Behavioural interventions can be delivered via online school canteens to improve healthy purchasing for students. However, no evaluations of the cost or cost-effectiveness of online canteen interventions have been conducted. The objective was to conduct a cost and cost-effectiveness analysis of implementing an online school canteen intervention to improve healthy purchasing.

The ‘Click & Crunch’ cluster randomised controlled trial was conducted in seventeen Australian primary schools.

Eight control schools (848 students) received the standard online canteen. Nine intervention schools (1359 students) received a behavioural intervention delivered through the online canteen.

Incremental cost-effectiveness ratios (ICER) were calculated for the cost per student to achieve (1) a unit decrease in the mean energy (kJ) content and (2) a percentage increase in the mean proportion of healthier ‘Everyday’ items purchased in their online lunch orders (from a health service and societal perspective).

It costs AUD$568 per school (range $343–$806) to implement. The ICER for mean energy content was AUD$0·06 and AUD$0·46 for mean proportion of ‘Everyday’ foods (from a health perspective). At a willingness to pay of AUD$0·20 and AUD$1·20 per student, the intervention would have a 95 % and 99 % probability of being cost-effective for the energy (kJ) content and proportion of ‘Everyday’ items, respectively.

‘Click & Crunch’ has the potential to be a cost-effective intervention to reduce the energy content and increase the proportion of ‘Everyday’ items from primary school online canteen lunch orders.

The research objectives were to evaluate factors that influence Canadian secondary school students’ milk and milk alternatives (MMA) consumption and to explore associations through age and gender lenses.

A qualitative design was used, consisting of semi-structured interviews and photo-elicitation methods. Analysis was guided by the Theory of Planned Behaviour (TPB). Deductive and inductive thematic analyses were used to generate themes, charting data based on attributes such as gender and age.

Interviews were held virtually or via telephone.

Participants were twenty-eight high school students from Ontario, Canada, diverse in terms of gender and age.

Both desirable and undesirable beliefs about the health outcomes of consuming MMA were commonly discussed. These included health benefits such as strong bones, muscular strength, and growth, and health consequences like unwanted skin conditions, weight gain, and diseases. While boys and girls associated MMA consumption with muscular strength, boys predominantly considered this favourable, while girls discussed outcomes like unwanted skin conditions and weight gain more often. Adolescents’ perspectives on taste/perceived enjoyment, environmentally friendly choices and animal welfare also influenced their MMA preferences. Parental influences were most cited among social factors, which appeared to be stronger during early adolescence. Factors involving cost, time and accessibility affected adolescents’ beliefs about how difficult it was to consume MMA.

Recommendations for shifting attitudes towards MMA are provided to address unfavourable beliefs towards these products. Interventions to increase MMA consumption among adolescents should include parents and address cost barriers.

This study aimed to (1) examine the clustering of energy balance-related behaviours (EBRB) and (2) investigate whether EBRB clusters, temperament and hair cortisol concentration (HCC) associate with overweight.

We assessed food consumption using food records, screen time (ST) using sedentary behaviour diaries, sleep consistency and temperament (negative affectivity, surgency, effortful control) using questionnaires and HCC using hair samples. Accelerometers were used to assess physical activity (PA) intensities, sleep duration and sleep efficiency. Researchers measured each child’s weight and height. We used finite mixture models to identify EBRB clusters and multilevel logistic regression models to examine the associations between EBRB clusters, temperament, HCC and overweight.

The cross-sectional DAGIS survey, data collected in 2015–2016.

Finnish 3–6-year-olds (n 864) recruited through preschools.

One-third of the participants were categorised into the cluster labelled ‘Unhealthy diet, excessive screen time’, characterised by unhealthy dietary choices (e.g. greater consumption of high-fat, high-sugar dairy products) and longer ST. Two-thirds were categorised into the second cluster, labelled ‘Healthy diet, moderate screen time’. PA and sleep were irrelevant for clustering. Higher negative affectivity and lower effortful control associated with the ‘Unhealthy diet, excessive screen time’ cluster. EBRB clusters and HCC did not associate with overweight, but surgency was positively associated with overweight (OR = 1·63, 95 % CI 1·17, 2·25).

Of the EBRB, food consumption and ST seem to associate. As temperament associates with EBRB clusters and overweight, tailored support acknowledging the child’s temperament could be profitable in maintaining a healthy weight.

To examine the extent and nature of food and non-alcoholic drink advertising displayed on public transport and infrastructure on school routes.

Audit of outdoor advertisements on government-controlled public transport and associated infrastructure (e.g. tram shelters, bus stops) on busy school routes in Victoria, Australia. Using a strict protocol, trained field workers collected data on the type and content of outdoor advertising during February 2023 (start of school year). Food/drink advertising was classified (unhealthy or healthy) according to the Council of Australian Governments Health Council National interim guide to reduce children’s exposure to unhealthy food and drink promotion (2018).

Government-controlled buses, trams and public transport infrastructure on routes from eleven of the busiest train stations in metropolitan Melbourne and regional Victoria, Australia to fifty public primary and secondary schools. Stations were chosen based on annual patronage, area-based socio-economic area (SEA) and regionality).

156 out of 888 advertisements were for food and non-alcoholic drinks. Of these, almost six in ten (58 %) were deemed unhealthy irrespective of SEA or regionality. Marketing appeals most featured were taste (31 %), convenience (28 %) and emotion (9 %). A significantly higher proportion of unhealthy advertisements were displayed within 500 m of schools v. outside this radius (91 % v. 57 %, P < 0·01).

Given the detrimental impacts of exposure to unhealthy food/drink advertising on children’s diets, the pervasive, powerful presence of such advertising across government public transport assets, particularly around schools, contradicts public health recommendations to protect children from exposure to and influence by this harmful marketing and warrants government action.

To investigate food consumption behaviour and self-perceived nutrition knowledge among university students, and to draw implications for nutrition education in contexts where formal nutrition education before university is limited.

A mixed-methods approach was adopted. A survey was first conducted to examine participants’ food consumption behaviour and self-perceived nutrition knowledge. Thirty-four participants were then invited to take part in semi-structured interviews to gain more in-depth insights into their self-declared knowledge and related behaviours.

Universities in China, representing a context of limited formal nutrition education in pre-university schooling.

190 university students.

Interview transcripts were reviewed to verify participants’ self-declared nutrition knowledge and identify misconceptions or gaps in understanding. Questionnaire data were analysed using descriptive statistics.

Students with higher education levels reported paying more attention to nutrition labels and selecting healthier snacks. However, interviews revealed that students who claimed to read nutritional claims during food purchases often misunderstood the meaning of sugar and fat content information. A significant ‘illusion of knowing’ was observed, and participants generally lacked awareness of authoritative food standards.

Illusion of knowing is common among students who have not received formal systematic nutrition education. Nutrition education programmes should prioritise raising students’ understanding of basic food concepts and improving their ability to interpret nutrition information accurately, as part of broader health promotion efforts.

To optimise school food baskets in Ghana to meet newly proposed food and nutrition targets while considering cultural acceptability and cost.

This was a modelling study. Data on existing school meal menus were collected from various regions to provide baseline inputs. Linear programming (LP) was used to model school meal baskets that satisfied minimum nutrient and food targets for school meals while meeting cost and acceptability constraints. Five LP models were tested, each varying in budget constraints and acceptability/food-based parameters.

Ghana.

NA.

Baseline school food baskets were significantly deficient in energy, protein, Fe, Zn, vitamin A, folate, vitamin B12 and vitamin C compared to food and nutrient standards for school meals in Ghana. Optimisation resulted in school food baskets that met cost, nutrient and food-based/acceptability targets but with substantial deviations from baseline. Achieving nutritional adequacy within cost limits increased reliance on animal-source foods and led to higher environmental impacts, indicating trade-offs between nutrition, affordability and environmental sustainability.

The study underscores LP’s potential for enhancing school meals in Ghana but highlights the need for increased financial investment for reaching dietary goals. Addressing local realities and cultural preferences is essential for implementing effective, sustainable school meal strategies and improving child health.

The double burden of malnutrition (DBM) – the coexistence of undernutrition and overweight/obesity – poses a critical global health challenge, particularly for children and adolescents. School meal programmes offer an opportunity to address the DBM by providing nutritious meals that support growth, development and lifelong health. However, limited school meal quality data hinders effective programme design. This study evaluates global school meal quality through nutrient composition analyses and the Global Diet Quality Score (GDQS)-Meal and -Menu metrics.

Data were collected from the Global Child Nutrition Foundation’s 2024 Global Survey of School Meal Programs, grey literature and in-country stakeholders. Nutritional content was compared to age-specific Nutrient Reference Values, including Harmonized Average Requirements. The nutritional quality of meals and menus was assessed using GDQS metrics based on twenty-five food groups.

Twenty-nine countries across diverse geographic and socio-economic contexts.

Not applicable.

Most menus met or exceeded 30 % of recommended daily nutrient values. Lunches had the highest nutrient contributions, followed by snacks and breakfasts. GDQS-Meal and -Menu scores revealed variability across age groups, meal types and countries. Overall, 57 % of menus achieved high nutritional quality, 37 % medium and 6 % low. Few menus lost points for unhealthy components. Diverse, balanced menus with healthy food groups scored higher, though fortification also proved nutritionally valuable.

While most menus contribute significantly to daily nutrient needs, variability across countries, age groups and meal types highlights opportunities for context-specific improvements through diversification and fortification. Learning from high-performing peer programmes can help identify feasible improvements.

To evaluate the effectiveness of the multicomponent intervention trial ‘Are You Too Sweet?’ in reducing discretionary foods and drinks intake among young schoolchildren.

The study was a 3·5-month two-arm cluster-randomised controlled trial among primary schoolchildren and their families. School health nurses provided guidance to families regarding discretionary foods and drinks for the children. Moreover, families were given a variety of knowledge- and capability-building materials to utilise at home. Dietary intake was assessed using a web-based 7-d dietary record. Linear mixed regression models were used to estimate intervention effects as changes in child intake of discretionary foods and drinks and sugar between groups.

Six schools from a Danish municipality were randomised to the intervention group (n 4) or the control group (n 2).

A total of 153 children aged 5–7 years.

No significant reduction in the children’s intake of total discretionary foods and drinks or discretionary foods alone was observed between the intervention and control group, while a decreased intake of discretionary drinks of 40·9 % (P = 0·045) was observed compared with control. Secondary subgroup analysis showed that children of parents with shorter educational level significantly reduced their intake of added sugar by 2·9 E% (P = 0·002).

The results of this study indicate that multicomponent interventions involving school health nurses may have some effects in reducing, especially, discretionary drinks.

This systematic review synthesises existing research on the relationship between dietary intake – specifically concerning food and beverage items promoted or restricted by US federal school nutrition policies – and child academic performance, a salient predictor of long-term health.

We used keywords to search three databases. Along with other inclusion criteria, studies had to assess and report (i) a measure of intake of food groups/nutrient promoted or restricted by US school nutrition policies; (ii) a measure of academic performance and (iii) a measure of the association between both.

We identified thirty-nine studies, all of which utilised observational designs, and seven of which were considered higher quality based off the Quality Assessment Tool for Observational Cohort and Cross-sectional Studies, published by the National Heart, Lung, and Blood Institute. Thirty-five studies reported evidence that children had better academic outcomes when they had an increased intake of food groups and nutrient items promoted by school nutrition policies and/or a decreased intake of the food groups and nutrient items limited by school nutrition policies.

These findings suggest that food groups and nutrient items governed by or promoted in school nutrition policies may shape children’s learning and contribute to downstream social determinants of health. Research on the population-level influences of school nutrition policies on children’s academic outcomes is warranted.

Rising levels of household food insecurity in England, and associated health and well-being impacts for children, have led to calls to expand access to free school meals (FSM). Policymakers have been hesitant to extend provision of FSM due to concerns surrounding acceptability, affordability and implementation challenges. The most effective strategies for expanding FSM are not yet fully understood. This work aims to fill this gap by examining school meal policy from the perspective of multiple stakeholders, to provide actionable recommendations for policies that could expand school meal provision.

A qualitative interview study design was used. The data were analysed using the Framework Method, underpinned by the Context and Implementation of Complex Interventions (CICI) framework. Themes were categorised into context-related and implementation-related factors. Detailed recommendations were discussed at the macro, meso and micro levels of the school food system.

The study was conducted in England, UK.

Seventeen stakeholders represented the views of local, regional and national government, policy, academia and schools.

Stakeholders indicated that policies should prioritise stigma reduction and integrate expansion of school meals with existing school policies where possible, including the monitoring of school food standards and ensuring maintenance of school food quality. Stakeholders also suggested improvement to the administrative process and communication with families and recommended a joined-up approach linking interventions with common goals across the whole food system.

Crucially, sufficient financial support is essential for successful implementation.

This study assessed macronutrient intake and associated factors among school adolescent girls in Meshenti, Northwest Ethiopia, 2020.

A cross-sectional study was conducted from 7 to 23 February 2020, among 401 randomly selected adolescent girls. Macronutrient intake was assessed using a 24-h dietary recall with portion size estimation. Nutrient data were analysed with Elizabeth Stewart Hands and Associates FOOD PROCESSOR software and compared with WHO/FAO recommendations – 2200 kcal for energy and 34–46 g for protein. Factors associated with inadequate macronutrient intake were identified using multivariable logistic regression.

The study was conducted in an institutional setting.

This study was conducted among school adolescent girls.

The median (IQR) energy intake was 2040·23 (1648·24–2744·51), and the mean (sd) protein intake was 63·88 (20·99). About 57·6 % (95 % CI: 52·9, 62·8 %) had inadequate energy intake, and 18·5 % (95 % CI: 14·7, 22·2 %) had inadequate protein intake. Inadequate energy intake was associated with dietary diversity (AOR = 4·31, 95 % CI: 2·20, 8·47), knowledge (AOR = 2·10, 95 % CI: 1·34, 3·28) and meal frequency (AOR = 2·5, 95 % CI: 1·06, 5·95). Factors linked to inadequate protein intake included early adolescence (AOR = 1·89, 95 % CI: 1·08, 3·31), residency (AOR = 0·27, 95 % CI: 0·15, 0·48), dietary diversity (AOR = 3·28, 95 % CI: 1·08, 9·98), knowledge (AOR = 1·82, 95 % CI: 1·04, 3·19) and meal frequency (AOR = 2·94, 95 % CI: 1·35, 6·37).

This study revealed high inadequate energy and protein intake. Contributing factors included dietary diversity, knowledge and meal frequency, with age and residence affecting protein intake. Emphasis is needed on early adolescent girls’ nutrition education.

This study aims at comparing two Italian case studies in relation to schoolchildren’s plate waste and its implications, in terms of nutritional loss, economic cost and carbon footprint.

Plate waste was collected through an aggregate selective weighting method for 39 d.

Children from the first to the fifth grade from four primary schools, two in each case study (Parma and Lucca), were involved.

With respect to the served food, in Parma, the plate waste percentage was lower than in Lucca (P < 0·001). Fruit and side dishes were highly wasted, mostly in Lucca (>50 %). The energy loss of the lunch meals accounted for 26 % (Parma) and 36 % (Lucca). Among nutrients, dietary fibre, folate and vitamin C, Ca and K were lost at most (26–45 %). Overall, after adjusting for plate waste data, most of the lunch menus fell below the national recommendations for energy (50 %, Parma; 79 %, Lucca) and nutrients, particularly for fat (85 %, Parma; 89 %, Lucca). Plate waste was responsible for 19 % (Parma) and 28 % (Lucca) of the carbon footprint associated with the food supplied by the catering service, with starchy food being the most important contributor (52 %, Parma; 47 %, Lucca). Overall, the average cost of plate waste was 1·8 €/kg (Parma) and 2·7 €/kg (Lucca), accounting respectively for 4 % and 10 % of the meal full price.

A re-planning of the school meals service organisation and priorities is needed to decrease the inefficiency of the current system and reduce food waste and its negative consequences.

To examine the impacts of school-based CalFresh Healthy Living (CFHL-California’s SNAP-Ed) interventions post-COVID-19-related school closures and whether student and school characteristics modified intervention impacts on student diet and physical activity (PA).

Quasi-experimental, two-group, pre-post, self-report.

CFHL-eligible public schools (nintervention = 51; ncomparison = 18).

4th/5th grade students (nintervention = 2115; ncomparison = 1102).

CFHL interventions were associated with an increase in consumption frequency of fruit (0·19 times/d (P = 0·015)) and vegetables (0·35 times/d (P = 0·006)). Differences in baseline diet and PA behaviours were observed by student race and gender and by whether the proportion of free and reduced-price meal (FRPM)-eligible students was above the state average. Notably, students in schools with FRPM above the state average reported more frequent consumption of sugar-sweetened beverages (Mean (se): 3·18 (0·10) v. 2·58 (0·11); P = 0·001) and fewer days/week with 60+ min of moderate-to-vigorous PA (MVPA) (Mean (se): 2·8 (0·10) v. 3·21 (0·12); P = 0·020) than those at schools with FRPM at/below the state average. Student gender, school urbanicity and school FRPM modified the relationship between the interventions and certain dietary and/or PA outcomes. Interventions were associated with greater increases in vegetable consumption in more urban schools (β (95 % CI) = 0·67 (0·15, 1·20)), and greater increases in fruit consumption (β (95 % CI) = 0·37 (0·07, 0·66)) and in MVPA in higher FRPM schools (β (95 % CI) = 0·86 (0·33, 1·39)).

Findings reaffirmed effectiveness of school-based CFHL interventions. We identified existing student and school-level disparities and then observed that interventions were associated with greater increases in MVPA in the highest FRPM schools. Findings can inform an equity-centred approach to delivery of school-based interventions that facilitate equal opportunity for all children to achieve lifelong health.

To identify and synthesise the educational and programmatic components most consistently associated with effective school-based food and nutrition education (SBFNE) for children and adolescents.

An umbrella review was conducted in accordance with PRISMA 2020 guidelines. We systematically searched CLEO, EDUCAT+, PubMed and Google Scholar through December 2022 for eligible literature reviews and meta-analyses. Included reviews were assessed for methodological quality and synthesised at the component level using qualitative and quantitative approaches.

Reviews examined SBFNE interventions delivered primarily during regular school hours in primary and secondary education settings worldwide.

Forty-four literature reviews synthesising evidence from 1651 primary studies (1115 unique) were included. Across reviews, twenty distinct educational and programmatic components were identified and standardised. Components most frequently reported and consistently associated with positive dietary behaviour change included Family Engagement, Focused on Behaviour Change, Multicomponent Approaches and Integration into the School Curriculum. Other components – such as Cultural Inclusivity and Environmental Sustainability – were examined less frequently but were associated with positive dietary outcomes in the reviews that assessed them. Several components, including Peer Support, Technology-Based Engagement and Community Engagement, demonstrated mixed or context-dependent effects, reflecting variability in implementation and study design. No component was consistently associated with adverse dietary outcomes.

By shifting the analytic focus from individual interventions to cross-cutting components, this umbrella review clarifies which elements most consistently underpin effective SBFNE. The resulting component framework provides a practical, transferable foundation to inform program design, evaluation and policy decision-making and to strengthen the equity, coherence and effectiveness of SBFNE initiatives.

To assess the nutritional composition, adequacy and environmental impact of menus served, consumed and wasted by 11–12-year-old students in public and charter schools in northern Spain.

A cross-sectional observational study (2017–2018) involving photographing menus before and after consumption, visual portion size estimation using a validated photographic catalogue and food waste assessment via the quarter-waste visual method. Nutritional composition was analysed using food composition databases and greenhouse gas emissions using life cycle assessment data.

Ten primary schools (five public and five charter) in northern Spain.

1000 school menus for students aged 11–12 years.

Menus served exceeded energy recommendations (791·5 (sd 176·7) kcal) and were high in fat (39·7 (sd 13·4) g), protein (29·7 (sd 10·0) g) and Na (980·4 (sd 302·2) mg) but low in carbohydrates (74·7 (sd 18·1) g), fibre (8·8 (sd 3·7) g) and several micronutrients. Food waste averaged 140·5 g per menu, mainly vegetables and fruit, leading to nutrient losses, particularly in fibre, vitamins A and C and Fe. The carbon footprint of menus averaged 1·489 kg CO2-eq, primarily from meat and fish, with waste contributing 0·298 kg CO2-eq. Public schools served more nutrient-dense food but had higher waste (public 151·5 (sd 112·3) g v. charter 129·5 (sd 86·3) g, P < 0·001); charter schools served more energy-dense food, with higher Na and fat (P < 0·001).

Menus showed nutritional imbalances, with excessive energy and Na and insufficient fibre and several micronutrients. Food waste worsened dietary adequacy while increasing environmental impact. Public schools offered more nutrient-rich food but faced greater waste compared with charter schools. Institutional differences suggest the need for tailored strategies to enhance both nutritional quality and sustainability.

During the COVID-19 pandemic, the US Department of Agriculture allowed all US schools to offer meals at no cost regardless of family income, a policy known as Universal Free Meals (UFM). Despite the recognised benefits of UFM, the policy expired in June 2022. The goal of this study was to gather perceptions of school staff in Arizona about school meals, UFM and the discontinuation of UFM.

This mixed-method study collected data using an online survey. Open-ended survey questions were analysed using qualitative thematic analysis, and closed-ended questions were analysed using descriptive analysis.

The survey was distributed to school staff in the two major metropolitan areas in Arizona between September and October 2022, soon after the UFM policy expired.

Survey responses were received from 1255 school staff, including teachers, cafeteria staff, administrators and other staff.

Most school staff (93 %) were supportive of UFM, and the support was consistent across all staff categories and across different political leanings. Thematic analysis demonstrated that staff felt UFM helped to meet students’ basic needs, reduced stigma and lessened the burden on teachers to use their own resources to provide food to students. Despite strong support, some staff reported concerns about food quality, programme waste and time available for lunch.

UFM policies were strongly supported by school staff, despite some concerns about programme implementation. Understanding these views is important to the discussion of expanding UFM policies in the USA and globally.

Schools are identified as a setting for food and nutrition education (FNE) in childhood. FNE is a key strategy to optimise child growth and development and impart life-long food skills. There is limited synthesis of the literature to understand the socio-ecological determinants of teachers and schools engaging in FNE.

For this scoping review, five databases (APA PsycInfo, ERIC, Medline, CINAHL and Scopus) were searched using the terms (and synonyms for) primary school teacher, self-efficacy and food and nutrition. A quality assessment using relevant Johanna Briggs tools was carried out for the included papers. Data were extracted using a modified socio-ecological model, and narrative themes were identified.

Primary (elementary) schools in high-income countries.

Primary-school teachers.

Forty-one papers were included in this review from ten countries (predominantly the USA). The narrative synthesis identified five themes that interact with teacher delivery of FNE. These were (i) perceived food and nutrition responsibilities of teachers, (ii) teacher beliefs and self-efficacy, (iii) opportunities to build teacher nutrition knowledge and self-efficacy, (iv) interpersonal contributors and (v) broader environmental, structural and policy contributors.

Multiple strategies are needed to build the capacity of teachers to undertake FNE within primary school settings. These strategies include a focus on learner-centred education that will build teacher agency, school leadership, ensuring the health and well-being of teachers and providing initial teacher education as well as innovative professional development for cross-curriculum integration. Strategies drawing from each level of the socio-ecological framework will increase opportunities for capacity building.

To examine how school food policies and perceived barriers influence food provision in New Zealand primary school canteens, using the ‘Healthy Food and Drink Guidance for Schools’.

Cross-sectional analyses of school food menus and school food policy and practices surveys completed by school leaders/principals.

New Zealand primary schools.

239 primary schools completed the school food policies and practices survey, and eighty schools provided canteen menus.

Most schools reported having a healthy food and drink policy in their school (76·2 %) and promoted healthy eating during school hours (87·4 %). Two-thirds (69·5 %) identified barriers to healthy food and drink provision, most commonly the convenience of ready-made foods (39·3 %), and resistance from parents (34·3 %). The number of reported barriers was not a significant predictor for the presence of a school food policy (OR-1·034, P = 0·841). School menus (n 80) consisted of 16·4 % ‘green’ items, 34·7 % ‘amber’ items and 36·8 % ‘red’ items. There was no relationship between the percentage of ‘green’, ‘amber’ and ‘red’ items and the presence of a school food policy or reported barriers. More than a third (38·9 %) of menus from schools that reported they had a ‘Plain Milk and Water’ only policy still contained sugar-sweetened beverages.

Although most New Zealand primary schools had healthy food policies, this was not consistently reflected in healthy food items on canteen menus. Further research is needed to understand how systemic barriers, such as cost, convenience and parental influence, affect policy implementation and school food provision.

To co-design a systems approach aimed at promoting the wide-scale adoption of whole-school approaches to food in UK primary schools to improve school food environments, food provision and dietary intake in children.

A systems framework (Action Scales Model) was used to guide the co-design of the systems approach. The process involved identifying leverage points within the UK primary school food system that, if influenced, could alter the way in which the system functions. Actions were then agreed upon to influence those leverage points.

Co-design workshops were held online between September 2021 and February 2022.

Members of the co-design team comprised twelve school stakeholders (headteachers, school food improvement officers, catering leads, representatives of UK school food organisations and a dietician) and a team of researchers with expertise in school food, systems thinking and intervention development. Our partnership board included decision-makers and advocates of the whole-school approach to food in England and Northern Ireland.

Identified leverage points included the priorities of headteachers, who are instrumental in instigating whole-school approach to food adoption. Direction from local and national policymakers was also identified. Actions to influence these leverage points included providing direct support to schools (through an online resource) and encouraging policymakers to monitor the adoption of the approach.

The methods described here can be replicated by others to promote the adoption of whole-school approaches to food in other contexts and contribute to the growing literature on developing systems-wide approaches to promote the adoption of public health initiatives.

To understand the relationship between adolescents’ unhealthy snacking behaviour during their school journey and their perceived and objective measures of food outlet availability in the school neighbourhood.

A cross-sectional survey enquired about socio-demographic information, school transport modes, perceived presence of food outlets in the school neighbourhood and unhealthy food purchase and consumption on the school journey. A geographical information system analysis of the food outlets within 500 m and 1000 m school buffers was undertaken. Data were analysed using generalised linear mixed modelling.

All twelve secondary schools in Dunedin, Aotearoa New Zealand, March 2020–June 2022.

Adolescents aged 13–18 years (n 725) who reported being familiar with their school neighbourhood.

Perceived availability of food outlets in the school neighbourhood was inversely correlated with distance to the closest food outlet from school and positively correlated with food outlet density within 500 m and 1000 m school buffers. Adolescents’ purchase and consumption of unhealthy snacks and drinks during the school journey were associated with perceived availability of food outlets and with shorter distance to the closest food outlet from school. Mixed transport users, girls and those living in high-deprivation neighbourhoods had higher odds of purchasing and consuming unhealthy snacks and drinks during the school journey than active transport users, boys and those living in low-deprivation neighbourhoods, respectively.

Adolescents perceptions of the food environment and close access to food outlets in the school neighbourhood may influence adolescents’ food purchase and consumption behaviours during the school journey.