr ov idi ng fr ee ve ge ta ble s r es ult ed in m or e f av ou ra ble at tit ud es to wa rd s v eg et ab le co ns um pt ion an d t ha t o ffe rin g a gr ea te r v ar iet y of ve ge ta ble s r es ult ed in an in cre as e i n v eg et ab le co ns um pt ion . Oo ste rh off et al ., 20 16 (8 9) 2.2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 85 RC Ts sh ow ed th at sc ho ol- ba se d lif es ty le int er ve nt ion s m ay re su lt in be ne fic ial ch an ge s i n c hil dr en ’s B M I an d b loo d p re ssu re ; p ar en ta l inv olv em en t a cce nt ua te d th es e b en efi cia l e ffe cts . Th e r ev iew in clu de d 9 1 pa pe rs, pu bli sh ed be tw ee n 19 85 an d 2 01 3, of 85 un iqu e RC Ts in wh ich th e c on tro l gr ou p d id no t r ec eiv e a ny int er ve nt ion be yo nd us ua l cu rri cu lar ac tiv iti es . In te rv en tio ns ai m ed at im pr ov ing di et (e .g. by ch an ge s i n f oo ds pr ov ide d b y sch oo l c an te en s/c afe te ria s), PA (e .g. in cre as ing ph ys ica l ed uc at ion or ac tiv ity du rin g o r a fte r s ch oo l ho ur s, re du cin g s ed en ta ry be ha vio ur ) o r e du ca tio n f or he alt hie r d iet ar y a nd ac tiv ity be ha vio ur s (e .g. th ro ug h p am ph let s, did ac tic m at er ial s a nd /o r lec tu re s). O f t he 85 RC Ts, 53 al so in clu de d a pa re nt al inv olv em en t c om po ne nt (e .g. by ne ws let te rs, in fo rm at ion se ssi on s, via ho m ew or k t as ks fo r c hil dr en ) a nd 19 in clu de d af te r-s ch oo l h ou rs ac tiv iti es . Ch ild re n ag ed 4– 11 ye ar s i n th e p rim ar y s ch oo l s et tin g in 29 d iff er en t c ou nt rie s. M os t s tu di es w er e c on du ct ed in Eu ro pe (3 7 s tu di es ) an d No rth A m er ica (3 3 st ud ie s), w he re as so m e we re co nd uc te d in O ce an ia (7 st ud ie s), A sia (5 st ud ie s), So ut h Am er ica (2 st ud ie s) an d No rth A fri ca (1 st ud ie s). Un iva ria te m ul til ev el an aly se s o f s tu dy ou tco m es sh ow ed si gn ifi ca nt be ne fic ial po ol ed eff ec t s ize s i n B M I ( -0 .0 72 ; 9 5% CI: -0 .1 06 , - 0. 03 8) , S BP (- 0. 18 3; 95 % CI: -0 .2 88 ; -0 .0 78 ) a nd D BP (- 0. 07 1; 95 % CI: -0 .1 85 , 0. 04 4) . U sin g a m ag ni tu de of SD s a ro un d t he BM I m ea n v alu es (o n a ve ra ge 3. 1 k g/ m 2). Th e a ut ho rs no te d t ha t t he re du cti on in B M I tra ns lat ed to ap pr ox im at ely -0 .2 2 k g/ m 2. Fo r bl oo d p re ss ur e, th e l ar ge r e ffe cts w er e s ee n fo r c hi ld re n w ho al re ad y h ad hi gh ba se lin e va les , w he re as fo r B M I t he eff ec ts ap pl ied to al l r eg ar dl es s o f b as eli ne w eig ht st at us . M ul tiv ar iab le m et a- re gr es sio n a na lys es of st ud y po pu lat ion , in te rv en tio ns an d s tu dy qu ali ty ch ar ac te ris tic s a s m od er at or s s ho we d t ha t pa re nt al in vo lve m en t s ig ni fic an tly im pr ov ed th e be ne fic ial eff ec ts on B M I a nd SB P ( P< 0. 05 ). 142 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Pe irs on et al ., 20 15 (4 8) 1.1 , 3 .2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 90 st ud ies of be ha vio ur al ob es ity pr ev en tio n i nt er ve nt ion s, of wh ich 71 w er e s ch oo l-b as ed , sh ow ed sm all re du cti on s in BM I, b ut no in te rv en tio n str at eg y c on sis te nt ly pr od uc ed be ne fit s. Th e r ev iew in clu de d 9 0 stu die s o f R CT s p ub lis he d be tw ee n 1 99 8 a nd 20 13 . T he re vie w wa s d es ign ed as an up da te of th e W at er s e t a l. (2 01 1) Co ch ra ne re vie w an d inc lud ed 28 of th os e s tu die s. Be ha vio ur ch an ge int er ve nt ion s f oc us ing in di et (1 6 s tu die s), PA (2 0 s tu die s), die t a nd PA (3 2 s tu die s) or lif es ty le (2 2 s tu die s). M os t sch oo l-b as ed in te rv en tio ns fo cu se d o n n ut rit ion ed uc at ion /m es sa ge s, alt ho ug h o th er ap pr oa ch es we re al so in clu de d ( e.g . alt er ing fo od en vir on m en t, pr ov idi ng su ga r-f re e be ve ra ge s, co un se llin g b y sch oo l h ea lth nu rse s, pa re nt inv olv em en t). Ch ild re n a nd ad ole sce nt s a ge d 0– 5 y ea rs (2 1 s tu die s), 6– 12 ye ar s ( 53 st ud ies ) a nd 13 –1 8 ye ar s ( 17 st ud ies ), lar ge ly in sch oo l s et tin gs (7 1 s tu die s), m os tly in th e U nit ed St at es an d Eu ro pe , b ut al so in Ca na da , Au str ali a, Br az il, Isr ae l, C hin a, Eg yp t, In dia , M ex ico , N ew Ze ala nd an d T ha ila nd . Th e o ve ra ll m et a- an aly sis of 76 st ud ies in vo lvi ng 56 34 2 c hil dr en in al l s et tin gs sh ow ed sm all bu t s ign ifi ca nt re du cti on s i n B M I o r B M I z -sc or e (S M D: -0 .07 ; 9 5% CI : - 0.1 0, -0 .03 ; 7 6 s tu die s), in BM I ( M D: -0 .09 kg /m 2; 95 % CI : - 0.1 6, -0 .03 ; 56 st ud ies ), an d i n o ve rw eig ht an d o be sit y pr ev ale nc e ( RR : 0 .94 ; 9 5% CI : 0 .89 , 0 .99 ; 30 st ud ies ). In va rio us su bg ro up an aly se s, diff er en ce s i n e ffe ct we re on ly se en fo r d iff er en t se tti ng s ( P= 0.0 4) , w he re on ly th e e du ca tio n se tti ng si gn ifi ca nt ly re du ce d B M I o r B M I z -sc or e (S M D: -0 .09 ; 9 5% CI : - 0.1 7, -0 .03 ; 5 1 s tu die s; 47 97 5 p ar tic ipa nt s). Fo r o th er su bg ro up an aly se s, th e o nl y s ub gr ou ps w ith si gn ifi ca nt re du cti on s in BM I a nd BM I z -sc or e w er e f or in te rv en tio ns of die t a nd ex er cis e c om bin ed (S M D: -0 .10 ; 9 5% CI : -0 .17 , - 0.0 3; 26 st ud ies ; 1 4 9 23 pa rti cip an ts) , a nd fo r t he ag e g ro up s 6 –1 2 y ea rs (S M D: -0 .06 ; 9 5% CI: -0 .10 , - 0.0 1; 42 st ud ies ; 3 6 9 16 pa rti cip an ts) an d 1 3– 18 ye ar s ( SM D: -0 .12 ; 9 5% CI : - 0.2 2, -0 .02 ; 1 7 s tu die s; 12 49 6 p ar tic ipa nt s). Pe ra lta , D ud ley an d C ot to n, 20 16 (1 20 ) 3.1 Th is sco pin g r ev iew of st ud ies of 32 nu tri tio n e du ca tio n re so ur ce s f ou nd th at su ch re so ur ce s a re le ss lik ely to em be d c ro ss- cu rri cu lar an d e xp er ien tia l le ar nin g ap pr oa ch es or co nt ing en t re inf or ce m en t a pp ro ac he s. Th e r ev iew in clu de d s tu die s of 32 nu tri tio n r es ou rce s pu bli sh ed in 49 pa pe rs be tw ee n 1 98 0 a nd 20 14 . Th e t ea ch ing st ra te gie s f or he alt hy ea tin g/ nu tri tio n m os tly in clu de d c ur ric ulu m ap pr oa ch es (3 2 r es ou rce s), cro ss- cu rri cu lar ap pr oa ch es (1 3 r es ou rce s), pa re nt al inv olv em en t ( 15 re so ur ce s), ex pe rie nt ial le ar nin g ap pr oa ch es (7 re so ur ce s); co nt ing en t r ein fo rce m en t ap pr oa ch es (2 re so ur ce s), lit er ar y a bs tra cti on ap pr oa ch es (2 2 r es ou rce s), ga m es -b as ed ap pr oa ch es (1 3 re so ur ce s) an d w eb -b as ed ap pr oa ch es (1 0 r es ou rce s). Pr im ar y s ch oo l a ge ch ild re n, in sch oo l s et tin gs la rg ely in th e Un ite d S ta te s a nd Au str ali a a s we ll a s t he U nit ed Ki ng do m an d N ew Ze ala nd . Th e s co pin g r ev iew fo un d t ha t m os t n ut rit ion ed uc at ion re so ur ce s i nc lud e c ur ric ulu m -b as ed as pe cts , d eli ve re d i n a nd ou tsi de th e c las sro om s, th ro ug h o nl ine , o r w eb -b as ed le ar nin g. Ho we ve r, th e r es ou rce s w er e l es s l ike ly to em be d c ro ss- cu rri cu lar , e xp er ien tia l le ar nin g a pp ro ac he s a nd co nt ing en t r ein fo rce m en t a cti vit ies . 143ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Pr ice et al ., 2 01 7 (1 19 ) 3.1 Th is sy ste m at ic re vie w of 9 stu die s o n n ut rit ion ed uc at ion an d B M I s ug ge ste d th at lo ng -te rm nu tri tio n ed uc at ion de liv er ed in th e sch oo l s et tin g c an pr ov ide ch ild re n w ith to ols to at ta in a he alt hy w eig ht st at us . Th e r ev iew in clu de d 9 st ud ies (R CT , n on -ra nd om ize d co nt ro lle d t ria ls, qu as i- ex pe rim en ta l, s ec on da ry da ta an aly sis ) p ub lis he d b et we en 20 08 an d 2 01 5. Nu tri tio n e du ca tio n f oll ow ing a s ch oo l c ur ric ulu m pa tte rn of te ac he r-l ed le ctu re an d fo llo w- up ac tiv iti es , s uc h a s a r eg ula r c las sro om le sso n, by tr ain ed te ac he rs. Al l th e i nt er ve nt ion s u til ize d a p ro gr am m e p re vio us ly de ve lop ed by a Un ite d St at es go ve rn m en t a nd /o r pr of es sio na l o rg an iza tio n. Ch ild re n i n k ind er ga rte n th ro ug h g ra de 5 (7 st ud ies ) an d g ra de s 6 –8 (2 st ud ies ) in th e s ch oo l s et tin g i n t he Un ite d S ta te s. Fo ur of th e 7 st ud ies on nu tri tio n e du ca tio n de liv er ed to th e y ou ng er ch ild re n d em on str at ed a sta tis tic all y s ign ifi ca nt de cre as e i n t he m ea su re d BM I p er ce nt ile , p rim ar ily a de cre as e i n t he ov er we igh t/o be se ca te go rie s i n t he pa rti cip an ts, wh er ea s 1 of th e 2 st ud ies in th e o lde r c hil dr en de m on str at ed a de cre as e i n t he hi gh es t c las s of ob es ity (> 95 pe rce nt ile ). In bo th gr ou ps , int er ve nt ion s w ith lo ng -te rm (g re at er th an 1 ye ar ) i m ple m en ta tio n s ho we d m or e p ro no un ce d re su lts w ith po sit ive im pa ct on re du cin g ov er we igh t/o be se BM I m ea su re s. Ro bin so n- O’ Br ien , St or y a nd H eim , 20 09 (1 23 ) 3.1 , 3 .3 Th is re vie w su gg es ts th at ga rd en -b as ed nu tri tio n ed uc at ion pr og ra m m es m ay ha ve po te nt ial to pr om ot e inc re as ed F& V i nt ak e am on g y ou th an d i nc re as ed wi llin gn es s t o t as te F& V am on g y ou ng er ch ild re n; ho we ve r, e m pir ica l e vid en ce in th is ar ea is re lat ive ly sc an t. Ele ve n s tu die s w er e i nc lud ed , of w hic h 5 w er e p re - a nd po st- te st wi th in te rv en tio n an d c on tro l o r c om pa ris on gr ou ps , 5 w er e p re - a nd po st- te sts w ith in th e s am e po pu lat ion an d 1 st ud y re po rte d t he m es fr om fo cu s g ro up s. St ud ies w er e pu bli sh ed be tw ee n 1 99 9 an d 2 00 7. Ga rd en -b as ed yo ut h n ut rit ion int er ve nt ion pr og ra m m es wi th or w ith ou t n ut rit ion ed uc at ion co m po ne nt . Ch ild re n a nd ad ole sce nt s ag ed 5– 15 ye ar s i n t he U nit ed St at es . F ive of 11 st ud ies w er e loc at ed on sc ho ol gr ou nd s an d w er e i nt eg ra te d w ith in th e s ch oo l c ur ric ulu m , 3 w er e co nd uc te d a s p ar t o f a n a fte r- sch oo l p ro gr am m e a nd 3 we re co m m un ity -b as ed . Of th e 8 sc ho ol or af te r-s ch oo l b as ed st ud ies , 2 o ut of 3 stu die s m ea su rin g F &V in ta ke re po rte d s ign ifi ca nt in cre as es as re su lt of ex po su re to ga rd en -b as ed nu tri tio n e du ca tio n, 2 o ut of 5 stu die s m ea su rin g w illi ng ne ss to try ve ge ta ble s r ep or te d i m pr ov em en ts, 3 ou t of 3 stu die s r ep or tin g o n w illi ng ne ss to ta ste ve ge ta ble s r ep or te d i m pr ov em en ts, an d 3 ou t of 5 m ea su rin g n ut rit ion kn ow led ge re po rte d im pr ov em en ts. 144 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ru gg ier i a nd Ba ss, 20 15 (1 76 ) 5.1 , 5 .2 Th is co m pr eh en siv e r ev iew of 24 st ud ies fo un d t ha t d es pit e th e c on tro ve rsi es su rro un din g sch oo l-b as ed BM I s cre en ing pr og ra m m es , t he y c an be va lua ble fo r s tu de nt s, pa re nt s an d s ch oo ls as pa rt of sc ho ol- ba se d o be sit y p re ve nt ion pr og ra m m es an d p ro m ot ion of he alt hy lif es ty les in sch oo ls. Th e r ev iew in clu de d 2 4 pa pe rs/ re po rts ev alu at ing m an da to ry BM I s cre en ing pr og ra m m es in th e U nit ed St at es be tw ee n 2 00 5 a nd 20 12 . R ela te d l eg isl at ion an d ot he r l ite ra tu re w er e a lso inc lud ed . M an da to ry sc ho ol- ba se d B M I scr ee nin g a nd su rv eil lan ce pr og ra m m es ou tli ne d i n Un ite d S ta te s s ta te le gis lat ion an d p oli cie s. Sc ho ol- ag ed ch ild re n ( ag e n ot sp ec ifi ed ) i n 7 U S s ta te s i n t he sch oo l s et tin g. Co nfi de nt ial ity is a pr im ar y c on ce rn of BM I scr ee nin g p ro gr am m es , s ha red by pa ren ts, sc ho ol nu rse s a nd sc ho ol ad m ini str ato rs. H ow ev er, m os t pa ren ts su pp or t s uc h p ro gr am m es pr ov ide d res pe ct for st ud en t p riv ac y i s m ain ta ine d a t a ll lev els . R es ou rce s r eq uir ed fo r s ta ff, m ea su rem en t to ols , o bt ain ing pa ren ta l p erm iss ion an d r ep or tin g da ta w ere re du ce d b y i nt ro du cin g b ien nia l ra th er th an an nu al scr ee nin g. BM I s cre en ing an d r efe rra ls led to in cre as ed fo llo w- up ac tio n b y p are nt s i n so m e s tu die s, wh ere as ot he r s tu die s r ep or ted th at on ly a s m all pe rce nt ag e o f p are nt s a ctu all y did so . O th er rep or ted ou tco m es w ere re du ce d us e o f d iet -p ills , e nc ou rag ing a he alt hy br ea kfa st, lim iti ng fa m ily co ns um pt ion of un he alt hy fo od s an d b ev era ge s, de cre as ing fa m ily m ea ls in fas t foo d r es ta ur an ts, lim iti ng un he alt hy fa ts for m ea ls co ok ed at ho m e, lim iti ng ch ild ren ’s s cre en tim e, giv ing m or e t im e f or PA , a nd re du cin g t he us e o f P A as pu nis hm en t. O ne st ud y r ep or ted st ab iliz ati on of BM I o ve r t he ye ars of th e p ro gr am m e. Sa cco et al ., 2 01 7 (1 59 ) 4.2 Th is sy ste m at ic re vie w of m en u l ab ell ing on fo od s a nd be ve ra ge s c ho se n b y c hil dr en or ad ole sce nt s f ou nd th at su ch la be llin g m ea su re s m ay be un de rst oo d a nd us ed by ch ild re n, in clu din g i n s ch oo l ca fet er ias . Th e r ev iew in clu de d 1 1 stu die s p ub lis he d b et we en 20 05 an d 2 01 5 d es cri bin g pr im ar y r es ea rch on m en u lab ell ing an d o f m ult ipl e re se ar ch de sig ns (p ro sp ec tiv e co ho rt de sig n, re pe at cr os s- se cti on al pr e- an d p os t-t es t de sig ns , R CT s, wi th in- pa rti cip an t e xp er im en ta l de sig n) . In te rv en tio ns in clu de d b ot h re al- wo rld m en u l ab ell ing wh ich le d t o a ctu al fo od pu rch as es (6 st ud ies ) a nd in ar tifi cia l s et tin gs w he re m en u l ab ell ing le d t o hy po th et ica l fo od pu rch as es (5 st ud ies ). Ch ild re n w er e or de rin g f or th em se lve s ( 7 stu die s) or pa re nt s w er e ch oo sin g o n b eh alf of th eir ch ild re n ( 5 s tu die s). M en u l ab ell ing in clu de d nu m er ic ca lor ie or nu tri en t co nt en t i nf or m at ion w ith or wi th ou t D RI , in te rp re ta tiv e inf or m at ion to ill us tra te (e .g. he alt hie r c ho ice or sy m bo ls) . Ch ild re n a ge d 6 –1 8 y ea rs an d p ar en ts. O f 1 1 s tu die s, 3 t oo k p lac e i n a m idd le or hig h s ch oo l s et tin g w hil e ot he rs to ok pl ac e i n r ea l-l ife re sta ur an t s et tin gs or hy po th et ica l s itu at ion s. Th e s tu die s o f h yp ot he tic al fo od pu rch as es in ar tifi cia l e nv iro nm en ts su gg es t t ha t m en u lab ell ing m ay be effi ca cio us in re du cin g c alo rie s pu rch as ed fo r o r b y c hil dr en an d a do les ce nt s. T he re al- wo rld st ud ies w er e l es s s up po rti ve , a lth ou gh 2 o f 3 sc ho ol- ba se d s tu die s s ug ge ste d a po te nt ial im pa ct of fa t o r e ne rg y c on te nt of pu rch as e, fo r ex am ple , a s a re su lt of m en u o r c alo rie la be llin g in th e s ch oo l c afe te ria . 145ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Sa lam et al ., 2 01 6 (1 83 ) 5.3 Th is sy ste m at ic re vie w an d m et a- an aly sis of 31 stu die s, 26 of w hic h w er e sch oo l-b as ed , fo un d t ha t sch oo l-b as ed m icr on ut rie nt su pp lem en ta tio n c an sig nifi ca nt ly re du ce an ae m ia pr ev ale nc e i n a do les ce nt s, es pe cia lly in gi rls . Fo r t he re vie w of m icr on ut rie nt su pp lem en ta tio n t o ad ole sce nt s, a t ot al of 31 stu die s w er e i nc lud ed , o f wh ich 16 w er e R CT s, 7 w er e qu as i-e xp er im en ta l a nd 8 we re be fo re -a fte r. S tu die s we re pu bli sh ed be tw ee n 19 96 an d 2 01 2. St ud ies ev alu at ed th e eff ec tiv en es s o f ir on , fo lic ac id, vi ta m in A, vi ta m in D, vi ta m in C, ca lci um , z inc an d M M N su pp lem en ta tio n to ad ole sce nt po pu lat ion s. Th irt ee n s tu die s e va lua te d th e i m pa ct of iro n/ iro n f oli c ac id su pp lem en ta tio n a lon e, 9 s tu die s e va lua te d t he im pa ct of iro n/ iro n f oli c a cid in co m bin at ion w ith ot he r m icr on ut rie nt s, 2 s tu die s ev alu at ed th e i m pa ct of M M N alo ne , 2 st ud ies ev alu at ed zin c s up ple m en ta tio n w hil e 5 s tu die s s up ple m en te d w ith ca lci um an d v ita m in D. Ad ole sce nt gi rls (a ll 3 1 s tu die s) an d b oy s ( 9 s tu die s) pr im ar ily in low - a nd m idd le- inc om e co un tri es . T we nt y- six st ud ies we re co nd uc te d i n t he sch oo l s et tin g a nd 5 we re co m m un ity -b as ed . Fin din gs su gg es t t ha t m icr on ut rie nt su pp lem en ta tio n a m on g a do les ce nt s (p re do m ina nt ly fem ale s) ca n s ign ifi ca nt ly de cre as e a na em ia pr ev ale nc e ( RR : 0 .69 ; 9 5% CI: 0. 62 , 0 .76 ). Su bg ro up an aly sis su gg es te d th at sc ho ol- ba se d d eli ve ry si gn ifi ca nt ly re du ce d an ae m ia (R R: 0. 67 ; 9 5% CI : 0 .60 , 0 .74 ) w hil e ev ide nc e f ro m co m m un ity -b as ed de liv er y wa s u nd er po we re d. Sc ho ol- ba se d d eli ve ry of iro n/ iro n f oli c a cid su pp lem en ta tio n a lon e or in co m bin at ion w ith ot he r m icr on ut rie nt su pp lem en ta tio n w as al so as so cia te d w ith im pr ov ed se ru m ha em og lob in (M D: 1. 94 g/ dl; 95 % CI : 1 .48 , 2 .41 ), fer rit in (M D: 3. 80 m cg /L ; 95 % CI : 2 .00 , 5 .59 ) a nd iro n ( M D: 6. 97 m m ol/ L; 95 % CI : 0 .19 , 1 3.7 6) . Z inc su pp lem en ta tio n – als o d eli ve re d t hr ou gh sc ho ols - led to im pr ov ed se ru m zi nc co nc en tra tio ns (M D: 0. 96 m cg /d l; 95 % CI : 0 .81 , 1 .12 ) w hil e c alc ium an d v ita m in D su pp lem en ta tio n d id no t h av e a cl ea r im pa ct on vi ta m in D lev els an d p ar at hy ro id ho rm on e. Ge nd er -sp ec ifi c s ub gr ou p a na lys is su gg es ts sig nifi ca nt im pr ov em en ts in bo th ge nd er s; ho we ve r, m os t o f t he st ud ies w er e c on du cte d o n ad ole sce nt gi rls . 146 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Sa ra f e t a l., 20 12 (1 03 ) 2.2 Th is re vie w fo un d t ha t sch oo l-b as ed in te rv en tio ns ha d m os tly po sit ive eff ec ts on re du cin g r isk fa cto rs of N CD s a nd th at po sit ive re su lts w er e m or e l ike ly wh en in te rv en tio ns in clu de d fam ilie s a nd co m m un iti es ra th er th an sc ho ols al on e. Th is re vie w inc lud ed 37 stu die s, 30 of w hic h w er e RC Ts an d t he re m ain ing w er e no n- ra nd om ize d t ria ls. NC D ris k f ac to r p re ve nt ion int er ve nt ion s i nc lud ing PA , nu tri tio n a nd ot he r N CD ris k fac to rs. O f t he 37 st ud ies , 8 fo cu se d o n n ut rit ion an d 9 fo cu se d n ut rit ion an d P A o r ot he r l ife sty le co m po ne nt , ab ou t h alf of th es e h ad a c om m un ity /p ar en t inv olv em en t c om po ne nt . Nu tri tio n i nt er ve nt ion ac tiv iti es in clu de d n ut rit ion ed uc at ion , P A, he alt h a nd lif e- sk ills ed uc at ion , c ha ng es in sch oo l fo od en vir on m en t an d p ar en ta l/f am ily co m po ne nt s. Ch ild re n a nd ad ole sce nt s a ge d 6– 18 ye ar s i n a sc ho ol- ba se d se tti ng in th e U nit ed St at es , Eu ro pe , A us tra lia , C hin a a nd In dia . Ov er all , 8 0% of st ud ies re po rte d a t l ea st so m e ev ide nc e o f a po sit ive in te rv en tio n e ffe ct on va rio us N CD ris k f ac to rs inc lud ing di et an d P A ( e.g . n ut rit ion kn ow led ge , s kin fo ld th ick ne ss, SS B c on su m pt ion , B M I, o ve rw eig ht pr ev ale nc e, PA le ve ls) . T he in te rv en tio ns pr om ot ing nu tri tio n w er e m or e s uc ce ssf ul if k no wl ed ge w as th e o ut co m e i nd ica to r a s co m pa re d w ith co ns um pt ion of F& V. A g re at er pr op or tio n o f s tu die s t ha t i nv olv ed fa m ilie s o r co m m un iti es sh ow ed po sit ive re su lts co m pa re d to in te rv en tio ns th at to ok pl ac e i n s ch oo l s et tin gs alo ne . Sc hr oe de r, T ra ve rs, an d S m ald on e, 20 16 (8 8) 2.2 , 5 .3 Th is sy ste m at ic re vie w an d m et a- an aly sis of 11 sc ho ol- ba se d o be sit y i nt er ve nt ion s inv olv ing sc ho ol he alt h nu rse s d em on str at ed sm all bu t s ign ifi ca nt eff ec ts on BM I a nd su gg es ts th at nu rse s c an pl ay a ke y r ole in im ple m en tin g e ffe cti ve sch oo l-b as ed ob es ity int er ve nt ion s. Th e r ev iew in clu de d 1 1 stu die s p ub lis he d b et we en 20 08 an d 2 01 5 ( 4 R CT , 7 qu as i-e xp er im en ta l), of wh ich 8 we re in clu de d i n t he m et a- an aly sis . Sc ho ol- ba se d i nt er ve nt ion s inv olv ing nu rse s i n a ro le be yo nd an th ro po m et ric m ea su re m en t ( e.g . s tu de nt ed uc at ion an d c ou ns ell ing , pa re nt pa rti cip at ion in co un se llin g, co ns ult at ion s o r ed uc at ion , s ta ff ed uc at ion an d P A) . Sc ho ol- ag ed ch ild re n a nd ad ole sce nt s i n t he U nit ed St at es , E ur op e a nd A sia in th e sch oo l s et tin g. Th e fi nd ing s s ho we d t ha t s ch oo l-b as ed int er ve nt ion s t ha t i nv olv e n ur se s l ed to sm all bu t s ign ifi ca nt de cre as es in BM I, B M I z -sc or e an d B M I p er ce nt ile . T he po ole d a na lys es of 8 stu die s d em on str at ed st at ist ica lly si gn ifi ca nt de cre as es in BM I ( -0 .48 ; 9 5% CI : - 0.8 4, -0 .12 ), BM I z -sc or e ( -0 .10 ; 9 5% CI : - 0.1 5, -0 .05 ) a nd BM I p er ce nt ile (- 0.4 1; 95 % CI : - 0.6 0, -0 .21 ). In su bg ro up an aly se s o f 6 st ud ies , 3 st ud ies wi th pa re nt in vo lve m en t ( -0 .72 ; 9 5% CI : - 1.4 2, -0 .01 ) r es ult ed in la rg er, th ou gh no t s ign ifi ca nt , de cre as es in BM I t ha n t he 3 stu die s w ith ou t t his co m po ne nt (- 0.0 6; 95 % CI : - 0.1 1, -0 .01 ). T he se re su lts ar e s im ila r t o o th er m et a- an aly se s o f sch oo l-b as ed ch ild ob es ity in te rv en tio ns an d t he au th or s s ug ge st th at nu rse s c an pl ay a ke y r ole in im ple m en tin g e ffe cti ve an d s us ta ina ble ch ild ob es ity in te rv en tio ns in sc ho ols . 147ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Sil dé n, 20 18 (4 2) 1.1 , 4 .1 Th is sy ste m at ic re vie w of 26 stu die s o n c om pe tit ive fo od s an d b ev er ag e i n s ch oo ls in th e U nit ed St at es , fo un d s uc h pr od uc ts to be pe rv as ive in th e s ch oo ls an d t ha t p oli cie s to re str ict th es e p ro du cts co uld im pr ov e c alo rie in ta ke an d B M I. Th e r ev iew in clu de d 2 6 stu die s o f v ar iou s d es ign s (cr os s-s ec tio na l d es ign s, qu as i-e xp er im en ta l, ex pe rim en ta l) pu bli sh ed be tw ee n 2 00 6 a nd 20 17 . Tw en ty st ud ies ad dr es se d t he av ail ab ilit y o f c om pe tit ive fo od s a nd be ve ra ge s i n sch oo ls, an d 1 4 s tu die s sp ec ifi ca lly an aly se d p oli cie s im ple m en te d i n s ch oo ls. Ch ild re n a ge d 1 3– 18 ye ar s i n th e s ch oo l s et tin g i n t he U nit ed St at es . R es ult s o f s tu die s w ith ot he r a ge gr ou ps (e .g. ag es 6– 12 ) w er e e xt ra po lat ed fo r inc lus ion in th e r ev iew . Th e r ev iew fo un d t ha t c om pe tit ive fo od s a nd be ve ra ge s a re pe rv as ive in sc ho ols , b ut ev ide nc e wa s l ac kin g t ha t s uc h p ro du cts ca us e o ve rw eig ht or ob es ity in ch ild re n. O n t he ot he r h an d, th e au th or no te d t ha t t he re w as a po sit ive as so cia tio n be tw ee n c om pe tit ive fo od an d b ev er ag es po lic ies an d w eig ht -re lat ed an d/ or di et ar y o ut co m es , a nd pr ov ide ex am ple s o f s ev er al ca se s w he re ca lor ic int ak e a nd ov er we igh t/o be sit y m ea su re s h ad im pr ov ed . Sil ve ira et al ., 20 11 (1 01 ) 2.2 , 3 .1 Th is sy ste m at ic re vie w fo un d ev ide nc e o f p os iti ve eff ec ts on an th ro po m et ry an d o f inc re as e i n F &V co ns um pt ion fro m in te rv en tio ns th at we re lo ng er th an 1 ye ar, int ro du ce d i nt o t he re gu lar ac tiv iti es of th e sch oo l, i nc lud ed pa re nt al inv olv em en t, int ro du ce d nu tri tio n e du ca tio n i nt o th e r eg ula r c ur ric ulu m , a nd pr ov ide d F &V by sc ho ol fo od se rv ice s. Th e r ev iew in clu de d 2 4 R CT s pu bli sh ed be tw ee n 1 99 8 an d 2 01 0. M os t s tu die s i nc lud ed sc ho ol- ba se d n ut rit ion ed uc at ion fo r th e p re ve nt ion or re du cti on of ov er we igh t a nd pr om ot ion F& V c on su m pt ion , w ith or wi th ou t c om ple m en ta ry int er ve nt ion s a dd re ssi ng th e s ch oo l e nv iro nm en t. Co m po ne nt s i nc lud ed cla ssr oo m ac tiv iti es , pa re nt al inv olv em en t, sch oo l nu tri tio n p oli cy , s ch oo l fo od se rv ice , e du ca tio na l ga m es , s oc ial m ar ke tin g an d e nv iro nm en ta l c ha ng es , ind ivi du al co un se llin g a nd int er ne t u se . Ch ild re n a nd ad ole sce nt s ag ed 5– 18 ye ar s i n s ch oo l a nd cla ssr oo m se tti ng s, m ain ly in Eu ro pe an d N or th Am er ica , b ut als o i n B ra zil , C hin a a nd N ew Ze ala nd . Th er e w as ev ide nc e o f p os iti ve eff ec ts on an th ro po m et ry (B M I, B M I z -sc or e) an d o f in cre as e in F& V c on su m pt ion . W hil e o nl y 1 ou t o f 9 st ud ies m ea su rin g B M I d em on str at ed im pr ov em en ts in av er ag e B M I, m os t o f t he in te rv en tio ns w ith a du ra tio n o f 1 –3 ye ar s d em on str at ed re du cti on s in th e p re va len ce of ov er we igh t a nd ob es ity w ith OR ra ng ing fr om 0. 61 –0 .69 . C ha ra cte ris tic s o f t he int er ve nt ion s t ha t d em on str at ed eff ec tiv en es s we re : d ur at ion > 1 y ea r, i nt ro du cti on in to th e r eg ula r a cti vit ies of th e s ch oo l, p ar en ta l inv olv em en t, int ro du cti on of nu tri tio n e du ca tio n int o t he re gu lar cu rri cu lum , a nd pr ov isi on of F& V by sc ho ol fo od se rv ice s. 148 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Sil ve ira et al ., 20 13 (1 13 ) 3.1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 8 stu die s fo un d e vid en ce th at sc ho ol- ba se d n ut rit ion ed uc at ion int er ve nt ion s a re eff ec tiv e i n re du cin g t he BM I o f c hil dr en an d a do les ce nt s, es pe cia lly int er ve nt ion s o f lo ng er du ra tio n. Th e r ev iew in clu de d 8 st ud ies of ra nd om ize d a nd no n- ra nd om ize d c on tro lle d t ria ls an d i nt er ru pt ed ti m e s er ies . Sc ho ol- ba se d n ut rit ion ed uc at ion in te rv en tio ns (e .g. ed uc at ion ga m es , c las sro om ac tiv iti es ) c om bin ed w ith ot he r c om po ne nt s ( e.g . sch oo l n ut rit ion po lic y, pa re nt al inv olv em en t, so cia l m ar ke tin g o r e nv iro nm en ta l ch an ge s). Ch ild re n a nd ad ole sce nt s ag ed 5– 18 ye ar s ( n= 87 22 ) in pr im ar y a nd se co nd ar y sch oo ls, la rg ely in Eu ro pe , b ut als o i n t he U nit ed St at es an d Br az il. In di vid ua lly , o nl y 2 st ud ies de m on str at ed a s ig ni fic an t r ed uc tio n i n B M I in ch ild re n an d a do les ce nt s i n t he in te rv en tio n g ro up s co m pa re d w ith th e c on tro l g ro up s. Th e p oo led re su lts fr om 8 stu di es ga ve an av er ag e e ffe ct of -0 .3 3 k g/ m 2 ( 95 % CI : - 0. 55 , - 0. 11 ; P = 0. 00 3) on BM I, w ith 84 % of th is eff ec t e xp lai ne d b y t he hi gh es t q ua lit y s tu di es . T he la rg es t e ffe ct on BM I w as ob se rv ed am on g t he 4 stu di es w ith a du ra tio n g re at er th an on e y ea r ( -0 .4 8 k g/ m 2; 95 % CI : - 0. 76 , - 0. 19 , P = 0. 00 1) . So bo l‐G old be rg , Ra bin ow itz an d Gr os s, 20 13 (5 2) 1.1 , 2 .2 Th is sy ste m at ic re vie w an d m et a- an aly sis fo un d co nv inc ing ev ide nc e t ha t sch oo l-b as ed ob es ity pr ev en tio n i nt er ve nt ion s ar e a t l ea st m ild ly eff ec tiv e in re du cin g B M I, w ith th e lar ge st eff ec ts se en in co m pr eh en siv e p ro gr am m es lon ge r t ha n 1 ye ar w ith pa re nt al su pp or t. Th e r ev iew in clu de d 3 2 R CT stu die s p ub lis he d b et we en 20 06 an d 2 01 2. Sc ho ol- ba se d o be sit y pr ev en tio n p ro gr am m es de sig ne d t o r ed uc e B M I in he alt hy ch ild re n t hr ou gh ch an gin g e at ing ha bit s, PA an d s ed en ta ry be ha vio ur. Of th e 3 2 s tu die s, 29 w er e inf or m at ive (e .g. in fo rm at ion on pr op er nu tri tio n a nd PA ), 28 be ha vio ur al (e .g. da ily m on ito rin g a nd re gis te rin g o f ea tin g p at te rn s a nd PA ), 26 en vir on m en ta l (e .g. m ak ing wa te r c oo ler s a cce ssi ble an d o rg an ize d s po rts ), 19 co gn iti ve (e .g. st ra te gie s fo r a tti tu de an d p er ce pt ion ch an ge ), an d 1 8 h ad pa re nt al su pp or t ( e.g . p ar en t c hil d ev en ts on he alt hy lif es ty le) . Th irt y o f t he se w er e m ult ico m po ne nt . Ch ild re n a nd ad ole sce nt s ag ed 5– 18 ye ar s ( n= 52 10 9) co nd uc te d i n s ch oo l s et tin gs m os tly in th e U nit ed St at es an d Eu ro pe , b ut al so in Au str ali a, Br az il a nd Ch ina . Sc ho ol- ba se d o be sit y p re ve nt ion pr og ra m m es re su lte d i n s ign ifi ca nt re du cti on s i n B M I in al l ag e g ro up s ( SM D: -0 .07 6; 95 % CI : - 0.1 23 , - 0.0 28 ; 32 RC Ts) an d i n c hil dr en al on e ( SM D: -0 .10 4; 95 % CI : - 0.1 95 , - 0.0 1; 18 RC Ts) , b ut ga ve a no n- sig nifi ca nt re du cti on of BM I in ad ole sce nt s (1 0 R CT s). M et a- re gr es sio n s ho we d a si gn ifi ca nt lin ea r h ier ar ch y o f s tu die s b as ed on th eir de sig n, wi th th e l ar ge st eff ec ts fo r c om pr eh en siv e pr og ra m m es w ith pa re nt al su pp or t ( SM D: -0 .15 1; 95 % CI : - 0.3 34 , - 0.0 31 ; 6 RC Ts) . C om pr eh en siv e pr og ra m m es in clu de d i nf or m at ion on nu tri tio n an d P A, at tit ud ina l c ha ng es , m on ito rin g e at ing an d P A a nd en vir on m en ta l m od ifi ca tio ns , a nd / or ai m ed to in cre as e P A, de cre as e s ed en ta ry ac tiv ity , im pr ov e h ea lth y e at ing an d d ec re as e un he alt hy ea tin g. In pe rfo rm ing th is re vie w, th e au th or s u pd at ed a pr ev iou s r ev iew by Ka m at h e t al. (2 00 8) an d n ot ed a re m ar ka ble im pr ov em en t. Th ey su gg es t t ha t u nl ike ea rli er st ud ies , m or e re ce nt st ud ies sh ow ed co nv inc ing ev ide nc e t ha t sch oo l-b as ed pr ev en tio n i nt er ve nt ion s a re at lea st m ild ly eff ec tiv e i n r ed uc ing BM I in ch ild re n, po ssi bly be ca us e o f b et te r p ro gr am m e d es ign an d t ha t n ew er st ud ies te nd to be lo ng er, m or e co m pr eh en siv e a nd in clu de pa re nt al su pp or t. 149ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W St ev en s, 20 10 (8 2) 1.1 Th is lit er at ur e r ev iew of m ult ico m po ne nt ob es ity pr ev en tio n p ro gr am m es in et hn ic- m ino rit y m idd le sch oo l a ge ad ole sce nt s f ou nd im pr ov ed di et ar y c ho ice s a nd inc re as ed F& V c on su m pt ion . Th e r ev iew in clu de d 8 stu die s, of w hic h 4 w er e RC Ts, 2 qu as i-e xp er im en ta l, 1 l on git ud ina l c on tro lle d ev alu at ion st ud y a nd 1 pr e- an d p os t-t es t d es ign . S tu die s we re pu bli sh ed be tw ee n 19 97 an d 2 00 7. M ult ico m po ne nt int er ve nt ion s f or ob es ity pr ev en tio n ( i.e . b ot h d iet an d PA ). Al l in clu de d n ut rit ion ed uc at ion in ad dit ion to ot he r in te rv en tio ns su ch as pa re nt al inv olv em en t, fo od se rv ice . Et hn ic m ino rit y m idd le sch oo l ag e a do les ce nt s ( Hi sp an ic, Af ric an Am er ica n, an d N at ive Am er ica n) ag ed 10 –1 4 y ea rs in th e U nit ed St at es , A us tra lia , Be lgi um an d C hil e. Six of 8 stu die s t oo k p lac e i n s ch oo ls, wh ile 1 stu dy to ok pl ac e i n cli nic s a nd 1 stu dy to ok pl ac e in a c om m un ity ce nt re. Th e r ev iew di d n ot id en tif y a ny eff ec tiv e int er ve nt ion s s pe cifi c t o e th nic m ino rit y c hil dr en bu t r es ult s a cro ss stu die s s ho we d i m pr ov em en ts in die ta ry ou tco m es (i .e. de cre as ing co ns um pt ion of hi gh fa t a nd ot he r h igh ca lor ic fo od s a s w ell as so ft dr ink s a nd in cre as ing co ns um pt ion of F& V) . US DA , 2 01 2 ( 13 0) 3.1 Th is sy ste m at ic re vie w of 24 stu die s o f n ut rit ion ed uc at ion de liv er ed vi a d igi ta l m ed ia an d/ or te ch no log y, 15 of wh ich w er e s ch oo l-b as ed , fo un d m od er at e e vid en ce th at su ch in te rv en tio ns m ay im pr ov e d iet ar y b eh av iou r. Th e r ev iew w as co nd uc te d a s pa rt of a se rie s o f s ys te m at ic re vie ws on va rio us as pe cts of nu tri tio n e du ca tio n i n co un tri es w ith a hig h o r v er y hig h h um an de ve lop m en t ind ex . Th e r ev iew in clu de d 2 4 stu die s, 19 RC Ts an d 5 no n- ra nd om ize d c on tro lle d t ria ls, pu bli sh ed be tw ee n 1 99 9 a nd 20 10 . T we lve st ud ies re ce ive d a p os iti ve qu ali ty ra tin g ( 12 RC Ts) , a nd 12 st ud ies re ce ive d a n eu tra l q ua lit y r at ing (7 RC Ts, 5 no n- RC Ts) . Co m pu te r- or in te rn et - ba se d n ut rit ion ed uc at ion int er ve nt ion s ( e.g . m ult im ed ia co m pu te r ga m es , in te ra cti ve w eb sit es , ed uc at ion al vid eo s, co m pu te r sto ry bo ok s, ad ve rg am es ) aim ing to ch an ge di et ar y be ha vio ur (e .g. in cre as e int ak e o f F &V , 1 00 % ju ice , fib re, w at er, da iry pr od uc ts, br ea kfa st, or de cre as e i nt ak e of fa t, ca lor ies , s of t d rin ks , fru it dr ink s ( no t 1 00 % ), so diu m , s na ck s, sw ee ts an d su ga rs) . Ch ild re n a ge d 5 –1 6 y ea rs in th e U nit ed St at es an d Eu ro pe ; 1 5 o f 2 4 s tu die s we re co nd uc te d i n t he sc ho ol se tti ng . M od er at e e vid en ce w as fo un d t ha t n ut rit ion ed uc at ion de liv er ed vi a d ig ita l m ed ia/ te ch no lo gy (c om pu te r- an d I nt er ne t-b as ed pr og ra m m es ) m ay be eff ec tiv e f or im pr ov in g di et ar y i nt ak e- re lat ed be ha vio ur s a m on g ch ild re n a nd ad ol es ce nt s; 21 of 24 st ud ies fo un d sig ni fic an t e ffe cts on di et ar y i nt ak e- re lat ed be ha vio ur s i n a t l ea st on e p op ul at ion su bg ro up , an d i n 1 5 o f t he se th e i nt er ve nt ion w as m or e eff ec tiv e t ha n t he co nt ro l ( no -in te rv en tio n o r ot he r d eli ve ry m et ho d) . N o p at te rn em er ge d re lat ed to th e d eli ve ry se tti ng . T he au th or s co nc lu de d t ha t d ig ita l m ed ia/ te ch no lo gy ca n b e de liv er ed to a wi de ra ng e o f c hi ld re n i n a va rie ty of se tti ng s w ith hi gh fi de lit y, wi th hi gh er lik eli ho od of su cc es s f ro m ta rg et in g s pe cifi c be ha vio ur s a nd pr ov id in g f re qu en t d os es ov er lo ng er pe rio ds of ti m e. 150 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W US DA , 2 01 2 ( 96 ) 2.2 , 3 .1 Th is sy ste m at ic re vie w of 10 in te rv en tio ns of pa re nt al in vo lve m en t i n n ut rit ion ed uc at ion , o f w hi ch 7 we re sc ho ol -b as ed , f ou nd lim ite d an d i nc on sis te nt ev id en ce th at in vo lvi ng pa re nt s in nu tri tio n e du ca tio n im pr ov es di et ar y b eh av iou r. Th e r ev iew w as co nd uc te d a s pa rt of a se rie s o f s ys te m at ic re vie ws on va rio us as pe cts of nu tri tio n e du ca tio n i n co un tri es w ith a hi gh or ve ry hi gh hu m an de ve lo pm en t in de x. Th is re vie w inc lud ed 10 RC Ts pu bli sh ed be tw ee n 1 99 5 a nd 20 09 . S ev en RC Ts re ce ive d a po sit ive qu ali ty ra tin g, an d 3 r ec eiv ed a ne ut ra l q ua lit y ra tin g. St ra te gie s t o i nv olv e p ar en ts in nu tri tio n e du ca tio n int er ve nt ion : d ire ct str at eg ies (e .g. fa ce -to -fa ce se ssi on s), ind ire ct (e .g. ne ws let te rs or ot he r e du ca tio na l m at er ial s se nt ho m e) , o r a co m bin at ion of di re ct an d i nd ire ct m et ho ds ). Ch ild re n a ge d 9 –1 8 y ea rs in th e U nit ed St at es , E ur op e a nd Au str ali a; 7 o f t he 10 st ud ies we re sc ho ol- ba se d. Th e a ut ho rs co nc lud ed th at lim ite d a nd inc on sis te nt ev ide nc e w as av ail ab le to as se ss th e e ffe cts of pa re nt al inv olv em en t i n n ut rit ion ed uc at ion on ch ild re n’s di et ar y b eh av iou r. O f th e 1 0 s tu die s, 6 f ou nd an ad de d b en efi t f ro m pa re nt al inv olv em en t i m pr ov em en ts in at lea st on e s ub gr ou p, of w hic h 3 st ud ies fo un d th is eff ec t i n g irl s b ut no t i n b oy s, 2 f ou nd im pr ov em en ts in bo th in te rv en tio n a nd co nt ro l gr ou p, an d 2 di d n ot fin d a ny eff ec t. On e o f t he 2 stu die s m ea su rin g t he eff ec t o f in cre as ing do se of pa re nt al inv olv em en t f ou nd a sig nifi ca nt do se - re sp on se re lat ion sh ip. St ud ies co nd uc te d o ut sid e of th e U nit ed St at es w er e m or e o fte n e ffe cti ve . US DA , 2 01 2 ( 13 2) 3.1 , 5 .3 Th is sy ste m at ic re vie w of th e e ffe ct of di ffe re nt ty pe s o f n ut rit ion ed uc at or s on ly id en tifi ed 1 stu dy an d c on clu de d t ha t t he re is in su ffi cie nt ev id en ce to de te rm in e w hi ch w ou ld be m os t e ffe cti ve in ch an gi ng ch ild re n’s di et ar y i nt ak e- re lat ed be ha vio ur s. Th e re vie w wa s c on du cte d a s pa rt of a se rie s o f s ys te m at ic re vie ws on va rio us as pe cts of nu tri tio n e du ca tio n i n co un tri es w ith a hi gh or ve ry hi gh hu m an de ve lo pm en t in de x. Th e r ev iew in clu de d 1 RC T pu bli sh ed in 20 07 . Nu tri tio n e du ca tio n int er ve nt ion de liv er ed by tw o diff er en t t yp es of ed uc at or s. Ch ild re n a ge d 1 0 y ea rs in pr im ar y s ch oo l in It aly . It wa s n ot po ssi ble to dr aw an y c on clu sio ns be ca us e t he lit er at ur e s ea rch id en tifi ed on ly 1 s tu dy . M or eo ve r, t he re su lts fr om th is stu dy we re m ixe d, re su lti ng fo r t he m os t p ar t i n n o sig nifi ca nt di ffe re nc es in di et ar y i nt ak e b et we en th e i nt er ve nt ion gr ou ps . H ow ev er, in cre as ed F& V i nt ak e w as se en in th e t ea ch er gr ou p w hil e de cre as ed in ta ke w as se en in th e n ut rit ion ist gr ou p. Fu rth er m or e, fo r a ll f oo d c at eg or ies , ch an ge s i n d iet ar y i nt ak e w er e s ign ifi ca nt ly gr ea te r in th e t ea ch er gr ou p c om pa re d t o t he nu tri tio nis t g ro up (P < 0.0 00 1) . T he au th or s n ot ed th at th e s tu dy ha d s ev er al lim ita tio ns in te rm s o f diff er en ce s i n t im ing , d os e a nd ed uc at ion fo rm at be tw ee n t he gr ou ps . 151ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W US DA , 2 01 2 ( 75 ) 1.1 , 3 .1 Th is sy ste m at ic re vie w of 5 s ch oo l-b as ed st ud ies fo un d c on sis te nt ev ide nc e th at co m bin ing nu tri tio n ed uc at ion w ith ch an ge s t o th e s ch oo l fo od en vir on m en t is m or e e ffe cti ve fo r im pr ov ing ch ild re n’s an d ad ole sce nt s’ d iet ar y i nt ak e th an m ak ing ch an ge s t o t he fo od en vir on m en t a lon e. Th e re vie w wa s c on du cte d a s pa rt of a se rie s o f s ys te m at ic re vie ws on va rio us as pe cts of nu tri tio n e du ca tio n i n co un tri es w ith a hig h o r v er y hig h h um an de ve lop m en t ind ex . Th is re vie w inc lud ed 5 stu die s ( 3 R CT s a nd 2 no n- ra nd om ize d c on tro lle d t ria ls) pu bli sh ed be tw ee n 2 00 2 an d 2 00 9. In te rv en tio ns in clu de d co m bin at ion of nu tri tio n ed uc at ion an d c ha ng es to th e sch oo l fo od en vir on m en t ( e.g . fre e b re ak fas t p ro vid ed at sch oo l, i nc re as ed av ail ab ilit y an d p ro m ot ion of F& V i n th e s ch oo l c afe te ria an d ve nd ing m ac hin es , fr ee F& V sn ac k, an d i nc re as ed ch oic e, m ar ke tin g, an d i m pr ov em en t in th e p re pa ra tio n a nd ap pe ar an ce of fis h s er ve d in th e s ch oo l c afe te ria ), co m pa re d t o e ith er on e o f th es e a lon e. Ch ild re n w ith m ea n a ge 12 –1 5 ye ar s i n t he sc ho ol se tti ng in No rth Am er ica an d E ur op e. Ov er all , a ll 5 st ud ies sh ow ed th at co m bin ing ch an ge s t o t he sc ho ol fo od en vir on m en t w ith nu tri tio n e du ca tio n w as eff ec tiv e f or im pr ov ing ch ild re n a nd ad ole sce nt ’s d iet ar y i nt ak e ( e.g . F &V int ak e, fis h c on su m pt ion ), an d s om e e vid en ce su gg es te d t ha t t he co m bin at ion le d t o g re at er im pr ov em en t t ha n e ith er in te rv en tio n a lon e. In 2 stu die s t he co m bin ed ap pr oa ch w as m or e eff ec tiv e, in 1 s tu dy it w as m or e e ffe cti ve in bo ys on ly an d i n 2 st ud ies bo th ap pr oa ch es w er e eff ec tiv e. US DA , 2 01 2 ( 12 1) 3.1 Th is sy ste m at ic re vie w of 14 st ud ies , 1 0 o f w hic h we re sc ho ol- ba se d, fo un d inc on sis te nt ev ide nc e o f m ult i v s. sin gle co m po ne nt nu tri tio n e du ca tio n int er ve nt ion s i n i m pr ov ing ch ild an d a do les ce nt di et ar y be ha vio ur, an d l im ite d ev ide nc e t ha t n ut rit ion ed uc at ion co m bin ed w ith ha nd s-o n e du ca tio na l co m po ne nt m ay be pa rti cu lar ly eff ec tiv e. Th e re vie w wa s c on du cte d a s pa rt of a se rie s o f s ys te m at ic re vie ws on va rio us as pe cts of nu tri tio n e du ca tio n i n co un tri es w ith a hig h o r v er y hig h h um an de ve lop m en t ind ex . Th e r ev iew in clu de d 1 4 stu die s ( 10 RC Ts an d 4 no n- ra nd om ize d c on tro lle d t ria ls) pu bli sh ed be tw ee n 1 99 6 an d 2 01 0. Nu tri tio n e du ca tio n a lon e as a sin gle co m po ne nt (e .g. us ua l c las sro om ed uc at ion or co un se llin g s es sio ns ) o r co m bin ed w ith su pp or tin g co m po ne nt s ( e.g . p ar en ta l inv olv em en t, we b- ba se d ac tiv ity , g ar de nin g o r t ak e ho m e m at er ial s). Ch ild re n w ith m ea n a ge 7– 15 ye ar s l ar ge ly in th e U nit ed St at es bu t a lso Ca na da an d Eu ro pe ; 1 0 o f 1 4 s tu die s w er e sch oo l-b as ed . Th er e w as no ag re em en t o n w he th er m ult ico m po ne nt nu tri tio n e du ca tio n pr og ra m m es in ch an gin g d iet ar y p ra cti ce s t ha n sin gle co m po ne nt pr og ra m m es an d s ev er al of th e s tu die s h ad m ixe d r es ult s. In to ta l, 6 stu die s f ou nd m ult ico m po ne nt in te rv en tio ns to be m or e e ffe cti ve th an si ng le co m po ne nt int er ve nt ion s f or at le as t o ne di et ar y o ut co m e, 3 s tu die s f ou nd si ng le- co m po ne nt nu tri tio n ed uc at ion in te rv en tio ns to be m or e e ffe cti ve th an m ult ico m po ne nt in te rv en tio ns fo r a t l ea st on e d iet ar y o ut co m e, an d 8 di d n ot fin d a ny diff er en ce be tw ee n t he m . T he au th or s n ot ed th at th e 3 st ud ies th at co ns ist en tly fo un d b et te r eff ec t o f m ult ico m po ne nt in te rv en tio ns w er e a lso th e o nl y o ne s t o c om bin e n ut rit ion ed uc at ion wi th a “h an ds -o n” co m po ne nt (e .g. co ok ing or ga rd en ing ). 152 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W US DA , 2 01 2 ( 75 ) 1.1 , 3 .1 Th is sy ste m at ic re vie w of 5 s ch oo l-b as ed st ud ies fo un d c on sis te nt ev ide nc e th at co m bin ing nu tri tio n ed uc at ion w ith ch an ge s t o th e s ch oo l fo od en vir on m en t is m or e e ffe cti ve fo r im pr ov ing ch ild re n’s an d ad ole sce nt s’ d iet ar y i nt ak e th an m ak ing ch an ge s t o t he fo od en vir on m en t a lon e. Th e re vie w wa s c on du cte d a s pa rt of a se rie s o f s ys te m at ic re vie ws on va rio us as pe cts of nu tri tio n e du ca tio n i n co un tri es w ith a hig h o r v er y hig h h um an de ve lop m en t ind ex . Th is re vie w inc lud ed 5 stu die s ( 3 R CT s a nd 2 no n- ra nd om ize d c on tro lle d t ria ls) pu bli sh ed be tw ee n 2 00 2 an d 2 00 9. In te rv en tio ns in clu de d co m bin at ion of nu tri tio n ed uc at ion an d c ha ng es to th e sch oo l fo od en vir on m en t ( e.g . fre e b re ak fas t p ro vid ed at sch oo l, i nc re as ed av ail ab ilit y an d p ro m ot ion of F& V i n th e s ch oo l c afe te ria an d ve nd ing m ac hin es , fr ee F& V sn ac k, an d i nc re as ed ch oic e, m ar ke tin g, an d i m pr ov em en t in th e p re pa ra tio n a nd ap pe ar an ce of fis h s er ve d in th e s ch oo l c afe te ria ), co m pa re d t o e ith er on e o f th es e a lon e. Ch ild re n w ith m ea n a ge 12 –1 5 ye ar s i n t he sc ho ol se tti ng in No rth Am er ica an d E ur op e. Ov er all , a ll 5 st ud ies sh ow ed th at co m bin ing ch an ge s t o t he sc ho ol fo od en vir on m en t w ith nu tri tio n e du ca tio n w as eff ec tiv e f or im pr ov ing ch ild re n a nd ad ole sce nt ’s d iet ar y i nt ak e ( e.g . F &V int ak e, fis h c on su m pt ion ), an d s om e e vid en ce su gg es te d t ha t t he co m bin at ion le d t o g re at er im pr ov em en t t ha n e ith er in te rv en tio n a lon e. In 2 stu die s t he co m bin ed ap pr oa ch w as m or e eff ec tiv e, in 1 s tu dy it w as m or e e ffe cti ve in bo ys on ly an d i n 2 st ud ies bo th ap pr oa ch es w er e eff ec tiv e. US DA , 2 01 2 ( 12 1) 3.1 Th is sy ste m at ic re vie w of 14 st ud ies , 1 0 o f w hic h we re sc ho ol- ba se d, fo un d inc on sis te nt ev ide nc e o f m ult i v s. sin gle co m po ne nt nu tri tio n e du ca tio n int er ve nt ion s i n i m pr ov ing ch ild an d a do les ce nt di et ar y be ha vio ur, an d l im ite d ev ide nc e t ha t n ut rit ion ed uc at ion co m bin ed w ith ha nd s-o n e du ca tio na l co m po ne nt m ay be pa rti cu lar ly eff ec tiv e. Th e re vie w wa s c on du cte d a s pa rt of a se rie s o f s ys te m at ic re vie ws on va rio us as pe cts of nu tri tio n e du ca tio n i n co un tri es w ith a hig h o r v er y hig h h um an de ve lop m en t ind ex . Th e r ev iew in clu de d 1 4 stu die s ( 10 RC Ts an d 4 no n- ra nd om ize d c on tro lle d t ria ls) pu bli sh ed be tw ee n 1 99 6 an d 2 01 0. Nu tri tio n e du ca tio n a lon e as a sin gle co m po ne nt (e .g. us ua l c las sro om ed uc at ion or co un se llin g s es sio ns ) o r co m bin ed w ith su pp or tin g co m po ne nt s ( e.g . p ar en ta l inv olv em en t, we b- ba se d ac tiv ity , g ar de nin g o r t ak e ho m e m at er ial s). Ch ild re n w ith m ea n a ge 7– 15 ye ar s l ar ge ly in th e U nit ed St at es bu t a lso Ca na da an d Eu ro pe ; 1 0 o f 1 4 s tu die s w er e sch oo l-b as ed . Th er e w as no ag re em en t o n w he th er m ult ico m po ne nt nu tri tio n e du ca tio n pr og ra m m es in ch an gin g d iet ar y p ra cti ce s t ha n sin gle co m po ne nt pr og ra m m es an d s ev er al of th e s tu die s h ad m ixe d r es ult s. In to ta l, 6 stu die s f ou nd m ult ico m po ne nt in te rv en tio ns to be m or e e ffe cti ve th an si ng le co m po ne nt int er ve nt ion s f or at le as t o ne di et ar y o ut co m e, 3 s tu die s f ou nd si ng le- co m po ne nt nu tri tio n ed uc at ion in te rv en tio ns to be m or e e ffe cti ve th an m ult ico m po ne nt in te rv en tio ns fo r a t l ea st on e d iet ar y o ut co m e, an d 8 di d n ot fin d a ny diff er en ce be tw ee n t he m . T he au th or s n ot ed th at th e 3 st ud ies th at co ns ist en tly fo un d b et te r eff ec t o f m ult ico m po ne nt in te rv en tio ns w er e a lso th e o nl y o ne s t o c om bin e n ut rit ion ed uc at ion wi th a “h an ds -o n” co m po ne nt (e .g. co ok ing or ga rd en ing ). AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Va n Ca uw en be rg he et al. , 2 01 0 ( 13 5) 3.1 , 4 .1 Th is sy ste m at ic re vie w of 42 Eu ro pe an sc ho ol- ba se d stu die s f ou nd ev ide nc e f or th e e ffe cti ve ne ss of ob es ity pr ev en tio n i nt er ve nt ion s pr om ot ing a he alt hy di et in sch oo l-a ge d c hil dr en on se lf- re po rte d d iet ar y b eh av iou r bu t e vid en ce fo r e ffe cti ve ne ss on an th ro po m et ric m ea su re s is lac kin g. Th e r ev iew in clu de d 4 2 stu die s ( re po rte d i n 5 3 ar tic les in clu din g g re y lit er at ur e) , o f w hic h 1 3 w er e RC Ts, 16 no n- ra nd om ize d co nt ro lle d t ria ls, 7 pr e- an d po st- te st an d 6 pr os pe cti ve co ho rts . S tu die s w er e pu bli sh ed be tw ee n 1 99 1 an d 2 00 8. In te rv en tio ns w er e a im ed at th e p rim ar y p re ve nt ion of ob es ity an d o be sit y- re lat ed di se as es ta rg et ing die ta ry be ha vio ur th ro ug h ed uc at ion al on e i n 2 0 s tu die s (e .g. cl as sro om le sso ns , ed uc at ion al m at er ial s), th ro ug h e nv iro nm en t a lon e in 8 s tu die s ( e.g . in cre as ed av ail ab ilit y a nd ac ce ssi bil ity of he alt hy fo od s, F& V o r br ea kfa st su bs cri pt ion or dis tri bu tio n p ro gr am m es , sch oo l lu nc h m od ifi ca tio ns ) or th ro ug h a co m bin at ion of ed uc at ion an d e nv iro nm en ta l m ea su re s i n 1 4 s tu die s. Pr im ar y s ch oo l c hil dr en ag ed 6– 12 ye ar s ( 29 st ud ies ) a nd ad ole sce nt s a ge d 1 3– 18 ye ar s (1 3 s tu die s) in sch oo l s et tin gs in Eu ro pe . M os t s tu die s i n b ot h a ge gr ou ps im pr ov ed di et ar y be ha vio ur (i nt ak e o f F &V , fa ts, so ft dr ink s, wa te r, fis h) , b ut on ly a f ew st ud ies in th e y ou ng es t a ge gr ou p i m pr ov ed bo dy co m po sit ion . In ch ild re n, str on g e vid en ce w as fo un d f or m ult ico m po ne nt int er ve nt ion s o n F &V in ta ke s i n c hil dr en ag ed 6– 12 ye ar s b ut lim ite d e vid en ce w as fo un d f or ed uc at ion al int er ve nt ion s o n b eh av iou r, a nd fo r en vir on m en ta l in te rv en tio ns on F& V i nt ak es . In ad ole sce nt s a ge d 1 3– 18 ye ar s, m od er at e ev ide nc e w as fo un d f or ed uc at ion al int er ve nt ion s on di et ar y b eh av iou r a nd lim ite d e vid en ce fo r m ult ico m po ne nt pr og ra m m es on di et ar y be ha vio ur. In ch ild re n a nd ad ole sce nt s, eff ec ts on an th ro po m et ric ou tco m es w er e o fte n n ot m ea su re d, an d t he re fo re ev ide nc e w as la ck ing or inc on clu siv e. Al l fi ve st ud ies on F& V s ub scr ipt ion or di str ibu tio n p ro gr am m es fo un d e ffe cti ve ne ss on F& V i nt ak es ; h ow ev er, on ly on e s tu dy sh ow ed a l as tin g e ffe ct an d t he re fo re ev ide nc e t ha t en vir on m en ta l in te rv en tio ns ca n i m pr ov e F &V int ak es is lim ite d. Va n L ipp ev eld e e t al. , 2 01 2 ( 10 4) 2.2 Th is sy ste m at ic re vie w to de te rm ine th e e ffe ct of pa re nt al inv olv em en t in sch oo l-b as ed ob es ity pr ev en tio n p ro gr am m es fo cu se d o n d iet an d P A f ou nd inc on clu siv e e vid en ce du e t o th e l ac k o f q ua lit y s tu die s a nd inc on sis te nt re su lts . Th e r ev iew in clu de d 5 stu die s ( 4 c lus te r R CT s, 1 R CT ) pu bli sh ed be tw ee n 1 99 0 an d 2 00 3. Sc ho ol- ba se d i nt er ve nt ion s to re du ce ob es ity us ing ed uc at ion al (e .g. he alt h ed uc at ion , P E, ta ilo re d fee db ac k) an d e nv iro nm en ta l (e .g. sc ho ol fo od se rv ice m od ifi ca tio ns , m or e op po rtu nit ies fo r P A o ut sid e of cl as se s) co m po ne nt s. Ch ild re n a ge d 6 –1 8 y ea rs in th e U nit ed St at es , A us tra lia an d B elg ium in sc ho ol se tti ng s. Fo ur ou t o f 5 st ud ies re po rte d p os iti ve eff ec ts of pa re nt al inv olv em en t o n d ive rse ou tco m es (d iet ar y k no wl ed ge , h ea lth be ha vio ur s, BM I, BM I z -sc or e, fat in ta ke ). T he au th or s c on clu de d th at th er e i s a la ck of ev ide nc e t o s up po rt th e cla im th at pa re nt al inv olv em en t i s i m po rta nt to im pr ov e e ffe cti ve ne ss of sc ho ol- ba se d be ha vio ur al nu tri tio n a nd PA in te rv en tio ns du e to a lac k o f s tu die s c om pa rin g i nt er ve nt ion s w ith an d w ith ou t a pa re nt al co m po ne nt an d d ue to lar ge ly inc on sis te nt re su lts in ex ist ing st ud ies . 153ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Va rg as -G ar cia et al. , 2 01 7 ( 11 7) 3.1 , 4 .11 Th e r es ult s o f t his sy ste m at ic re vie w an d m et a- an aly sis of 40 st ud ies , 1 5 o f w hic h w er e sch oo l-b as ed , s ug ge st th at sch oo l-b as ed in te rv en tio ns m ay re du ce SS B c on su m pt ion in ch ild re n. Th e r ev iew in clu de d 5 0 stu die s ( RC Ts, cl us te r R CT s an d n on -ra nd om ize d co nt ro lle d t ria ls) in th e qu ali ta tiv e r ev iew , o f w hic h 40 st ud ies w er e i nc lud ed in th e m et a- an aly sis . Sc ho ol int er ve nt ion s w er e pr im ar ily nu tri tio n e du ca tio n (e .g. le sso ns , h om ew or k, pr int ed m at er ial s) to re du ce ob es ity or SS B c on su m pt ion . Fiv e i nt er ve nt ion s i nc lud ed ins ta lla tio n o f w at er fo un ta ins or pr ov isi on of wa te r b ot tle s. Al l p op ula tio n g ro up s. Of th e 40 st ud ies , 2 3 h ad da ta on ch ild re n ( n= 10 96 4) , 5 on ad ole sce nt s ( n= 31 17 ) a nd 12 on ad ult s ( n= 24 24 ), wi th on e stu dy re po rti ng in ta ke s o n bo th ch ild re n a nd ad ult s a nd an ot he r o n b ot h a do les ce nt s an d a du lts . C ou nt rie s i nc lud ed Au str ali a, Be lgi um , B ra zil , Ca na da , C hil e, Ge rm an y, M ala ys ia, M ex ico , N ew Ze ala nd , N or wa y, Po rtu ga l, Tu rke y, th e N et he rla nd s, th e U nit ed Ki ng do m an d Un ite d S ta te s. Se tti ng s o f int er ve nt ion s w er e s ch oo l- ba se d ( 15 st ud ies ), ho m e (9 st ud ies ), co m m un ity (1 1 stu die s) an d c lin ics (5 st ud ies ). In te rv en tio ns si gn ifi ca nt ly de cre as ed co ns um pt ion of SS B in ch ild re n b y 7 6 m L/ d (9 5% CI : - 10 5, -4 6; 23 st ud ies ; P < 0. 01 ) a nd in ad ol es ce nt s ( -6 6 m L/ d; 95 % CI : - 13 0, -2 ; 5 stu di es ; P = 0. 04 ). Po ol ed es tim at es of w at er in ta ke s s ho we d i nc re as ed co ns um pt ion in ch ild re n ( M D: + 67 m L/ d; 95 % CI : 6 , 1 28 ; 7 s tu di es ; P = 0. 04 ), bu t s ub gr ou p a na lys is by se tti ng w as no t p os sib le. In su bg ro up an aly se s o f s tu di es w ith ch ild re n, si gn ifi ca nt de cre as es in SS B co ns um pt ion w er e s ee n f or co m m un ity ( -7 2 m L/ d; 95 % CI : - 11 5, -3 0; P< 0. 01 ) a nd sc ho ol -b as ed (- 28 m L/ d; 95 % CI : -4 2, -1 2; P< 0. 01 ) s tu di es bu t n ot fo r c lin ica l o r ho m e- ba se d. Fo r c hi ld re n, th er e w as ev id en ce to su gg es t t ha t m od ell in g/ de m on str at in g th e b eh av iou r h elp ed to re du ce SS B in ta ke . M od ell in g t he be ha vio ur w as as so cia te d w ith gr ea te r e ffe cti ve ne ss in re du cin g S SB af te r un iva ria te m et a- re gr es sio ns w er e c on du cte d ac ro ss al l a ge gr ou ps (- 12 4 m L/ d; 95 % CI : - 22 1, -2 7; P= 0. 01 ), wi th si m ila r e ffe cts fo un d f or stu di es in ch ild re n ( -1 73 m L/ d; 95 % CI : - 31 5, -3 1; P= 0. 02 ). Va sq ue s e t a l., 20 14 (5 0) 1.1 , 2 .2, 3. 2 Th is sy ste m at ic re vie w an d m et a- an aly sis on th e effi ca cy of sc ho ol- ba se d a nd af te r-s ch oo l in te rv en tio n pr og ra m m es to ad dr es s ob es ity on BM Is of ch ild re n an d a do les ce nt s f ou nd th at pa re nt al inv olv em en t w as a m od er at or th at in cre as ed th e eff ec t s ize . Th e r ev iew in clu de d 5 2 stu die s o f R CT s o r n on - ra nd om ize d c lin ica l t ria ls pu bli sh ed be tw ee n 2 00 0 an d 2 01 1. In te rv en tio ns in clu de d P A, die t o r l ife sty le ch an ge s us ing m ult ipl e a pp ro ac he s inc lud ing ed uc at ion (n ut rit ion ed uc at ion , int er ac tiv e m ult im ed ia, en co ur ag em en t o f P A a nd he alt h- re lat ed se ssi on s), nu tri tio n p oli cy , fo od se rv ice m od ifi ca tio ns an d s oc ial m ar ke tin g. Ha lf o f t he int er ve nt ion s i nv olv ed fa m ily m em be rs or pa re nt s ( e.g . ou tre ac h, ed uc at ion ). Ch ild re n a nd ad ole sce nt s a ge d 5– 18 ye ar s ( n= 28 23 6) in sch oo l o r a fte r-s ch oo l s et tin gs . Ge og ra ph ica l lo ca tio n w as no t re po rte d. Th e o ve ral l e ffe ct siz e o n B MI w as sm all fo r a ll stu die s i nc lud ed (0 .06 8; P< 0.0 01 ), s ch oo l-b as ed (r= 0.0 69 ) a s w ell as af ter -sc ho ol int er ve nt ion s (r= 0.0 65 ), a cco rd ing to Co he n w he re th e sig nifi ca nc e o f e ffe ct siz es is co ns ide red sm all fo r les s t ha n 0 .10 , m ed ium fo r le ss th an 0. 30 , a nd lar ge fo r le ss th an 0. 50 . P ro gr am m es co m bin ing PA an d d iet w ere th e m os t s uc ce ssf ul wi th a m ed ium eff ec t s ize (r =0 .14 8) , w he rea s t ho se co nc en tra tin g on ly on PA le ve ls we re les s s uc ce ssf ul wi th a sm all eff ec t s ize (r =0 .02 9) . T he eff ec t s ize s f or pa ren ta l inv olv em en t w ere al l s m all , b ut in cre as ed as th e l ev el of pa ren ta l in vo lve m en t in cre as ed : n o inv olv em en t ( r= 0.0 47 ), m ini m al inv olv em en t (r= 0.0 57 ), m od era te inv olv em en t ( r= 0.0 82 ) an d h igh in vo lve m en t ( r= 0.0 94 ) w ith si gn ific an t diff ere nc es be tw ee n g ro up s ( P= 0.0 01 ). 154 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Va rg as -G ar cia et al. , 2 01 7 ( 11 7) 3.1 , 4 .11 Th e r es ult s o f t his sy ste m at ic re vie w an d m et a- an aly sis of 40 st ud ies , 1 5 o f w hic h w er e sch oo l-b as ed , s ug ge st th at sch oo l-b as ed in te rv en tio ns m ay re du ce SS B c on su m pt ion in ch ild re n. Th e r ev iew in clu de d 5 0 stu die s ( RC Ts, cl us te r R CT s an d n on -ra nd om ize d co nt ro lle d t ria ls) in th e qu ali ta tiv e r ev iew , o f w hic h 40 st ud ies w er e i nc lud ed in th e m et a- an aly sis . Sc ho ol int er ve nt ion s w er e pr im ar ily nu tri tio n e du ca tio n (e .g. le sso ns , h om ew or k, pr int ed m at er ial s) to re du ce ob es ity or SS B c on su m pt ion . Fiv e i nt er ve nt ion s i nc lud ed ins ta lla tio n o f w at er fo un ta ins or pr ov isi on of wa te r b ot tle s. Al l p op ula tio n g ro up s. Of th e 40 st ud ies , 2 3 h ad da ta on ch ild re n ( n= 10 96 4) , 5 on ad ole sce nt s ( n= 31 17 ) a nd 12 on ad ult s ( n= 24 24 ), wi th on e stu dy re po rti ng in ta ke s o n bo th ch ild re n a nd ad ult s a nd an ot he r o n b ot h a do les ce nt s an d a du lts . C ou nt rie s i nc lud ed Au str ali a, Be lgi um , B ra zil , Ca na da , C hil e, Ge rm an y, M ala ys ia, M ex ico , N ew Ze ala nd , N or wa y, Po rtu ga l, Tu rke y, th e N et he rla nd s, th e U nit ed Ki ng do m an d Un ite d S ta te s. Se tti ng s o f int er ve nt ion s w er e s ch oo l- ba se d ( 15 st ud ies ), ho m e (9 st ud ies ), co m m un ity (1 1 stu die s) an d c lin ics (5 st ud ies ). In te rv en tio ns si gn ifi ca nt ly de cre as ed co ns um pt ion of SS B in ch ild re n b y 7 6 m L/ d (9 5% CI : - 10 5, -4 6; 23 st ud ies ; P < 0. 01 ) a nd in ad ol es ce nt s ( -6 6 m L/ d; 95 % CI : - 13 0, -2 ; 5 stu di es ; P = 0. 04 ). Po ol ed es tim at es of w at er in ta ke s s ho we d i nc re as ed co ns um pt ion in ch ild re n ( M D: + 67 m L/ d; 95 % CI : 6 , 1 28 ; 7 s tu di es ; P = 0. 04 ), bu t s ub gr ou p a na lys is by se tti ng w as no t p os sib le. In su bg ro up an aly se s o f s tu di es w ith ch ild re n, si gn ifi ca nt de cre as es in SS B co ns um pt ion w er e s ee n f or co m m un ity ( -7 2 m L/ d; 95 % CI : - 11 5, -3 0; P< 0. 01 ) a nd sc ho ol -b as ed (- 28 m L/ d; 95 % CI : -4 2, -1 2; P< 0. 01 ) s tu di es bu t n ot fo r c lin ica l o r ho m e- ba se d. Fo r c hi ld re n, th er e w as ev id en ce to su gg es t t ha t m od ell in g/ de m on str at in g th e b eh av iou r h elp ed to re du ce SS B in ta ke . M od ell in g t he be ha vio ur w as as so cia te d w ith gr ea te r e ffe cti ve ne ss in re du cin g S SB af te r un iva ria te m et a- re gr es sio ns w er e c on du cte d ac ro ss al l a ge gr ou ps (- 12 4 m L/ d; 95 % CI : - 22 1, -2 7; P= 0. 01 ), wi th si m ila r e ffe cts fo un d f or stu di es in ch ild re n ( -1 73 m L/ d; 95 % CI : - 31 5, -3 1; P= 0. 02 ). Va sq ue s e t a l., 20 14 (5 0) 1.1 , 2 .2, 3. 2 Th is sy ste m at ic re vie w an d m et a- an aly sis on th e effi ca cy of sc ho ol- ba se d a nd af te r-s ch oo l in te rv en tio n pr og ra m m es to ad dr es s ob es ity on BM Is of ch ild re n an d a do les ce nt s f ou nd th at pa re nt al inv olv em en t w as a m od er at or th at in cre as ed th e eff ec t s ize . Th e r ev iew in clu de d 5 2 stu die s o f R CT s o r n on - ra nd om ize d c lin ica l t ria ls pu bli sh ed be tw ee n 2 00 0 an d 2 01 1. In te rv en tio ns in clu de d P A, die t o r l ife sty le ch an ge s us ing m ult ipl e a pp ro ac he s inc lud ing ed uc at ion (n ut rit ion ed uc at ion , int er ac tiv e m ult im ed ia, en co ur ag em en t o f P A a nd he alt h- re lat ed se ssi on s), nu tri tio n p oli cy , fo od se rv ice m od ifi ca tio ns an d s oc ial m ar ke tin g. Ha lf o f t he int er ve nt ion s i nv olv ed fa m ily m em be rs or pa re nt s ( e.g . ou tre ac h, ed uc at ion ). Ch ild re n a nd ad ole sce nt s a ge d 5– 18 ye ar s ( n= 28 23 6) in sch oo l o r a fte r-s ch oo l s et tin gs . Ge og ra ph ica l lo ca tio n w as no t re po rte d. Th e o ve ral l e ffe ct siz e o n B MI w as sm all fo r a ll stu die s i nc lud ed (0 .06 8; P< 0.0 01 ), s ch oo l-b as ed (r= 0.0 69 ) a s w ell as af ter -sc ho ol int er ve nt ion s (r= 0.0 65 ), a cco rd ing to Co he n w he re th e sig nifi ca nc e o f e ffe ct siz es is co ns ide red sm all fo r les s t ha n 0 .10 , m ed ium fo r le ss th an 0. 30 , a nd lar ge fo r le ss th an 0. 50 . P ro gr am m es co m bin ing PA an d d iet w ere th e m os t s uc ce ssf ul wi th a m ed ium eff ec t s ize (r =0 .14 8) , w he rea s t ho se co nc en tra tin g on ly on PA le ve ls we re les s s uc ce ssf ul wi th a sm all eff ec t s ize (r =0 .02 9) . T he eff ec t s ize s f or pa ren ta l inv olv em en t w ere al l s m all , b ut in cre as ed as th e l ev el of pa ren ta l in vo lve m en t in cre as ed : n o inv olv em en t ( r= 0.0 47 ), m ini m al inv olv em en t (r= 0.0 57 ), m od era te inv olv em en t ( r= 0.0 82 ) an d h igh in vo lve m en t ( r= 0.0 94 ) w ith si gn ific an t diff ere nc es be tw ee n g ro up s ( P= 0.0 01 ). AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ve laz qu ez , B lac k an d P ot vin Ke nt , 20 17 (1 67 ) 4.3 Th is re vie w of fo od an d be ve ra ge m ar ke tin g i n sch oo ls fo un d t ha t s tu de nt s ar e c om m on ly ex po se d to fo od an d b ev er ag e m ar ke tin g t hr ou gh a va rie ty of st ra te gie s, an d f ou nd nu m er ou s i ns ta nc es of no n- co m pli an t m ar ke tin g. Th e r ev iew in clu de d 2 7 cro ss- se cti on al stu die s ( 20 un iqu e d at a s ou rce s). M or e th an ha lf ( 15 ) o f t he st ud ies we re pu bli sh ed be tw ee n 20 10 an d 2 01 5, an d o nl y t wo be fo re 20 05 . In ad dit ion to as se ssi ng th e e xp os ur e t o m ar ke tin g of fo od s a nd be ve ra ge s in sch oo ls an d p re se nc e (o r a bs en ce ) o f p oli cie s o r gu ide lin es re gu lat ing fo od an d b ev er ag e m ar ke tin g in sch oo ls, tw o s tu die s ev alu at ed th e c om pli an ce to ex ist ing m ar ke tin g r es tri cti on law s. Ele m en ta ry, m idd le an d h igh sch oo ls in th e U nit ed St at es , Ca na da , Ir ela nd an d P ola nd . Th e r ev iew fo un d t ha t s tu de nt s a re co m m on ly ex po se d t o f oo d a nd be ve ra ge m ar ke tin g th ro ug h a va rie ty of st ra te gie s, es pe cia lly am on g h igh sc ho ols an d s ch oo ls wi th lo we r so cio -e co no m ic sta tu s. M ar ke tin g e xp os ur e is of te n a sso cia te d w ith fo od pu rch as ing an d co ns um pt ion , e sp ec ial ly fo r f oo d i te m s o f lo w nu tri tio na l q ua lit y ( e.g . s alt y s na ck s, ca nd y, SS Bs ). Fu rth er m or e, m an y s ch oo ls do no t h av e po lic ies or gu ide lin es re gu lat ing su ch m ar ke tin g. Tw o s tu die s e va lua tin g t he co m pli an ce w ith la ws or po lic ies re str ict ing sc ho ol fo od an d b ev er ag e m ar ke tin g a nd fo un d n um er ou s i ns ta nc es of no n- co m pli an t m ar ke tin g. Ve rja ns -Ja ns se n e t al. , 2 01 8 ( 95 ) 2.2 Th is sy ste m at ic re vie w of 25 sc ho ol- ba se d n ut rit ion an d P A i nt er ve nt ion s w ith a dir ec t p ar en ta l in vo lve m en t co m po ne nt fo un d f av ou ra ble re su lts on BM I o r B M I z -sc or e bu t m ixe d r es ult s o n n ut rit ion be ha vio ur ou tco m es . Th e r ev iew in clu de d 2 5 stu die s o f v ar iou s d es ign s ( 10 RC Ts, 11 qu as i-e xp er im en ta l stu die s, 3 p re - a nd po st- te st de sig n, 1 re pe at ed cr os s- se cti on al stu dy ) p ub lis he d be tw ee n 1 99 6 a nd 20 18 . In te rv en tio ns ta rg et ed m ult ipl e e nv iro nm en ta l t yp es (e .g. so cia l s ch oo l a nd fa m ily , ph ys ica l s ch oo l, p oli tic al sch oo l e nv iro nm en ts) ad dr es sin g n ut rit ion (3 stu die s), PA (2 st ud ies ) or bo th (2 0 s tu die s) an d we re im ple m en te d w ith a dir ec t p ar en t i nv olv em en t co m po ne nt (e .g. re qu es tin g pa re nt s t o a tte nd ed uc at ion se ssi on s, as kin g p ar en ts to at te nd or pa rti cip at e i n fam ily be ha vio ur co un se llin g or pa re nt tr ain ing se ssi on s). Ch ild re n a ge d 4 –1 2 y ea rs in th e s ch oo l s et tin g, lar ge ly in th e U nit ed St at es as w ell as in ot he r c ou nt rie s. Ele ve n o f 1 8 s tu die s m ea su rin g B M I o r B M I z-s co re fo un d f av ou ra ble re su lts . O f t he 11 stu die s w ith fa vo ur ab le re su lts , 6 ha d s m all eff ec t s ize s ( ES -0 .04 to -0 .27 ), 2 h ad m od er at e eff ec t s ize s ( ES -0 .34 an d - 0.4 8) an d 1 ha d a lar ge eff ec t o n B M I ( ES -0 .79 ). Re su lts re ga rd ing nu tri tio n b eh av iou r ( e.g . S SB in ta ke , F &V in ta ke , ad de d s ug ar in ta ke , re d m ea t c on su m pt ion ) we re in co nc lus ive , w ith 5 of 11 st ud ies re po rti ng fav ou ra ble re su lts an d a no th er 5 ou t o f 1 1 re po rti ng m ixe d r es ult s o n o ut co m es m ea su re d. Th e t wo in te rv en tio ns ta rg et ing on ly PA an d th e t hr ee in te rv en tio ns ta rg et ing on ly nu tri tio n be ha vio ur w er e m ain ly ine ffe cti ve . 155ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ve rst ra et en et al ., 20 12 (6 9) 1.1 , 3 .1, 3. 2 Th is sy ste m at ic re vie w of 25 sch oo l-b as ed di et an d P A int er ve nt ion s i n m idd le- an d low -in co m e c ou nt rie s f ou nd th at m os t i nt er ve nt ion s h ad a p os iti ve eff ec t o n d iet an d PA be ha vio ur s a nd BM I a nd th at eff ec tiv e i nt er ve nt ion s ta rg et ed bo th di et an d PA , in vo lve d m ult ipl e sta ke ho lde rs, an d i nt eg ra te d ed uc at ion al ac tiv iti es in to th e sch oo l c ur ric ulu m . Th e r ev iew in clu de d 25 st ud ies , p rim ar ily clu ste r-c on tro lle d t ria ls wi th (8 st ud ies ) o r w ith ou t ra nd om iza tio n ( 10 st ud ies ), pr es en te d i n 2 9 p ub lic at ion s wh ich w er e p ub lis he d be tw ee n 1 99 2 a nd 20 10 . Th irt ee n i nt er ve nt ion s inv olv ed sc ho ol sta ff, co m m un iti es , p ar en ts, ch ild re n a nd fa m ilie s an d a re re fer re d t o as m ult ico m po ne nt int er ve nt ion s. Fo ur of th e 25 in te rv en tio ns pr ov ide d an in div idu al co un se llin g co m po ne nt , 4 st ud ies w er e die t-o nl y i nt er ve nt ion s (e .g. nu tri tio n e du ca tio n, sch oo l b re ak fas t), 10 st ud ies we re PA -o nl y i nt er ve nt ion s (in cre as ed ph ys ica l ed uc at ion ), an d 1 1 s tu die s inv olv ed bo th di et an d P A. Ch ild re n, m ea n a ge 6. 5 t o 1 8.4 ye ar s, in sch oo ls, gl ob all y b ut pr im ar ily in La tin Am er ica (1 3 stu die s) an d A sia (8 st ud ies ). Tw o- th ird s o f t he st ud ies we re co nd uc te d i n m ajo r ur ba n a re as (1 6 s tu die s) an d 2 stu die s i nc lud ed a ru ra l a re a. Of th e 2 2 s tu die s r ep or tin g u se fu l st ati sti cs on th e ou tco m es of in ter es t, 1 8 h ad a po sit ive eff ec t o n on e o r m ore of th e o ut co m es (8 2% ). T he 3 die t int er ve nt ion s h ad sig nifi ca nt eff ec t o n d iet ar y be ha vio ur (e .g. pr efe ren ce s o r c on su m pt ion of so da , fa st foo d, sn ac ks ) ( 2 o ut of 3 m ea su rin g th is) , b ut di d n ot ha ve an eff ec t o n a ny of th e BM I-r ela ted ou tco m es . T he 9 PA in ter ve nt ion s w ere su cce ssf ul at inc rea sin g P A ( 6 o f 7 m ea su rin g P A) an d l ow eri ng BM I in at le as t o ne of th e s tu die d su bg rou ps (3 of 4 ev alu ati ng BM I), bu t d id no t ha ve an eff ec t o n t he pr ev ale nc e o f o ve rw eig ht or ob es ity (0 of 2 ev alu ati ng th is o ut co m e). Th e 1 0 co m bin ed in ter ve nt ion s h ad a sig nifi ca nt eff ec t on al l o ut co m es : d iet (3 of 3 m ea su rin g d iet ), P A (2 of 2 m ea su rin g P A) , B MI (5 of 6 in at lea st on e su bg rou p) , a nd th e p rev ale nc e o f o ve rw eig ht or ob es ity (3 of 3 ev alu ati ng th is o ut co m e). Eff ec tiv e int er ve nt ion s t arg ete d b ot h d iet an d P A, in vo lve d m ult ipl e s tak eh old ers , a nd in teg rat ed ed uc ati on al ac tiv itie s in to th e s ch oo l c ur ric ulu m w ith ed uc ati on - ba se d i nt er ve nt ion s d eli ve red by te ac he rs an d pr ov idi ng ad dit ion al PA se ssi on s o r in teg rat ed cla sse s a bo ut he alt hy fo od s, n ut rit ion , o r P A t o en co ur ag e c hil dr en to ad op t a he alt hy lif es tyl e. Vé zin a- Im et al ., 20 17 (4 5) 1.1 , 3 .1, 4. 1 Th is sy ste m at ic re vie w of 36 st ud ies on sc ho ol- ba se d int er ve nt ion s a im ed at re du cin g S SB co ns um pt ion am on g a do les ce nt s fo un d t ha t m os t ( 70 % ) int er ve nt ion s w er e e ffe cti ve in de cre as ing co ns um pt ion , wi th th e h igh es t s uc ce ss ra te (9 0% ) a m on g l eg isl at ive / en vir on m en ta l a pp ro ac he s. Th e r ev iew in clu de d 3 6 stu die s ( RC Ts, qu as i- ex pe rim en ta l s tu die s a nd on e- gr ou p p re - a nd po st- te st stu die s) pu bli sh ed be tw ee n 19 98 an d 2 01 5. Of th e 3 6 s tu die s a im ed at red uc ing SS B c on su m pt ion , 2 0 we re ed uc ati on al/ be ha vio ura l (e. g. ex pla ini ng ne ga tiv e h ea lth co ns eq ue nc es, be ha vio ura l g oa l se tti ng , w rit ten or el ec tro nic co m m un ica tio n) , 1 0 w ere leg isla tiv e/ en vir on m en tal (e .g. ba nn ing SS B o r re pla cin g S SB wi th he alt hie r a lte rn ati ve s) an d 6 i nc lud ed bo th co m po ne nt s. In m os t o f th e s tu die s, r ed uc ing SS B c on su m pt ion w as pa rt of a bro ad er go al to im pro ve di et an d he alt h. Ch ild re n a ge d 1 2– 17 ye ar s (n = 15 2 0 01 ) i n t he sc ho ol se tti ng . M os t w er e c on du cte d in th e U nit ed St at es bu t a lso in Ca na da , A us tra lia , B elg ium , Br az il, Ch ina , In dia , R ep ub lic of Ko re a a nd th e N et he rla nd s. In fo rm at ion w as co lle cte d o n i nd ivi du al SS B co ns um pt ion in clu din g f re qu en cy or am ou nt , a nd co m pr ise d o f s of t d rin ks , n on -1 00 % fr uit ju ice s, en er gy dr ink s, sp or ts dr ink s, sw ee te ne d t ea an d co ffe e, an d a ny ot he r b ev er ag es w ith ad de d su ga r. T he 10 le gis lat ive /e nv iro nm en ta l s tu die s ha d t he hi gh es t s uc ce ss ra te (9 0% ), wh ile th e 20 ed uc at ion al/ be ha vio ur al int er ve nt ion s a nd th e 6 in te rv en tio ns th at w er e b ot h e du ca tio na l/ be ha vio ur al an d l eg isl at ive /e nv iro nm en ta l h ad alm os t e qu all y e ffe cti ve su cce ss ra te s ( 65 % an d 66 .7% , re sp ec tiv ely ). 156 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Ve rst ra et en et al ., 20 12 (6 9) 1.1 , 3 .1, 3. 2 Th is sy ste m at ic re vie w of 25 sch oo l-b as ed di et an d P A int er ve nt ion s i n m idd le- an d low -in co m e c ou nt rie s f ou nd th at m os t i nt er ve nt ion s h ad a p os iti ve eff ec t o n d iet an d PA be ha vio ur s a nd BM I a nd th at eff ec tiv e i nt er ve nt ion s ta rg et ed bo th di et an d PA , in vo lve d m ult ipl e sta ke ho lde rs, an d i nt eg ra te d ed uc at ion al ac tiv iti es in to th e sch oo l c ur ric ulu m . Th e r ev iew in clu de d 25 st ud ies , p rim ar ily clu ste r-c on tro lle d t ria ls wi th (8 st ud ies ) o r w ith ou t ra nd om iza tio n ( 10 st ud ies ), pr es en te d i n 2 9 p ub lic at ion s wh ich w er e p ub lis he d be tw ee n 1 99 2 a nd 20 10 . Th irt ee n i nt er ve nt ion s inv olv ed sc ho ol sta ff, co m m un iti es , p ar en ts, ch ild re n a nd fa m ilie s an d a re re fer re d t o as m ult ico m po ne nt int er ve nt ion s. Fo ur of th e 25 in te rv en tio ns pr ov ide d an in div idu al co un se llin g co m po ne nt , 4 st ud ies w er e die t-o nl y i nt er ve nt ion s (e .g. nu tri tio n e du ca tio n, sch oo l b re ak fas t), 10 st ud ies we re PA -o nl y i nt er ve nt ion s (in cre as ed ph ys ica l ed uc at ion ), an d 1 1 s tu die s inv olv ed bo th di et an d P A. Ch ild re n, m ea n a ge 6. 5 t o 1 8.4 ye ar s, in sch oo ls, gl ob all y b ut pr im ar ily in La tin Am er ica (1 3 stu die s) an d A sia (8 st ud ies ). Tw o- th ird s o f t he st ud ies we re co nd uc te d i n m ajo r ur ba n a re as (1 6 s tu die s) an d 2 stu die s i nc lud ed a ru ra l a re a. Of th e 2 2 s tu die s r ep or tin g u se fu l st ati sti cs on th e ou tco m es of in ter es t, 1 8 h ad a po sit ive eff ec t o n on e o r m ore of th e o ut co m es (8 2% ). T he 3 die t int er ve nt ion s h ad sig nifi ca nt eff ec t o n d iet ar y be ha vio ur (e .g. pr efe ren ce s o r c on su m pt ion of so da , fa st foo d, sn ac ks ) ( 2 o ut of 3 m ea su rin g th is) , b ut di d n ot ha ve an eff ec t o n a ny of th e BM I-r ela ted ou tco m es . T he 9 PA in ter ve nt ion s w ere su cce ssf ul at inc rea sin g P A ( 6 o f 7 m ea su rin g P A) an d l ow eri ng BM I in at le as t o ne of th e s tu die d su bg rou ps (3 of 4 ev alu ati ng BM I), bu t d id no t ha ve an eff ec t o n t he pr ev ale nc e o f o ve rw eig ht or ob es ity (0 of 2 ev alu ati ng th is o ut co m e). Th e 1 0 co m bin ed in ter ve nt ion s h ad a sig nifi ca nt eff ec t on al l o ut co m es : d iet (3 of 3 m ea su rin g d iet ), P A (2 of 2 m ea su rin g P A) , B MI (5 of 6 in at lea st on e su bg rou p) , a nd th e p rev ale nc e o f o ve rw eig ht or ob es ity (3 of 3 ev alu ati ng th is o ut co m e). Eff ec tiv e int er ve nt ion s t arg ete d b ot h d iet an d P A, in vo lve d m ult ipl e s tak eh old ers , a nd in teg rat ed ed uc ati on al ac tiv itie s in to th e s ch oo l c ur ric ulu m w ith ed uc ati on - ba se d i nt er ve nt ion s d eli ve red by te ac he rs an d pr ov idi ng ad dit ion al PA se ssi on s o r in teg rat ed cla sse s a bo ut he alt hy fo od s, n ut rit ion , o r P A t o en co ur ag e c hil dr en to ad op t a he alt hy lif es tyl e. Vé zin a- Im et al ., 20 17 (4 5) 1.1 , 3 .1, 4. 1 Th is sy ste m at ic re vie w of 36 st ud ies on sc ho ol- ba se d int er ve nt ion s a im ed at re du cin g S SB co ns um pt ion am on g a do les ce nt s fo un d t ha t m os t ( 70 % ) int er ve nt ion s w er e e ffe cti ve in de cre as ing co ns um pt ion , wi th th e h igh es t s uc ce ss ra te (9 0% ) a m on g l eg isl at ive / en vir on m en ta l a pp ro ac he s. Th e r ev iew in clu de d 3 6 stu die s ( RC Ts, qu as i- ex pe rim en ta l s tu die s a nd on e- gr ou p p re - a nd po st- te st stu die s) pu bli sh ed be tw ee n 19 98 an d 2 01 5. Of th e 3 6 s tu die s a im ed at red uc ing SS B c on su m pt ion , 2 0 we re ed uc ati on al/ be ha vio ura l (e. g. ex pla ini ng ne ga tiv e h ea lth co ns eq ue nc es, be ha vio ura l g oa l se tti ng , w rit ten or el ec tro nic co m m un ica tio n) , 1 0 w ere leg isla tiv e/ en vir on m en tal (e .g. ba nn ing SS B o r re pla cin g S SB wi th he alt hie r a lte rn ati ve s) an d 6 i nc lud ed bo th co m po ne nt s. In m os t o f th e s tu die s, r ed uc ing SS B c on su m pt ion w as pa rt of a bro ad er go al to im pro ve di et an d he alt h. Ch ild re n a ge d 1 2– 17 ye ar s (n = 15 2 0 01 ) i n t he sc ho ol se tti ng . M os t w er e c on du cte d in th e U nit ed St at es bu t a lso in Ca na da , A us tra lia , B elg ium , Br az il, Ch ina , In dia , R ep ub lic of Ko re a a nd th e N et he rla nd s. In fo rm at ion w as co lle cte d o n i nd ivi du al SS B co ns um pt ion in clu din g f re qu en cy or am ou nt , a nd co m pr ise d o f s of t d rin ks , n on -1 00 % fr uit ju ice s, en er gy dr ink s, sp or ts dr ink s, sw ee te ne d t ea an d co ffe e, an d a ny ot he r b ev er ag es w ith ad de d su ga r. T he 10 le gis lat ive /e nv iro nm en ta l s tu die s ha d t he hi gh es t s uc ce ss ra te (9 0% ), wh ile th e 20 ed uc at ion al/ be ha vio ur al int er ve nt ion s a nd th e 6 in te rv en tio ns th at w er e b ot h e du ca tio na l/ be ha vio ur al an d l eg isl at ive /e nv iro nm en ta l h ad alm os t e qu all y e ffe cti ve su cce ss ra te s ( 65 % an d 66 .7% , re sp ec tiv ely ). AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W vo n P hil ips bo rn et al. , 2 01 9 ( 44 ) 1.1 , 4 .1, 4. 2, 4.6 Th is Co ch ra ne re vie w of 58 stu die s, 20 of w hic h w er e sch oo l-b as ed , fo un d t ha t lim iti ng th e a va ila bil ity of SS B i n s ch oo ls (e .g. re pla cin g SS Bs w ith w at er in sc ho ol ca fet er ias ) c an be eff ec tiv e i n re du cin g S SB in ta ke . Th e r ev iew in clu de d 5 8 stu die s ( 22 RC Ts, 3 no n- ra nd om ize d c on tro lle d tri als , 1 4 C BA st ud ies an d 19 in te rru pt ed ti m e s er ies stu die s) pu bli sh ed be tw ee n 20 06 an d 2 01 8. Th e r ev iew in clu de d int er ve nt ion s i m ple m en te d at an en vir on m en ta l le ve l, re po rti ng eff ec ts on di re ct or in dir ec t m ea su re s o f S SB int ak e i nc lud ing la be llin g, nu tri tio n s ta nd ar ds in pu bli c ins tit ut ion s, ec on om ic to ols , w ho le fo od su pp ly int er ve nt ion s, re ta il a nd fo od se rv ice in te rv en tio ns , int er se cto ra l a pp ro ac he s a nd ho m e- ba se d i nt er ve nt ion s. Al l t he nu tri tio n s ta nd ar ds in pu bli c i ns tit ut ion s int er ve nt ion s w er e c on du cte d in sch oo ls an d i nc lud ed re du ce d a va ila bil ity of SS Bs , im pr ov ed ac ce ss to dr ink ing wa te r, a wa rd ed pr ize s f or th e s ele cti on of he alt hie r op tio ns in sc ho ol ca fet er ias , im pr ov ed pl ac em en t o f he alt hie r b ev er ag es in sch oo l c afe te ria s a nd pr ov ide d f ru its . O ne la be llin g int er ve nt ion co nc er ne d t he us e o f e m ot ico ns in sc ho ol ca fet er ias , a nd on e r et ail an d fo od se rv ice in te rv en tio n co nc er ne d v en din g m ac hin es in wo rk pla ce s a nd sc ho ols . Th e f ull re vie w co ns ide rs all ag e g ro up s i n a va rie ty of se tti ng s, inc lud ing sc ho ols , re ta ilin g a nd fo od se rv ice es ta bli sh m en ts. Th e 2 0 s ch oo l- ba se d s tu die s w er e c on du cte d in ele m en ta ry, m idd le an d hig h s ch oo ls, la rg ely in th e Un ite d S ta te s, bu t a lso in Br az il, Ge rm an y, Ita ly, N or wa y an d t he N et he rla nd s. No m et a- an aly sis w as pe rfo rm ed on th e s ch oo l- ba se d s tu die s. Am on g t he 7 stu die s o n n ut rit ion sta nd ar ds in sc ho ols , t he re w as lo w- ce rta int y ev ide nc e f ro m 5 stu die s t ha t r ed uc ed av ail ab ilit y of SS B i n s ch oo ls is as so cia te d w ith re du ce d S SB int ak e; 3 o ut of 5 stu die s o n l ar ge ly go ve rn m en t po lic ies to re du ce th e a va ila bil ity of SS Bs re po rte d sig nifi ca nt eff ec t o n t ot al int ak e o f S SB an d ot he r b ev er ag es ex clu de d b y t he st an da rd or th e s ha re of st ud en ts co ns um ing SS Bs . T he re wa s i nd ica tio n t ha t r ed uc ed av ail ab ilit y o f SS B i n s ch oo ls m ay be in cre as e c om pe ns at or y SS B c on su m pt ion ou tsi de sc ho ol am on g o lde r stu de nt s b ut m ay no t a ffe ct bo dy di ssa tis fac tio n or di et ing be ha vio ur. Th er e w as al so ve ry lo w- ce rta int y e vid en ce fr om 4 stu die s t ha t i m pr ov ed av ail ab ilit y o f d rin kin g w at er in sc ho ols m ay be as so cia te d w ith de cre as ed SS B i nt ak e; 2 o ut of 4 stu die s r ep or te d s ign ifi ca nt eff ec t o n t ot al int ak e o f S SB or su ga r-s we et en ed m ilk an d 2 ou t o f 3 th at al so m ea su re d w eig ht -re lat ed ou tco m es re po rte d s ign ifi ca nt eff ec ts. Th er e wa s i nd ica tio n t ha t i m pr ov ed w at er av ail ab ilit y m ay de cre as e t ot al m ilk se lec tio n. Th er e w as low - t o m od er at e- ce rta int y e vid en ce fr om 3 stu die s t ha t s m all pr ize s a nd to ke n r ew ar ds fo r th e s ele cti on of pl ain m ilk in el em en ta ry sc ho ol ca fet er ias ar e a sso cia te d w ith de cre as ed se lec tio n of su ga r-s we et en ed m ilk ; a ll o f t he se 3 stu die s re po rte d s ign ifi ca nt eff ec t o n n um be r o f m ea ls wi th un he alt hy be ve ra ge s s ele cte d o r c ho co lat e m ilk pu rch as es . 157ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Th er e w as in dic at ion th at to ta l m ilk se lec tio n m ay be aff ec te d t ho ug h t he di re cti on of re po rte d eff ec ts va rie d, an d t ha t t he in te rv en tio n m ay be as so cia te d w ith an in cre as e i n f oo d w as te . Th er e w as lo w- ce rta int y e vid en ce fr om 1 stu dy th at im pr ov ed pl ac em en t o f p lai n m ilk in sc ho ol ca fet er ias is no t a sso cia te d w ith th e s ele cti on an d co ns um pt ion of su ga r-s we et en ed an d p lai n m ilk , wi th no si gn ifi ca nt eff ec ts re po rte d. Th er e w as ve ry lo w- ce rta int y e vid en ce fr om 2 stu die s t ha t fru it pr ov isi on in sc ho ols m ay be as so cia te d w ith de cre as ed SS B c on su m pt ion by st ud en ts; bo th stu die s r ep or te d s ign ifi ca nt eff ec t. T he re vie w als o c on sid er ed ot he r a cti on ar ea s t o r ed uc e S SB int ak e i nc lud ing on e s tu dy th at fo un d p os iti ve eff ec t o f e m ot ico n ( sm ile y f ac e) la be ls on pl ain m ilk an d s m all pr ize s i n r ew ar d f or se lec tin g s uc h m ilk ov er su ga r-s we et en ed m ilk (l ow -c er ta int y ev ide nc e) . T he au th or s n ot ed th at th is m ay dr ive up fo od w as te if ch ild re n c ho os e t he pl ain m ilk bu t d o n ot dr ink it. An ot he r a na lys is co nc er ne d th e h ea lth ier se lec tio ns in ve nd ing m ac hin es in wo rk pla ce s a nd sc ho ols , w he re th e o ne st ud y i n sch oo ls re po rte d s ign ifi ca nt eff ec t o n S SB in ta ke . W an g a nd St ew ar t, 20 13 (7 2) 1.1 Th is sy ste m at ic re vie w of 19 nu tri tio n p ro m ot ion pr og ra m m es us ing th e H PS ap pr oa ch fo un d t ha t i t c an im pr ov e p ar tic ipa nt s’ d iet s an d h elp de ve lop re lat ed he alt hy be ha vio ur s s uc h a s hy gie ne an d f oo d s afe ty. Th e r ev iew in clu de d 1 9 stu die s p ub lis he d b et we en 19 93 an d 2 01 1, of w hic h 8 we re RC Ts, 4 we re co nt ro lle d tri als , 6 w er e p re - a nd po st- te st de sig n a nd 1 wa s a n int er ru pt ed ti m e s er ies . In te rv en tio ns th at us ed on e o r m or e o f t he H PS fra m ew or k a re as (i .e. sc ho ol po lic y, cu rri cu lum , fa m ily / co m m un ity in vo lve m en t). Of th e 1 9 s tu die s, 3 f oc us ed on nu tri tio n p oli cy , 6 on nu tri tio n e du ca tio n a nd 10 us ed a ho lis tic H PS ap pr oa ch . Ch ild re n a nd ad ole sce nt s a t pr im ar y a nd se co nd ar y s ch oo ls ar ou nd th e w or ld. Th e a ut ho rs re po rte d t ha t n ut rit ion pr om ot ion pr og ra m m es us ing a fu ll o r p ar tia l H PS ap pr oa ch ca n i nc re as e d iet an d n ut rit ion kn ow led ge , at tit ud e a nd sk ills , a nd be ha vio ur. Th e au th or s n ot ed im pr ov em en ts in pa rti cip an ts’ co ns um pt ion of hi gh -fi br e f oo ds , h ea lth ier sn ac ks , w at er, m ilk an d F &V . It ca n a lso re du ce pa rti cip an ts’ “b re ak fas t s kip pin g,” re du ce co ns um pt ion of nu tri en t-p oo r f oo ds in clu din g fat ty fo od s a nd sw ee t d rin ks an d r ed uc e e at ing dis or de rs. Fu rth er m or e, th es e p ro gr am m es m ay he lp to de ve lop hy gie nic ha bit s a nd im pr ov ed fo od sa fet y b eh av iou rs. 158 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Th er e w as in dic at ion th at to ta l m ilk se lec tio n m ay be aff ec te d t ho ug h t he di re cti on of re po rte d eff ec ts va rie d, an d t ha t t he in te rv en tio n m ay be as so cia te d w ith an in cre as e i n f oo d w as te . Th er e w as lo w- ce rta int y e vid en ce fr om 1 stu dy th at im pr ov ed pl ac em en t o f p lai n m ilk in sc ho ol ca fet er ias is no t a sso cia te d w ith th e s ele cti on an d co ns um pt ion of su ga r-s we et en ed an d p lai n m ilk , wi th no si gn ifi ca nt eff ec ts re po rte d. Th er e w as ve ry lo w- ce rta int y e vid en ce fr om 2 stu die s t ha t fru it pr ov isi on in sc ho ols m ay be as so cia te d w ith de cre as ed SS B c on su m pt ion by st ud en ts; bo th stu die s r ep or te d s ign ifi ca nt eff ec t. T he re vie w als o c on sid er ed ot he r a cti on ar ea s t o r ed uc e S SB int ak e i nc lud ing on e s tu dy th at fo un d p os iti ve eff ec t o f e m ot ico n ( sm ile y f ac e) la be ls on pl ain m ilk an d s m all pr ize s i n r ew ar d f or se lec tin g s uc h m ilk ov er su ga r-s we et en ed m ilk (l ow -c er ta int y ev ide nc e) . T he au th or s n ot ed th at th is m ay dr ive up fo od w as te if ch ild re n c ho os e t he pl ain m ilk bu t d o n ot dr ink it. An ot he r a na lys is co nc er ne d th e h ea lth ier se lec tio ns in ve nd ing m ac hin es in wo rk pla ce s a nd sc ho ols , w he re th e o ne st ud y i n sch oo ls re po rte d s ign ifi ca nt eff ec t o n S SB in ta ke . W an g a nd St ew ar t, 20 13 (7 2) 1.1 Th is sy ste m at ic re vie w of 19 nu tri tio n p ro m ot ion pr og ra m m es us ing th e H PS ap pr oa ch fo un d t ha t i t c an im pr ov e p ar tic ipa nt s’ d iet s an d h elp de ve lop re lat ed he alt hy be ha vio ur s s uc h a s hy gie ne an d f oo d s afe ty. Th e r ev iew in clu de d 1 9 stu die s p ub lis he d b et we en 19 93 an d 2 01 1, of w hic h 8 we re RC Ts, 4 we re co nt ro lle d tri als , 6 w er e p re - a nd po st- te st de sig n a nd 1 wa s a n int er ru pt ed ti m e s er ies . In te rv en tio ns th at us ed on e o r m or e o f t he H PS fra m ew or k a re as (i .e. sc ho ol po lic y, cu rri cu lum , fa m ily / co m m un ity in vo lve m en t). Of th e 1 9 s tu die s, 3 f oc us ed on nu tri tio n p oli cy , 6 on nu tri tio n e du ca tio n a nd 10 us ed a ho lis tic H PS ap pr oa ch . Ch ild re n a nd ad ole sce nt s a t pr im ar y a nd se co nd ar y s ch oo ls ar ou nd th e w or ld. Th e a ut ho rs re po rte d t ha t n ut rit ion pr om ot ion pr og ra m m es us ing a fu ll o r p ar tia l H PS ap pr oa ch ca n i nc re as e d iet an d n ut rit ion kn ow led ge , at tit ud e a nd sk ills , a nd be ha vio ur. Th e au th or s n ot ed im pr ov em en ts in pa rti cip an ts’ co ns um pt ion of hi gh -fi br e f oo ds , h ea lth ier sn ac ks , w at er, m ilk an d F &V . It ca n a lso re du ce pa rti cip an ts’ “b re ak fas t s kip pin g,” re du ce co ns um pt ion of nu tri en t-p oo r f oo ds in clu din g fat ty fo od s a nd sw ee t d rin ks an d r ed uc e e at ing dis or de rs. Fu rth er m or e, th es e p ro gr am m es m ay he lp to de ve lop hy gie nic ha bit s a nd im pr ov ed fo od sa fet y b eh av iou rs. AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W an g e t a l., 20 13 (5 8) 1.1 , 2 .2, 3. 2 Th is co m pa ra tiv e eff ec tiv en es s r ev iew an d m et a- an aly sis of 12 4 o be sit y pr ev en tio n i nt er ve nt ion s, of wh ich 10 4 w er e s ch oo l- ba se d, fo un d g re at er st re ng th of ev ide nc e f or co m bin ed di et an d P A i nt er ve nt ion s t ha n fo r d iet or PA in te rv en tio ns alo ne . Th e r ev iew in clu de d 1 24 int er ve nt ion st ud ies de scr ibe d i n 1 31 ar tic les , co ns ist ing of RC Ts, qu as i- ex pe rim en ta l s tu die s a nd na tu ra l e xp er im en ts. In te rv en tio ns in clu din g m od ifi ca tio n o f d iet , m od ifi ca tio n o f P A o r se de nt ar y a cti vit y o r a co m bin at ion of th es e. Ch ild re n a nd ad ole sce nt s ag ed 2– 18 ye ar s i n t he U nit ed St at es . O f t he 12 4 s tu die s, 10 4 we re sc ho ol- ba se d. Ov er ha lf o f t he sc ho ol- ba se d i nt er ve nt ion s re po rte d s ta tis tic all y s ign ifi ca nt be ne fic ial int er ve nt ion eff ec t i n a t l ea st so m e w eig ht - re lat ed ou tco m es (B M I, B M I z -sc or e, pr ev ale nc e of ov er we igh t a nd ob es ity , w ais t c irc um fer en ce , sk inf old th ick ne ss an d p er ce nt bo dy fa t). In th e an aly sis of se tti ng an d i nt er ve nt ion co m po ne nt s, hig h s tre ng th of ev ide nc e f or th e b en efi ts of sch oo l-b as ed ap pr oa ch es w as fo un d f or PA al on e wi th a ho m e c om po ne nt , a nd co m bin ed di et an d PA w ith a ho m e a nd a co m m un ity co m po ne nt , wh ile m od er at e s tre ng th of ev ide nc e w as fo un d fo r d iet al on e, PA al on e, an d c om bin ed di et an d PA w ith a ho m e o r a co m m un ity co m po ne nt . Th e a ut ho rs no te d t ha t d es pit e t he la rg e n um be r of st ud ies co m bin ing di et an d P A i nt er ve nt ion s in sch oo ls, th e e vid en ce re m ain s i ns uffi cie nt as re su lt of in co ns ist en cy ac ro ss stu die s. On ly few sc ho ol- ba se d c om bin ed di et an d P A s tu die s inc lud ed su ffi cie nt da ta an d h om og en eit y f or m et a- an aly sis : 4 st ud ies sh ow ed im pr ov ed BM I z-s co re (- 0.0 8; 95 % CI : - 0.1 4, -0 .02 ; P = 0.0 09 ) wh er ea s 7 st ud ies sh ow ed im pr ov ed BM I ( -0 .32 kg /m 2; 95 % CI : - 0.4 9, -0 .16 ; P < 0.0 01 ). An ot he r 8 sch oo l-b as ed co m bin ed di et an d P A s tu die s w ith a h om e c om po ne nt di d n ot sh ow im pr ov ed BM I in m et a- an aly se s. Co m pr eh en siv e i nt er ve nt ion s ad dr es sin g t he en vir on m en t ( e.g . m od ifi ed fo od an d b ev er ag e i te m s o ffe re d i n s ch oo l c afe te ria or str uc tu ra l c ha ng es in sc ho ol PA ) a nd in div idu als ' kn ow led ge an d a tti tu de s ( e.g . n ut rit ion ed uc at ion ) w er e m or e l ike ly to be su cce ssf ul th an th os e a dd re ssi ng ei th er on e a lon e. Ed uc at ion al int er ve nt ion s w er e l es s l ike ly to be eff ec tiv e t ha n en vir on m en ta l c ha ng es . 159ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W an g e t a l., 20 15 (9 3) 2.2 Th is sy ste m at ic re vie w an d m et a- an aly sis of 13 9 o be sit y pr ev en tio n s tu die s, 11 5 o f wh ich w er e s ch oo l-b as ed , fo un d t ha t t he re is at le as t m od er at e e vid en ce th at su ch pr og ra m m es re du ce BM I, es pe cia lly m ult ico m po ne nt str at eg ies th at in clu de ho m e inv olv em en t. Th e r ev iew in clu de d 1 39 int er ve nt ion st ud ies pu bli sh ed in 14 7 a rti cle s be tw ee n 1 98 5 a nd 20 14 , wh ich in clu de d R CT s, qu as i- ex pe rim en ta l s tu die s a nd na tu ra l e xp er im en ts. St ud ies ev alu at ed co m bin ed die t a nd PA in te rv en tio ns (1 04 st ud ies ), PA on ly (2 8 stu die s) an d d iet on ly (7 stu die s). Ch ild re n a ge d 2 –1 8 y ea rs, in HI C, es pe cia lly th e U nit ed St at es . O f t he 13 9 s tu die s (co ve rin g < 10 00 00 ch ild re n) , 11 5 w er e m ain ly ca rri ed ou t in th e s ch oo l s et tin g a nd 24 m ain ly in ot he r s et tin gs (h om e, pr im ar y h ea lth ca re fa cil iti es , c hil d c ar e o r co m m un iti es ). Th er e w as at le as t m od er at e e vid en ce to su pp or t th e e ffe cti ve ne ss of sc ho ol- ba se d i nt er ve nt ion s. Ov er all , s tu die s o f m ult ipl e s et tin g i nt er ve nt ion s ha d b en efi cia l a nd si gn ifi ca nt re su lts m or e o fte n th an si ng le se tti ng in te rv en tio ns . S ta tis tic all y sig nifi ca nt eff ec ts we re re po rte d b y a bo ut ha lf of th e s ch oo l-b as ed st ud ies , e sp ec ial ly th os e wi th a ho m e i nv olv em en t c om po ne nt . M et a- an aly se s w er e p er fo rm ed w he n m or e t ha n 3 co m pa ra ble st ud ies w er e a va ila ble fo r a gi ve n int er ve nt ion st ra te gy an d s et tin g. Th e p oo led re su lts w er e s ign ifi ca nt fo r s ch oo l-b as ed on ly int er ve nt ion s f or BM I z -sc or e ( -0 .05 ; 5 st ud ies ; P= 0.0 25 ) a nd BM I ( M D: -0 .3 kg /m 2; 9 s tu die s; P< 0.0 01 ), bu t i ns ign ifi ca nt fo r s ch oo l-b as ed int er ve nt ion s c om bin ing di et an d P A w ith a h om e c om po ne nt . T he po ole d e ffe ct siz es es tim at ed in th es e m et a- an aly se s w er e s til l sm all co m pa re d t o t he re ce nt in cre as e i n B M I an d o ve rw eig ht ra te s i n m an y c ou nt rie s. T he au th or s c on clu de d t ha t t he st re ng th of ev ide nc e wa s h igh fo r P A- on ly int er ve nt ion s d eli ve re d i n sch oo ls wi th ho m e i nv olv em en t o r c om bin ed die t/P A i nt er ve nt ion s d eli ve re d i n s ch oo ls wi th bo th ho m e a nd co m m un ity co m po ne nt s. St re ng th of ev ide nc e w as m od er at e f or sc ho ol- ba se d i nt er ve nt ion s t ar ge tin g e ith er di et or PA , co m bin ed in te rv en tio ns de liv er ed in sc ho ols w ith ho m e o r c om m un ity co m po ne nt s o r c om bin ed int er ve nt ion s d eli ve re d i n t he co m m un ity w ith a sch oo l c om po ne nt . 160 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W an g e t a l., 20 18 (1 12 ) 3.1 Th is sy ste m at ic re vie w an d m et a- an aly sis of 30 st ud ies of ed uc at ion al int er ve nt ion s on he alt h f ou nd po sit ive eff ec ts on BM I, B M I z -sc or e, wa ist ci rcu m fer en ce , tri ce ps sk inf old , S BP , t ot al ch ole ste ro l, a nd tr igl yc er ide s. Th e r ev iew in clu de d 3 0 R CT s, pu bli sh ed be tw ee n 1 98 9 an d 2 01 6. He alt h e du ca tio n int er ve nt ion s t o p re ve nt ob es ity , a s s ing le or m ult ico m po ne nt in te rv en tio n wi th a du ra tio n o f 6 –7 2 m on th s. Ch ild re n ( n= 35 29 6) , m ea n ag e r an ge d f ro m 2. 5 t o 1 1.8 ye ar s. Se ve nt ee n s tu die s we re fr om Eu ro pe , 1 1 w er e fro m Am er ica an d 2 w er e m ult i-c ou nt ry st ud ies . A la rg e m ajo rit y o f s tu die s a pp ea r to ha ve be en co nd uc te d i n sch oo ls. Ed uc at ion al int er ve nt ion s o n h ea lth ha d sig nifi ca nt eff ec ts on th e w eig ht ed m ea n diff er en ce of BM I ( -0 .15 kg /m 2; 95 % CI : - 0.2 4, -0 .05 ; P = 0.0 03 ), BM I z -sc or e ( -0 .03 ; 9 5% CI : -0 .05 , - 0.0 2; P< 0.0 01 ), wa ist ci rcu m fer en ce (-0 .97 ; 9 5% CI : - 1.9 5, -0 .00 ; P = 0.0 50 ), tri ce ps sk inf old (- 1.3 9; 95 % CI : - 2.4 1, -0 .37 ; P = 0.0 08 ), SB P ( -1 .13 ; 9 5% CI : - 2.2 0, -0 .07 ; P = 0.0 37 ), to ta l c ho les te ro l ( -4 .04 ; 9 5% CI : - 7.1 8. -0 .90 ; P= 0. 01 2) an d t rig lyc er ide (- 2.6 2; 95 % CI : -4 .33 , - 0.9 0, P= 0. 00 3) . H ow ev er, ed uc at ion al int er ve nt ion s w er e f ou nd to ha ve lit tle or no sig nifi ca nt im pa ct on th e w ais t-t o- hip ra tio , DB P, hig h- de ns ity lip op ro te in an d l ow -d en sit y lip op ro te in. W ar d e t a l., 20 17 (1 00 ) 2.2 Th is sy ste m at ic re vie w of 43 pr es ch oo l-b as ed ob es ity pr ev en tio n s tu die s f ou nd te nt at ive ev ide nc e t ha t m ult ico m po ne nt , m ult i-l ev el int er ve nt ion s w ith pa re nt al en ga ge m en t a re m os t lik ely to be eff ec tiv e w ith an th ro po m et ric ou tco m es . Th e r ev iew in clu de d 4 3 stu die s o f v ar iou s d es ign s, inc lud ing RC Ts, no n- ex pe rim en ta l p re - a nd po st- te st de sig n, w ith in su bje ct cro ss- ov er an d l on git ud ina l fo llo w- up s t o R CT s, pu bli sh ed be tw ee n 2 01 0 a nd 20 15 . In te rv en tio ns in clu de d str at eg ies to im pr ov e h ea lth y ea tin g ( e.g . m en u c ha ng es , nu tri tio n e du ca tio n, ch an gin g m ea l s er vic e a pp ro ac he s, an d f oo d t as tin gs ) a nd /o r PA , w ith or w ith ou t p ar en ta l inv olv em en t ( e.g . w or ks ho ps , co ok ing cl as se s, ne ws let te rs, CD s, ha nd ou ts) . Ch ild re n a ge d 2 –6 ye ar s in pr es ch oo ls in th e U nit ed St at es , E ur op e, Au str ali a, as we ll a s C hil e. Th irt ee n o f 1 8 s tu die s i nc lud ing a die ta ry int ak e m ea su re an d 1 0 o f 2 4 s tu die s w ith an an th ro po m et ric m ea su re de m on str at ed at le as t on e s uc ce ssf ul int er ve nt ion eff ec t. T o i nv es tig at e th e e ffe ct of in div idu al int er ve nt ion el em en ts, an in te rv en tio n s tre ng th w as de fin ed us ing ap pr oa ch es co m m on in co m m un ity he alt h eff or ts fo r o be sit y ( i.e . a s a co m po sit e o f t he nu m be r o f in te rv en tio n s tra te gie s u se d) , t he ir po te nt ial im pa ct (h igh fo r p oli cy /e nv iro nm en t, low fo r e du ca tio n) , a nd th e f re qu en cy an d du ra tio n ( hig he r t he m or e f re qu en t e xp os ur es or lo ng er du ra tio n) of th eir us e. Th is str en gt h wa s p os iti ve ly co rre lat ed w ith re po rti ng of po sit ive an th ro po m et ric ou tco m es fo r P A, di et an d c om bin ed in te rv en tio ns . F ur th er m or e, pa re nt en ga ge m en t c om po ne nt s i nc re as ed th e str en gt h o f t he se re lat ion sh ips . T hu s, th e a ut ho rs co nc lud ed th at te nt at ive ev ide nc e e xis ts th at m ult ico m po ne nt , m ult i-l ev el int er ve nt ion s w ith pa re nt al en ga ge m en t w er e m os t l ike ly to be eff ec tiv e w ith an th ro po m et ric ou tco m es . 161ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W at er s e t a l., 20 11 (4 7) 1.1 , 3 .2 Th is re vie w an d m et a- an aly sis of 55 pr og ra m m es to pr ev en t c hil dh oo d o be sit y, 43 of w hic h w er e s ch oo l-b as ed , fo un d s tro ng ev ide nc e t o su pp or t b en efi cia l e ffe cts of ch ild ob es ity pr ev en tio n pr og ra m m es on BM I, pa rti cu lar ly fo r p ro gr am m es ta rg et ed at ch ild re n a ge d 6– 12 ye ar s. Th e r ev iew in clu de d 5 5 stu die s t ha t u se d a co nt ro lle d stu dy de sig n ( wi th or w ith ou t ra nd om iza tio n) , 3 7 o f w hic h we re in clu de d i n t he m et a- an aly sis . T he st ud ies w er e pu bli sh ed be tw ee n 1 99 3 an d 2 00 9. In te rv en tio ns or pr og ra m m es th at in vo lve d d iet an d nu tri tio n, ex er cis e a nd PA , lif es ty le an d s oc ial su pp or t. Nu tri tio n i nt er ve nt ion s we re la rg ely be ha vio ur al, fo cu sin g o n c ur ric ulu m or ed uc at ion /m es sa ge s, alt ho ug h s om e s tu die s a lso inc lud ed en vir on m en ta l int er ve nt ion s ( e.g . a dd re ssi ng fo od se rv ice ) a nd /o r se rv ice -le ve l in te rv en tio ns inv olv ing sc ho ol nu rse s. In te rv en tio ns w ith in th e co m m un ity , s ch oo l a nd ca re ou tsi de of sc ho ol ho ur s, ho m e, ch ild ca re or pr es ch oo l/ nu rse ry /k ind er ga rte n w er e eli gib le. Th ey in clu de d stu die s t ha t c om pa re d d iet or PA in te rv en tio ns , o r b ot h wi th a no n- int er ve nt ion co nt ro l g ro up w ho re ce ive d us ua l c ar e o r a no th er ac tiv e int er ve nt ion . Ch ild re n a ge d 0 –1 8 y ea rs in va rio us se tti ng s, m os tly sch oo l-b as ed (4 3 o f 5 5 stu die s), bu t a lso ho m e, co m m un ity , h ea lth se rv ice an d c ar e s et tin gs . M os t s tu die s to ok pl ac e i n H IC, es pe cia lly th e U nit ed St at es an d E ur op e, bu t a lso U M IC an d L M IC. A m et a- an aly sis of 37 st ud ies (n = 27 94 6 ch ild re n) sh ow ed si gn ifi ca nt re du cti on s i n ad ipo sit y ( m ea su re d a s B M I o r B M I z -sc or e) (S M D: -0 .15 ; 9 5% CI : - 0.2 1, -0 .09 ). Su bg ro up an aly se s by se tti ng sh ow ed th e g re at es t o ve ra ll e ffe ct fo r e du ca tio n o nl y ( SM D: -0 .14 ; 9 5% CI : - 0.2 1, -0 .08 ) c om pa re d t o e du ca tio n p lus ot he r s et tin g an d n on -e du ca tio n s et tin g. Su bg ro up an aly se s by in te rv en tio n t yp e s ho we d t he gr ea te st ov er all eff ec t f or co m bin ed PA an d d iet ar y i nt er ve nt ion s (S M D: -0 .18 ; 9 5% CI : - 0.2 7, -0 .09 ) c om pa re d t o PA al on e ( SM D: -0 .11 ; 9 5% CI: -0 .19 , - 0.0 2) w ith no si gn ifi ca nt eff ec t f or di et al on e. 162 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W at kin s e t a l., 20 15 (1 54 ) 4.1 Th is lit er at ur e r ev iew of nu tri tio n a sp ec ts of 10 sc ho ol fee din g p ro gr am m e s tu die s co nc lud ed th at th es e c an be eff ec tiv e i n i m pr ov ing an th ro po m et ric an d bio ch em ica l n ut rit ion -re lat ed ou tco m es . Th e r ev iew in clu de d 1 0 R CT s an d C BA st ud ies pu bli sh ed be tw ee n 1 96 2 a nd 20 11 . In -sc ho ol m ea ls an d t ak e- ho m e r at ion s ( e.g . s ch oo l lun ch or br ea kfa st, m ilk su pp lem en t, m ea t o r o th er loc all y a va ila ble fo od s t ha t ar e r ich in en er gy /p ro te in as su pp lem en ts) . Pr im ar y s ch oo l c hil dr en in low - a nd m idd le- inc om e co un tri es in Af ric a, As ia an d La tin Am er ica Am on g t he 10 st ud ies m ea su rin g a nt hr op om et ric ou tco m es , s m all bu t s ign ifi ca nt im pr ov em en ts we re se en in he igh t/s tu nt ing in 7 of th e s tu die s an d o n w eig ht /u nd er we igh t i n 6 of th e s tu die s. Ne ar ly all of th e f oo d s up ple m en ta tio n s tu die s th at re po rte d s ign ifi ca nt he igh t a nd w eig ht ga ins in clu de d a n a nim al- ba se d p ro du ct, no t us ua lly in clu de d i n s ch oo l fe ed ing pr og ra m m es in low -in co m e c ou nt rie s. Sp ill- ov er be ne fit s w er e ob se rv ed fo r y ou ng er si bli ng s i nd ica tin g t ha t sch oo l fe ed ing co uld ha ve in im po rta nt ro le in pr om ot ing th e h ea lth of th e n ex t g en er at ion . In co rp or at ing fin din gs fr om ot he r r ela te d sy ste m at ic re vie ws , t he au th or s c on clu de d t ha t th e l ar ge st eff ec ts we re se en fo r in -sc ho ol m ea ls on he igh t/s tu nt ing , w eig ht /u nd er we igh t a nd fo r M M N fo rti fic at ion on iro n s ta tu s, an ae m ia/ ha em og lob in, vi ta m in A s ta tu s a nd zi nc st at us . Ta ke -h om e r at ion s, M M N fo rti fic at ion or po wd er ha d s om e e ffe ct on he igh t/s tu nt ing an d w eig ht / un de rw eig ht , w he re as in -sc ho ol m ea ls an d M M N po wd er s h ad so m e e ffe ct on va rio us m icr on ut rie nt -re lat ed ou tco m es . 163ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W hit te m or e e t a l., 20 13 (1 27 ) 3.1 , 3 .3 Th is sy ste m at ic re vie w of 12 sc ho ol- ba se d o be sit y pr ev en tio n p ro gr am m es us ing In te rn et -b as ed int er ve nt ion s f ou nd th at su ch pr og ra m m es w er e e ffe cti ve at im pr ov ing he alt h b eh av iou rs of ad ole sce nt s, at le as t i n t he sh or t t er m . Th e r ev iew in clu de d 1 2 em pir ica l s tu die s w ith a co m pa ris on gr ou p, pu bli sh ed be tw ee n 1 99 9 a nd 20 12 . In te rn et -b as ed ob es ity pr ev en tio n p ro gr am m es inc lud ing co nt en t o n nu tri tio n a nd PA (a lon e o r co m bin ed ); 5 c om pa re d a sch oo l-b as ed In te rn et ob es ity pr ev en tio n p ro gr am m e t o a no -tr ea tm en t c on tro l g ro up , 3 stu die s c om pa re d a n I nt er ne t pr og ra m m e t o t ra dit ion al cla ssr oo m ed uc at ion , 2 s tu die s c om pa re d a n In te rn et pr og ra m m e t o a p rin t p ro gr am , a nd 2 co m pa re d 2 di ffe re nt In te rn et pr og ra m m es . Ad ole sce nt s a ge d 1 2– 18 ye ar s (m ea n a ge 14 .7) at sc ho ols in th e U nit ed St at es an d E ur op e Ov er all , s ch oo l-b as ed In te rn et ob es ity pr ev en tio n pr og ra m m es w er e e ffe cti ve in im pr ov ing he alt h be ha vio ur s o f a do les ce nt s i n t he sh or t t er m (< 3– 6 m on th s). Im pr ov em en ts in se lf- re po rte d die ta ry be ha vio ur (e .g. in ta ke of SS B, F& V, fas t fo od , fi br e o r f at , s kip pin g m ea ls) an d P A w er e re po rte d f or th e m ajo rit y o f p ro gr am m es (1 0 pr og ra m m es , 8 3% ), bu t o nl y 7 pr og ra m m es (5 8% ) d em on str at ed po sit ive ou tco m es in th e I nt er ne t g ro up co m pa re d t o t he co nt ro l. Im pr ov em en ts in ad ole sce nt s’ s elf -e ffi ca cy fo r h ea lth y e at ing or be ing ph ys ica lly ac tiv e we re re po rte d i n p ro gr am m es th at ta rg et ed se lf- effi ca cy (3 pr og ra m m es ). Of th e 4 st ud ies in wh ich th e e ffe ct on BM I w as ev alu at ed , o nl y 1 s tu dy sh ow ed a sig nifi ca nt de cre as e. In 2 of th e 3 st ud ies th at co m pa re d t he In te rn et ob es ity pr ev en tio n p ro gr am m e t o t ra dit ion al ed uc at ion , th e I nt er ne t i nt er ve nt ion ha d a gr ea te r e ffe ct on se lec t h ea lth be ha vio ur s i nc lud ing in cre as ed PA an d d ec re as ed BM I. W illi am s e t a l., 20 13 (6 7) 1.1 , 2 .2, 4. 1 Th is m et a- an aly sis of 21 sch oo l-b as ed di et - a nd PA -re lat ed pr og ra m m es fo un d t ha t t he m ult ifa ce te d pr og ra m m es ga ve si gn ifi ca nt re du cti on s i n w eig ht -re lat ed ou tco m es . Th e r ev iew in clu de d 2 1 stu die s ( RC Ts, CB A s tu die s, co ho rts an d c ro ss- se cti on al stu die s) pu bli sh ed in 20 03 or la te r. Of th e 2 1 s tu die s, 10 fo cu se d on di et (e .g. sc ho ol lun ch es , br ea kfa st pr og ram m es , s ch oo l nu tri tio n g uid eli ne s), 5 on PA (e .g. PE , w alk ing sc ho ol bu s s ch em e) an d 6 on bo th pr og ram m e a rea s. Of th e 16 di et -re lat ed st ud ies , 1 3 foc us ed on po lic ies , o f w hic h 8 e va lua ted sc ho ol lun ch or br ea kfa st pr og ram m es ta rg et ing ch ild ren at ris k of un de rn ut rit ion an d 5 ev alu ate d p oli cie s r ela ted to in cre as ing av ail ab ilit y of he alt hy fo od s a nd be ve rag es or de cre as ing th at of un he alt hy fo od s a nd be ve rag es w ith in sch oo ls. Ch ild re n a ge d 4 –1 1 y ea rs in pr im ar y s ch oo ls, m os tly in th e U nit ed St at es bu t a lso in Au str ali a, Ca na da , It aly , M ex ico an d t he U nit ed Ki ng do m . Th e p oo led re su lts of di et -re lat ed pr og ram m es we re m ixe d. Pa rti cip ati on in th e l un ch pr og ram m e ga ve a no n- sig nifi ca nt ris e i n B MI z- sco re, w hil e pa rti cip ati on in th e s ch oo l b rea kfa st pr og ram m e ga ve si gn ific an tly lo we r B MI z- sco re (-0 .08 ; 9 5% CI: -0 .14 3, -0 .01 7; 5 s tu die s). Ex po su re to ot he r die t-r ela ted po lic ies ta rg et ing th e a va ila bil ity of foo ds in sc ho ols ga ve a no n- sig nifi ca nt de cre as e i n B MI z- sco re. Th e p oo led re su lts of PA pr og ram m es sh ow ed a sm all bu t n on -si gn ific an t de cre as e i n B MI z- sco re. Th e h ete ro ge ne ity of co m bin ed di et an d P A p ro gr am m es pr ev en ted fu rth er m et a- an aly sis , b ut th es e p ro gr am m es de m on str ate d p ro m isi ng re su lts , p ar tic ula rly th os e w hic h i nv olv ed st ak eh old er inv olv em en t an d e ng ag em en t w ith fa m ilie s. T he au th or s co nc lud e t ha t d iet - a nd PA -re lat ed po lic ies ne ed to be si tu ate d w ith in m or e c om ple x a pp ro ac he s t o pr ev en tin g o be sit y a nd to fo cu s o n m ult ipl e f ac to rs an d l ev els of in flu en ce . 164 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W hit te m or e e t a l., 20 13 (1 27 ) 3.1 , 3 .3 Th is sy ste m at ic re vie w of 12 sc ho ol- ba se d o be sit y pr ev en tio n p ro gr am m es us ing In te rn et -b as ed int er ve nt ion s f ou nd th at su ch pr og ra m m es w er e e ffe cti ve at im pr ov ing he alt h b eh av iou rs of ad ole sce nt s, at le as t i n t he sh or t t er m . Th e r ev iew in clu de d 1 2 em pir ica l s tu die s w ith a co m pa ris on gr ou p, pu bli sh ed be tw ee n 1 99 9 a nd 20 12 . In te rn et -b as ed ob es ity pr ev en tio n p ro gr am m es inc lud ing co nt en t o n nu tri tio n a nd PA (a lon e o r co m bin ed ); 5 c om pa re d a sch oo l-b as ed In te rn et ob es ity pr ev en tio n p ro gr am m e t o a no -tr ea tm en t c on tro l g ro up , 3 stu die s c om pa re d a n I nt er ne t pr og ra m m e t o t ra dit ion al cla ssr oo m ed uc at ion , 2 s tu die s c om pa re d a n In te rn et pr og ra m m e t o a p rin t p ro gr am , a nd 2 co m pa re d 2 di ffe re nt In te rn et pr og ra m m es . Ad ole sce nt s a ge d 1 2– 18 ye ar s (m ea n a ge 14 .7) at sc ho ols in th e U nit ed St at es an d E ur op e Ov er all , s ch oo l-b as ed In te rn et ob es ity pr ev en tio n pr og ra m m es w er e e ffe cti ve in im pr ov ing he alt h be ha vio ur s o f a do les ce nt s i n t he sh or t t er m (< 3– 6 m on th s). Im pr ov em en ts in se lf- re po rte d die ta ry be ha vio ur (e .g. in ta ke of SS B, F& V, fas t fo od , fi br e o r f at , s kip pin g m ea ls) an d P A w er e re po rte d f or th e m ajo rit y o f p ro gr am m es (1 0 pr og ra m m es , 8 3% ), bu t o nl y 7 pr og ra m m es (5 8% ) d em on str at ed po sit ive ou tco m es in th e I nt er ne t g ro up co m pa re d t o t he co nt ro l. Im pr ov em en ts in ad ole sce nt s’ s elf -e ffi ca cy fo r h ea lth y e at ing or be ing ph ys ica lly ac tiv e we re re po rte d i n p ro gr am m es th at ta rg et ed se lf- effi ca cy (3 pr og ra m m es ). Of th e 4 st ud ies in wh ich th e e ffe ct on BM I w as ev alu at ed , o nl y 1 s tu dy sh ow ed a sig nifi ca nt de cre as e. In 2 of th e 3 st ud ies th at co m pa re d t he In te rn et ob es ity pr ev en tio n p ro gr am m e t o t ra dit ion al ed uc at ion , th e I nt er ne t i nt er ve nt ion ha d a gr ea te r e ffe ct on se lec t h ea lth be ha vio ur s i nc lud ing in cre as ed PA an d d ec re as ed BM I. W illi am s e t a l., 20 13 (6 7) 1.1 , 2 .2, 4. 1 Th is m et a- an aly sis of 21 sch oo l-b as ed di et - a nd PA -re lat ed pr og ra m m es fo un d t ha t t he m ult ifa ce te d pr og ra m m es ga ve si gn ifi ca nt re du cti on s i n w eig ht -re lat ed ou tco m es . Th e r ev iew in clu de d 2 1 stu die s ( RC Ts, CB A s tu die s, co ho rts an d c ro ss- se cti on al stu die s) pu bli sh ed in 20 03 or la te r. Of th e 2 1 s tu die s, 10 fo cu se d on di et (e .g. sc ho ol lun ch es , br ea kfa st pr og ram m es , s ch oo l nu tri tio n g uid eli ne s), 5 on PA (e .g. PE , w alk ing sc ho ol bu s s ch em e) an d 6 on bo th pr og ram m e a rea s. Of th e 16 di et -re lat ed st ud ies , 1 3 foc us ed on po lic ies , o f w hic h 8 e va lua ted sc ho ol lun ch or br ea kfa st pr og ram m es ta rg et ing ch ild ren at ris k of un de rn ut rit ion an d 5 ev alu ate d p oli cie s r ela ted to in cre as ing av ail ab ilit y of he alt hy fo od s a nd be ve rag es or de cre as ing th at of un he alt hy fo od s a nd be ve rag es w ith in sch oo ls. Ch ild re n a ge d 4 –1 1 y ea rs in pr im ar y s ch oo ls, m os tly in th e U nit ed St at es bu t a lso in Au str ali a, Ca na da , It aly , M ex ico an d t he U nit ed Ki ng do m . Th e p oo led re su lts of di et -re lat ed pr og ram m es we re m ixe d. Pa rti cip ati on in th e l un ch pr og ram m e ga ve a no n- sig nifi ca nt ris e i n B MI z- sco re, w hil e pa rti cip ati on in th e s ch oo l b rea kfa st pr og ram m e ga ve si gn ific an tly lo we r B MI z- sco re (-0 .08 ; 9 5% CI: -0 .14 3, -0 .01 7; 5 s tu die s). Ex po su re to ot he r die t-r ela ted po lic ies ta rg et ing th e a va ila bil ity of foo ds in sc ho ols ga ve a no n- sig nifi ca nt de cre as e i n B MI z- sco re. Th e p oo led re su lts of PA pr og ram m es sh ow ed a sm all bu t n on -si gn ific an t de cre as e i n B MI z- sco re. Th e h ete ro ge ne ity of co m bin ed di et an d P A p ro gr am m es pr ev en ted fu rth er m et a- an aly sis , b ut th es e p ro gr am m es de m on str ate d p ro m isi ng re su lts , p ar tic ula rly th os e w hic h i nv olv ed st ak eh old er inv olv em en t an d e ng ag em en t w ith fa m ilie s. T he au th or s co nc lud e t ha t d iet - a nd PA -re lat ed po lic ies ne ed to be si tu ate d w ith in m or e c om ple x a pp ro ac he s t o pr ev en tin g o be sit y a nd to fo cu s o n m ult ipl e f ac to rs an d l ev els of in flu en ce . AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W W olf en de n e t a l., 20 16 (1 06 ) 2.3 Th is Co ch ra ne sy ste m at ic re vie w of 10 st ud ies of st ra te gie s a im ed at im pr ov ing im ple m en ta tio n of he alt hy po lic ies , p ra cti ce s or pr og ra m m es in ch ild ca re ce nt re s w as un ab le to fin d su ffi cie nt ev ide nc e t o d ra w co nc lus ion s r eg ar din g t he ir eff ec tiv en es s a t i m pr ov ing im ple m en ta tio n o r im pr ov ing sta ff kn ow led ge , a tti tu de s an d s kil ls. Th e r ev iew in clu de d 1 0 ra nd om ize d a nd no n- ra nd om ize d s tu die s w ith a p ar all el co nt ro l g ro up , pu bli sh ed be tw ee n 2 00 2 an d 2 01 4. In te rv en tio ns to im pr ov e t he im ple m en ta tio n o f c hil d c ar e- ba se d h ea lth y e at ing , P A a nd ob es ity pr ev en tio n p oli cie s, pr ac tic es or pr og ra m m es (fo cu sin g o n h ea lth y e at ing , PA or bo th ) b y s ta ff of ce nt re - ba se d c hil d c ar e s er vic es (e .g. ed uc at ion al m at er ial s, ed uc at ion al m ee tin gs , au dit an d f ee db ac k, op ini on lea de rs, sm all in ce nt ive s o r gr an ts, ed uc at ion al ou tre ac h vis its or ac ad em ic de ta ilin g) . Ch ild re n a ge d < 6 y ea rs (n = 28 29 ) a nd ch ild ca re se rv ice sta ff (n = 45 7) in ce nt re - ba se d c hil d c ar e s er vic es (i. e. pr es ch oo ls, nu rse rie s, lon g d ay -c ar e s er vic es an d kin de rg ar te ns ) i n t he U nit ed St at es , A us tra lia an d I re lan d) . Th e e vid en ce w as in su ffi cie nt to dr aw co nc lus ion s re ga rd ing th e e ffe cti ve ne ss of th e i nt er ve nt ion s to im pr ov e t he im ple m en ta tio n o f p oli cie s, pr ac tic es or pr og ra m m es th at pr om ot e c hil d he alt hy ea tin g, PA an d/ or ob es ity pr ev en tio n, or to im pr ov e t he kn ow led ge , s kil ls or at tit ud es of ch ild ca re se rv ice st aff . N o s tra te gy im pr ov ed th e i m ple m en ta tio n o f a ll c om po ne nt s o f t he ob es ity pr ev en tio n p ro gr am m es ; h ow ev er, 7 ou t of 8 stu die s t ha t c om pa re d a n i m ple m en ta tio n str at eg y t o u su al pr ac tic e ( as op po se d t o alt er na tiv e s tra te gie s) re po rte d s ign ifi ca nt im pr ov em en ts in th e i m ple m en ta tio n o f a t l ea st on e c om po ne nt re lat ive to co nt ro l. T he re su lts of th e 3 st ud ies in ve sti ga tin g e ffe ct on ch ild nu tri tio n o ut co m es (f at in ta ke , e ne rg y i nt ak e, BM I, b od y w eig ht ) w er e a m big uo us , w ith on ly 1 stu dy re po rti ng im pr ov em en ts. W olf en de n e t a l., 20 17 (1 07 ) 2. 3 Th is Co ch ra ne sy ste m at ic re vie w of 27 st ud ies of str at eg ies fo r e nh an cin g t he im ple m en ta tio n o f s ch oo l- ba se d p oli cie s o r p ra cti ce s ta rg et ing ris k f ac to rs fo r ch ro nic di se as e w as un ab le to fin d s uffi cie nt ev ide nc e t o dr aw co nc lus ion s r eg ar din g th eir eff ec tiv en es s a t im pr ov ing im ple m en ta tio n o r im pr ov ing st aff kn ow led ge , at tit ud es an d s kil ls. Th e r ev iew in clu de d 2 7 ra nd om ize d a nd no n- ra nd om ize d s tu die s w ith a p ar all el co nt ro l g ro up , pu bli sh ed be tw ee n 1 98 8 an d 2 01 7. In te rv en tio ns to im pr ov e th e i m ple m en ta tio n o f ch ild ca re -b as ed ob es ity pr ev en tio n p ro gr am m es (fo cu sin g o n d iet , P A, ov er we igh t o r o be sit y, to ba cco or al co ho l u se ) b y sch oo l s ta ff (e .g. ed uc at ion al m at er ial s, ed uc at ion al m ee tin gs , t he us e o f o pin ion lea de rs, ex te rn al fu nd ing , loc al co ns en su s p ro ce sse s a nd ta ilo re d i nt er ve nt ion s). Ch ild re n a ge d 5 –1 8 y ea rs (n = 29 18 1) an d s ch oo l s ta ff (n = 13 47 ) i n t he U nit ed St at es , In dia , A us tra lia , C an ad a a nd So ut h A fri ca . Th e e vid en ce w as in su ffi cie nt to dr aw co nc lus ion s re ga rd ing th e e ffe cti ve ne ss of th e i nt er ve nt ion s to im pr ov e t he im ple m en ta tio n o f p oli cie s, pr ac tic es or pr og ra m m es fo r P A, he alt hy ea tin g, ob es ity pr ev en tio n, to ba cco us e p re ve nt ion or al co ho l u se pr ev en tio n, or to im pr ov e t he kn ow led ge , s kil ls or at tit ud es of ch ild ca re se rv ice st aff . T he m ajo rit y o f in clu de d t ria ls re po rte d s ign ifi ca nt im pr ov em en ts in at le as t o ne im ple m en ta tio n o ut co m e m ea su re. Re ga rd ing die ta ry in ta ke ou tco m es , 1 1 o ut of 14 st ud ies re po rte d i m pr ov em en ts on at le as t o ne m ea su re of di et ar y i nt ak e. Re ga rd ing ov er we igh t, ob es ity or ad ipo sit y o ut co m es , 7 of 9 stu die s r ep or te d n o be tw ee n- gr ou p d iff er en ce s. 165ANNEX 2: DETAILS OF REVIEWS INCLUDED AU TH OR , Y EA R NF SI ES SE NT IA L CR IT ER IA ON E S EN TE NC E S UM M AR Y ST UD Y T YP ES ST UD Y I NT ER VE NT IO NS ST UD Y S ET TI NG A ND PO PU LA TI ON RE SU LT S O F R EV IE W Yip et al ., 2 01 6 (1 33 ) 3.1 , 3 .3 Th is sy ste m at ic re vie w of 17 pr es ch oo l a nd sc ho ol- ba se d stu die s i nd ica te d t ha t pe er -le d n ut rit ion ed uc at ion pr og ra m m es ha ve po te nt ial to im pr ov e k no wl ed ge , se lf- effi ca cy an d a tti tu de s to wa rd s h ea lth y e at ing . Th e r ev iew in clu de d 1 7 stu die s p ub lis he d i n 2 00 2 o r lat er of pr og ra m m es w ith a p ee r c om po ne nt , m os t of w hic h u se d a pr e- an d po st- te st de sig n a nd co nt ro l gr ou ps . Pe er -le d n ut rit ion ed uc at ion pr og ra m m es . Ch ild re n a ge d 5 –1 8 y ea rs in kin de rg ar te n, pr im ar y a nd se co nd ar y s ch oo ls in Ca na da an d t he U nit ed St at es . Po sit ive re su lts w er e o bs er ve d i n t ho se st ud ies m ea su rin g s at isf ac tio n a nd be ha vio ur ch an ge as we ll a s s elf -e ste em . Im pr ov em en ts im m ed iat ely po st- pr og ra m m e w er e o bs er ve d f or in ta ke of F& V, SS Bs or fa ts, an d s elf -re po rte d h ab it an d be ha vio ur sc or es in 11 ou t o f 1 3 s tu die s r ep or tin g on th es e m ea su re s, bu t t he se ch an ge s w er e n ot m ain ta ine d o ve r t im e. Sig nifi ca nt re su lts w er e ob se rv ed fo r a nt hr op om et ric m ea su re s ( BM I, B M I z-s co re, w ais t c irc um fer en ce ) o r o th er bi olo gic al m ea su re s ( blo od pr es su re ) i n a ll 4 st ud ies m ea su rin g t he se ou tco m es . A do se -im pa ct re lat ion sh ip wa s o bs er ve d i n 3 ou t o f 3 st ud ies inv es tig at ing th is as pe ct. 166 NUTRITION ACTION IN SCHOOLS: A REVIEW OF EVIDENCE RELATED TO THE NUTRITION-FRIENDLY SCHOOLS INITIATIVE

ISBN 978–92–4–151630–3 ISBN 978–92–4–151696–9 For more information, please contact: Department of Nutrition and Food Safety World Health Organization Avenue Appia 20, CH-1211 Geneva 27, Switzerland Email: nutrition@who.int www.who.int/teams/nutrition-and-food-safety