Effects of school-based nutrition literacy Intervention on obesity prevention in school-age students: a systematic review and meta-analysis

Abstract

Background

Obesity among school-age students is a form of malnutrition resulting from excessive body fat relative to a child’s growth. Nutrition literacy plays an important role in shaping healthy eating behaviors and promoting long-term health. Although interest in this area has increased, evidence on the effectiveness and design of school-based programs to improve nutrition literacy remains limited. Therefore, this study aimed to (1) assess the effectiveness of interventions to promote nutrition literacy among school-age students for obesity prevention, and (2) identify the strategies, activity components, and implementation methods used in these interventions.

Methods

A systematic review and meta-analysis were conducted of randomized and non-randomized controlled trials conducted in school settings. Nine databases—CINAHL Complete, ProQuest, PubMed, Scopus, Web of Science, ThaiJo, ThaiLIS, TCI, and Google Scholar (English and Thai leaguages)—were searched for studies published in English or Thai up to 14 February 2025. Studies were included if they evaluated school-based interventions designed to improve nutrition literacy among school-age students aged 6 to 12 years old for obesity prevention and reported at least one nutrition literacy, behavioral, or anthropometric outcome. Two reviewers independently screened the studies, extracted the data, and assessed methodological quality using the Effective Public Health Practice Project tool. Meta-analysis was performed when the data were sufficiently comparable.

Results

A total of 21 studies were included in the review, of which 18 were included in the meta-analysis. Overall, the findings showed that school-based interventions improved food and nutrition literacy among school-age students, particularly in cognitive domains such as knowledge and related competencies. Interventions were commonly delivered through classroom-based, teacher-led approaches and were often supplemented with practical activities such as cooking, tasting, gardening, label-reading tasks, and family or school support. The pooled analysis showed a significant improvement in food and nutrition literacy outcomes, whereas the effect on healthy eating behavior was small and not statistically significant. Across studies, behavioral and anthropometric outcomes were less consistent than literacy outcomes, and substantial heterogeneity was observed.

Conclusions

School-based interventions can improve food and nutrition literacy among school-age students, but their effects on healthy eating behavior and anthropometric outcomes remain limited and inconsistent. Interventions that combine classroom teaching with practical skill-based activities and support from the family or school environment may be more promising than information-based approaches alone. Further studies with longer follow-up, clearer reporting of intervention content and implementation, and more consistent outcome measures are needed to clarify which intervention components are most effective for obesity prevention.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-026-27909-w.

Background

Obesity among school-age students is an important public health problem. In this review, school-age students are defined as children aged 6–12 years. Obesity is a form of malnutrition in which excess body fat occurs in relation to a child’s growth. Evidence on childhood obesity indicates that it is not caused only by an imbalance between energy intake and energy expenditure, but is also influenced by genetic, biological, behavioral, and environmental factors [1–3]. Childhood obesity has continued to increase worldwide. Between 2020 and 2035, the prevalence is predicted to double from 10% to 20% in boys and from 8% to 18% in girls [4]. In the United States, overweight or obesity among children aged 6 to 11 years could reach 33% by 2030 [5, 6]. In Asia, the prevalence is also expected to increase substantially by 2030 [7]. The ages of 6–12 years are an important period for the development of eating habits. Obesity among school-age students is often related to the frequent consumption of snack foods, fast foods, and sugar-sweetened beverages, along with low fruit and vegetable intake. Irregular meal patterns, large portion sizes, and limited nutrition knowledge may also lead to greater consumption of energy-dense, nutrient-poor foods. These eating patterns are part of the growing global burden of obesity [8–18]. Focusing on a single eating behavior may not be sufficient for obesity prevention. School-age students need support to make appropriate food choices in everyday settings [19]. Nutrition literacy concerns how students understand and use nutrition information in daily life [20]. School-based nutrition education has therefore been used to develop nutrition literacy and encourage healthier eating practices [21–23].

Nutrition literacy, as a subset of health literacy, focuses on the ability to interpret and use nutritional information to make informed dietary choices, such as understanding food labels and guidelines [24, 25]. It includes functional, interactive, and critical literacy levels, which together enable individuals to read, communicate, and critically evaluate nutrition-related information [24]. Nutbeam [26] stated that health literacy is the cognitive and social skills with which people can take action to maintain health. In Thailand, nutrition literacy has been described across six domains: access, cognitive, communication, self-management, media literacy, and decision-making [27]. This review included interventions that addressed one or more of these domains. Access skills refer to the ability to locate and verify reliable nutrition information. Cognitive skills relate to knowledge and understanding of appropriate dietary practices. Communication skills involve conveying and interpreting nutrition information in daily contexts. Self-management and decision-making skills focus on goal setting, planning, and making appropriate food choices. Media literacy involves assessing the credibility of nutrition information and avoiding misleading content (Health Education Division, 2017).

Schools are an important setting for health promotion because eating behaviors developed in childhood often continue into adulthood [19]. However, nutrition education is not consistently provided. Recent evidence indicates that teaching in this area remains limited, constrained by training, funding, and policy support [28]. Implementation may also be affected by limited facilities, time, and resources [29]. These conditions may reduce children’s opportunities to develop nutrition literacy. This is important during the primary school years, when food-related knowledge and skills are still developing [30]. School-based interventions can improve dietary behaviors, but their delivery differs across settings [31]. Clearer evidence is therefore needed to identify effective approaches and to guide the integration of nutrition education into school practice.

School-based interventions are crucial for promoting healthy eating habits and preventing childhood obesity by leveraging their extensive reach to include curriculum-based nutrition education, improve canteen options, and encourage supportive parenting [32, 33]. Key components of these programs include the active participation by health experts, educational sessions on dietary concepts, and comprehensive health assessment and feedback for parents [34, 35]. This holistic approach, school-based intervention integrating family, school, and personal interventions has the potential to provide significant long-term public health benefits and reduce childhood obesity rates [36, 37]. Consequently, schools serve as vital platforms for education and preventive healthcare practices.

Although previous reviews have examined school-based nutrition interventions in children, most have focused on dietary intake, healthy eating behavior, or general nutrition outcomes rather than nutrition literacy as the primary target. Existing reviews have also provided limited detail on how these interventions were designed and delivered, including their strategies, activity components, curriculum content, and implementation methods. The present review was therefore conducted to address this gap by focusing specifically on nutrition literacy interventions for obesity prevention among school-age students and by examining both their effectiveness and intervention characteristics [38–40].

This present systematic review and meta-analysis focuses on identifying interventional studies that promote nutrition literacy among school-age students by systematically searching the relevant literature. Following data selection and extraction, the review will identify and report the components, implementation methods, and effectiveness of these interventions. The objectives of this study are to (1) assess the effectiveness of interventions in promoting nutrition literacy among school-age students for obesity prevention, and (2) identify the strategies, activity components, and implementation methods of these interventions. The findings will serve as a guide for developing future interventions to improve school-age students ‘s nutrition literacy.

Method

This detailed systematic review and meta-analysis were carried out in alignment with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. This systematic review and meta-analysis have been appropriately registered with PROSPERO, the International Prospective Register of Systematic Reviews and Meta-Analyses (CRD42024612790).

Study eligibility criteria

The literature search was conducted between 28 October 2024 and 14 February 2025. Studies were considered eligible if they met all inclusion criteria and none of the exclusion criteria. The inclusion and exclusion criteria are presented in Table 1.

Table 1.

Study eligibility and exclusion criteria based on the PICOS elements Studies

PICOS	Inclusion Criteria	Exclusion criteria
Population	School age students aged 6 to 12 years old	Studies in which all participants had a specific or familial disease were excluded. Studies that combined children with other age groups without separate data for children were also excluded.
Intervention	1. School-based intervention. 2. The study included interventions focusing on any components of food and nutrition literacy, enhancing competence across multiple skill domains[27] 3. Intervention included being aware of and knowledgeable about snack foods and their nutrition, having the skill to gather information from a range of sources, preparing snacks, increasing the intake of fruits, vegetables and sugary beverages while reducing the intake of prepared foods, selecting healthy snack foods, and being able to read and analyze food labels.	Studies without a clearly described intervention and targeting children with specific clinical conditions were excluded.
Comparison	Usual care/ routine school health activities or delayed intervention	Studies without a comparator group were excluded.
Outcomes	1. Primary outcomes: Healthy eating diet knowledge about obesity prevention which will be categorized according to the above-mentioned six domains of nutrition literacy. 2. Secondary outcomes: 2.1 Dietary behavior from dietary record or survey, 2.2 Body weight status including: BMI, BMI Z–score.	Studies that did not report outcomes relevant to the review or provided insufficient data for analysis were excluded.
Study design	1. Randomized controlled trials and cluster randomized controlled trials (teachers, classrooms, school levels), 2. Non-randomized trials	Studies with inaccessible full texts or insufficient data after two attempts to contact the corresponding author were excluded. Non-comparative designs were also excluded.
Language	English and Thai	Studies for which the full text could not be obtained or translated into usable data were excluded.

Studies published in English or Thai were included. English-language studies were included to cover evidence from the international literature, whereas Thai-language studies were included to avoid missing relevant studies conducted in the local context. These languages were selected because the review team could accurately screen, extract data from, and interpret studies published in these languages.

Search strategy

The search was conducted from 28 October 2024 to 14 February 2025. The bibliographic search strategy focused on articles published in peer-reviewed journals in English and Thai. The databases to search were CINAHL Complete, ProQuest, PubMed, Scopus, Web of Science, ThaiJo, ThaiLis, TCI and Scholar (English and Thai language). The review team (SP, SW, AP and SWB) thoroughly designed and carried out recommended search methodologies or queries that were aligned with the data bank across a range of databases, adhering to the PICOS (Participants, Intervention, Comparison, Outcome, and Setting) framework and the Medical Subject Headings (MeSH) databases. The keywords to use were be: (1) English data base: (Child OR Students) AND (Primary school OR Elementary school) AND School-based AND (Sugary beverages OR Vegetable OR Fruit OR Snack OR Healthy diet OR Dietary OR Nutrition) AND (Information OR Knowledge OR Education OR Literacy), (2) Thai data base: Health literacy for the prevention of school-age children and the prevention of childhood obesity.

Study selection

All literature search results were imported into EndNote 21 and systematically deduplicated to create a consolidated library. Using a pretested eligibility criteria checklist, two reviewers (SP and SW) independently screened titles and abstracts for eligibility. Full-text articles of all potentially eligible studies were then obtained and independently assessed by the same two reviewers against the eligibility criteria. Any disagreements at either stage were resolved through discussion. Final inclusion decisions followed this checklist, and reasons for exclusion were recorded. Any disagreements were resolved through discussion and, when needed, consultation with a third reviewer (AP). The selection process was documented using the PRISMA flowchart, as seen in Fig. 1.

Fig. 1.

PRISMA flow diagram: the results of the search

Data extraction

Data were extracted into a standardized data extraction form in Microsoft Excel. The data of interest comprised author, title, year of publication, study design, setting, duration, number of participating schools, participant characteristics, intervention characteristics (including personnel, duration), control, primary and secondary outcome data at baseline and follow-up time points, method of statistical analysis and a summary of study findings. This was carried out by SP for all studies, and each form was verified by SW. Any discrepancies were identified and resolved through discussion or adjudicated by a third author (AP).

Data analysis

The quality of the included studies was independently assessed by two reviewers working in duplicate using the validated Quality Assessment Tool for Quantitative Studies developed by the Effective Public Health Practice Project (EPHPP) [41]. The EPHPP tool includes six components used for the global rating: selection bias, study design, confounders, blinding, data collection methods, and withdrawals/dropouts. Each study was rated as strong if no component was rated weak, moderate if one component was rated weak, and weak if two or more components were rated weak.

In addition, intervention integrity and analytical appropriateness were examined descriptively. Intervention integrity was considered based on the proportion of participants receiving the allocated intervention, the consistency of intervention delivery, and the potential presence of co-intervention or contamination. Analytical appropriateness was assessed with respect to alignment between the unit of allocation and the unit of analysis, the suitability of the statistical methods, and the use of intention-to-treat analysis. These components were not included in the global rating.

Any disagreements were resolved through performed by two reviewers: SP quality appraisal data and SW double-checked for accuracy. Due to the nature of the intervention, performance bias was not considered because it is not possible to blind study personnel or participants as seen in Fig. 2.

Fig. 2.

Quality assessment using the EPHPP. (N = 21)

Publication bias was accounted for in the meta-analysis. Funnel plots and statistical tests for funnel plot asymmetry were planned when a sufficient number of studies were available for an outcome. For outcomes with a small number of studies, formal assessment of publication bias was not performed because the results would be difficult to interpret.

Data synthesis

Descriptive synthesis was used to summaries the characteristics and quality of studies evaluating school-based interventions to improve nutrition literacy for obesity prevention among school-age students. Intervention components (e.g., contact time, curriculum, behavior change strategies) were extracted and summarized in a matrix table, and their frequency was examined. Meta-analysis was conducted when studies were comparable with respect to population, intervention, outcomes, and study design. Outcomes were included when sufficient data were available to estimate effects. Data were pooled in Stata to produce summary estimates. Reporting followed PRISMA 2020 [42].

Results

Study selection

The search included studies published up to February 14, 2025. In total, 3,956 records were identified from the selected databases: Scopus (196), PubMed (1,947), Web of Science (706), CINAHL Complete (644), ProQuest (200), Google Scholar (English) (200), TCI (2), ThaiJo (28), ThaiLIS (3), and Thai Google Scholar (30). After removing duplicates, 2,270 records remained. Screening of titles and abstracts excluded 2,159 records. The full texts of 111 articles were then reviewed. Of these, 90 did not meet the inclusion criteria. The remaining 21 studies were included in the qualitative synthesis, detailed in Fig. 1.

Study characteristics

Of the 21 studies included, nine studies were cluster- or individually randomized controlled trials (RCTs), whereas the remaining 12 studies were quasi-experimental or other non-randomized designs. Seven trials were located in the United States, two each from China, Iran, and the Netherlands, single studies from Australia, Canada, Germany, Italy, Scotland, Taiwan, Thailand and the United Kingdom. Low- and middle-income countries, therefore, account for roughly one-third of the sample. Six studies targeted school-age students aged 9–12 years, while four studies focused specifically on the 9–10 and 10–12 age groups. One additional study included participants aged 6–11 years. (Supplementary files 2; Table 1 provides key characteristics of studies included in the review).

The effectiveness of interventions in promoting nutrition literacy on obesity prevention among school-age students

This review synthesized 21 of interventions aimed at improving nutrition literacy among school-age students for obesity prevention. Outcomes are examined across cognitive, affective, behavioral, and anthropometric or broader health domains, with particular attention to whether the observed effects were sustained beyond the intervention period.

Cognitive outcomes

Including food and nutrition literacy, general nutrition knowledge, and obesity-related knowledge, generally improved following intervention, although the magnitude and consistency of effects varied across studies [43–55]. Marked improvements in comprehensive literacy measures were observed, particularly significant gains across all subdomains of food and nutrition literacy [43]. Consistent improvements in knowledge-related outcomes were also reported across several studies [45, 48, 51–55]. In some studies, these effects appeared to be sustained over time, and nutritional knowledge was identified as a mediator of behavioral change [54, 55]. More selective improvements were reported in other studies. Significant gains were observed in several cognitive indicators, including understanding of healthy eating, diet–disease relationships, and food categorization [47]. Modest improvements were also found in knowledge related to tasting and healthy-eating concepts, as well as in the number of foods recognized [49]. In contrast, one study found improvement limited to understanding of the food system (farm-to-plate), with no significant changes in other knowledge measures [44]. Another study reported improvements in nutrition knowledge in both intervention and control groups, but no significant between-group difference was observed [46]. No change was found in knowledge regarding the importance of fruit and vegetable consumption for disease prevention [50]. (Supplementary files 2; Table 2 summary of findings from included studies).

Affective outcomes—including attitudes, preferences, subjective norms, and self-efficacy—showed mixed effects across studies. Improvements were reported in several interventions, particularly in vegetable preference, attitudes toward cooking, and self-efficacy [56], as well as in subjective norms supporting fruit and vegetable intake and reduced preference for energy-dense snacks [47]. Positive changes in attitudes toward healthy foods, taste preferences, and overall perceptions of healthy eating were also observed [51, 53, 57]. Modest improvements in affect-related behavioral determinants, such as intentions to try new foods and perceived social influences, were additionally reported [49]. However, not all interventions demonstrated significant effects. Some studies found no changes in attitudes, perceptions of the social environment, or willingness to try fruits and vegetables [50], while others reported improvements in both the intervention and control groups, without significant between-group differences in acceptance or intention-related outcomes [46] (Supplementary files 2; Table 2 summary of findings from included studies).

Behavioral outcome

Positive effects were reported for several specific dietary behaviors, including increased vegetable intake [58], increased fruit intake [47, 50, 59], reduced sugar-sweetened beverage and salty snack consumption [59], increased water intake without significant changes in juice or soft drink intake [60], increased intake of several fruits and vegetables on recall measures [48], improved tasting-related behavioral determinants and intention to taste or eat healthily [49], and favorable changes in dietary patterns, with greater consumption of unprocessed foods and lower consumption of ultra-processed foods.

However, total energy intake did not change [61]. Additional improvements were observed in food-related practices, as cooking behaviors predicted higher dietary fiber and vegetable intake, and gardening behaviors predicted higher dietary fiber intake [62]. Another study also showed improvements in dietary behaviors, with increased consumption of meat and nuts, and further demonstrated that nutritional knowledge mediated both healthy and unhealthy dietary behaviors [55]. In contrast, no significant effect was found on the processing profile of foods brought from home [63], fruit and vegetable intake and cooking-skill measures [44], fruit and vegetable consumption in a large school-based intervention [57], or overall healthy eating behavior [53]. Taken together, the evidence suggests that these interventions improved selected eating behaviors and food-related practices, but effects were inconsistent across studies and not sustained across all outcomes [44, 47–50, 53–55, 57–63] (Supplementary files 2; Table 2 summary of findings from included studies).

Anthropometric and health outcomes

Anthropometric and health outcomes were assessed in only a limited number of studies and showed minimal effects overall. One study reported a reduction in overweight prevalence after 1 school year; however, no significant between-group difference in BMI z-score was observed [8]. Other studies found no significant changes in body composition, blood pressure, or total energy intake despite improvements in dietary behaviors [4, 5]. In addition, it was suggested that anthropometric changes may require longer or more intensive interventions [16], while one study did not assess anthropometric outcomes [1]. Overall, the evidence indicates that these interventions had a limited impact on anthropometric and health outcomes (Supplementary files 2; Table 2 summary of findings from included studies).

Sustainability of effects

Sustainability of NL gains remain a challenge; It was found that while short-term effects on knowledge and preferences were promising, most improvements faded without reinforcement beyond three to six months [47, 61]. Interventions with follow-up beyond one year provide critical insight into the need for sustained, multi-level engagement to maintain behavior change [54] (Supplementary files 2; Table 2 summary of findings from included studies).

Characteristics of interventions: strategies, components, and implementation

Theoretical foundations and intervention strategies

A variety of theoretical models informed the interventions, each providing a tailored approach to enhancing nutrition literacy. Social Cognitive Theory was the most commonly applied, featured in 7 studies [44, 47, 48, 52, 55, 56]. Other frameworks appeared less frequently, including the Theory of Planned Behavior (3 studies [51, 52, 63]), social-ecological models (2 studies [59, 64]), and single mentions such as the Environmental Intervention Model [47], Experiential Learning Theory [56], Intervention Mapping (IM) [43], and the PRECEDE–PROCEED Model [48]. Four studies did not specify a guiding theoretical framework [46, 50, 54, 62] (Supplementary files 2; Table 1 key characteristics of studies included in the review).

Intervention components and curriculum content

Intervention components and curriculum content were delivered predominantly through school-based, classroom-oriented formats, which were reported in 18 studies [43–47, 49, 51–59, 61–63]. Face-to-face delivery was described in 13 studies [43–46, 50, 53, 56–61, 63], whereas less common formats included hybrid online and classroom delivery during the COVID-19 period [48], app-based gamified learning [57], environmental modification combined with brief classroom teaching [51], and school- and family-based education supported by digital messaging [54].

Across studies, curriculum content consistently addressed foundational nutrition knowledge, including food and nutrient concepts, healthy eating, food groups, daily caloric intake, hydration, obesity-related knowledge, and balanced diet principles [43, 49, 51, 53–55, 62, 63]. Several interventions extended this content to food processing, food classification, and label-related competencies, including label literacy, nutrition label comprehension, classification of healthy and unhealthy foods, recognition of processed and ultra-processed foods, and understanding of the health and environmental implications of food choices [43, 44, 53, 54, 59, 62, 63].

A substantial proportion of interventions incorporated experiential and skill-based learning, particularly cooking, food preparation, tasting, and sensory activities [45–48, 50, 52, 56, 59–61, 63]. In several studies, these activities were supplemented by gardening, harvesting, or farm-based experiences designed to strengthen understanding of food origins and encourage fruit and vegetable consumption [47, 50, 52, 59–61]. Behavioral and decision-oriented components were likewise common and included food choice, planning, goal-setting, portion awareness, self-efficacy, intention formation, and family or peer reinforcement [43, 44, 46, 47, 50, 52–56, 61–63]. A smaller number of studies further incorporated broader themes such as food waste and sustainability [48]. Taken together, the included interventions were characteristically multicomponent, combining classroom instruction with practical skill development and social or environmental reinforcement (Supplementary files 2; Table 1 the key characteristics of studies included in the review).

Nutrition education—covering definitions, food groups, and nutrient functions—was included in 20 intervention programs [43–59, 62–64]. Practical skill practice (gardening, cooking, label reading, game-based challenges) appeared in 13 studies [43, 45, 47–50, 52, 53, 55, 58, 61, 63, 64]. Behavioral counselling (goal-setting, self-monitoring) was explicit in only four studies. Every intervention claimed to build cognitive nutrition literacy. More than half also aimed at decision-making or self-management skills, and almost half at communication skills (e.g., peer or family dialogue). Only a minority modified access to healthier food or drink (two studies [43, 44]) or tackled media-literacy (three studies [43, 47, 58, 63]) (Supplementary files 2; Table 1 key characteristics of studies included in the review).

In 18 of the 21 studies [43–59, 61–64], the stated primary outcome was improved knowledge of nutrition or obesity prevention, highlighting the field’s didactic focus. About half of the studies tracked dietary behavior only as a secondary endpoint, while just four measured BMI and two reported BMI-z. This imbalance reflects both the practical ease of testing knowledge and the longer, more challenging pathway required to demonstrate changes in weight status (Supplementary files 2; Table 1 key characteristics of studies included in the review).

The implementation methods of the interventions

Intervention durations varied, with nine studies lasting not longer than three months [45–50, 53, 57, 62], four studies lasting 4–6 months [43, 55, 56, 58], and eight studies lasting 7–12 months [44, 51, 52, 54, 59, 61, 63, 64]. Weekly sessions were the most common schedule (12 studies [43–46, 48, 50, 53, 55, 56, 59, 62, 63]). Daily exposure (mainly environmental changes) appeared in only two studies, and seven used a monthly session. Where specified, most lessons lasted 40–60 min (10 studies [43, 46, 51–53, 55, 57, 59, 62, 64]); six studies scheduled 1–2 h blocks, while two studies into ≤ 30 min mini-lessons [54, 61]. Classroom teachers were the principal facilitators in 19 studies [43–49, 51–57, 59, 61–64], often after brief training. Smaller subsets added specialist nutritionists (five studies [43, 45, 50, 56, 57]), researchers (six studies [44, 53, 58, 61–63]) or community experts such as gardeners (two studies [50, 59]), chefs (one study [47]) or parents (one study [54]) (Supplementary files 2; Table 1 key characteristics of studies included in the review).

All studies relied on self-administered questionnaires to gauge nutrition knowledge, attitudes, or self-reported intake. Roughly one-third supplemented these with food-frequency diaries or recalls, and 5 studies collected full 24-h dietary recalls. Truly objective tools—digital lunch-box photos, vending-machine sales logs, or water-fountain flow meters—were used only sporadically. Anthropometric measures appeared in four trials, but BMI was seldom treated as a primary outcome (Supplementary files 2; Table 1 key characteristics of studies included in the review).

Nutrition curriculum content

The 21 reviewed studies applied varied but overlapping nutrition curricula, commonly covering food groups, healthy eating, label reading, and behavior change. Most combined theoretical instruction with experiential learning and addressed five key themes: food and nutrient knowledge, food preparation, label/media literacy, decision-making, and sustainability (Supplementary files 2; Table 3 the nutrition curriculum content).

Food and nutrient knowledge

Curricula consistently introduced basic food groups and their nutritional value, then built toward a clearer understanding of macronutrients, micronutrients, fiber, and water [43, 55, 59]. Several programs distinguished whole from processed foods, natural from added sugar, and “healthy” from “junk” foods, linking these concepts to everyday school meals and snacks [44, 57, 59]. Hydration and beverage choice were treated as core nutrition topics, with lessons on the water cycle, hydration physiology, recommended fluid intake, and strategies to choose water over sugary drinks [51, 54]. Several programs aligned content with national dietary guidelines, the Mediterranean Diet Pyramid, or similar models to reinforce consistency with public health recommendations [48, 57, 61].

Food preparation, cooking, and sensory activities

Practical cooking was used to translate abstract nutrition content into concrete skills. Children prepared simple dishes rich in fruits, vegetables, and whole grains, often eating together afterwards [45, 48, 50]. Lessons commonly included food safety, basic preparation skills, and working with recipes, with some classes compiling a cookbook from student work [48, 63]. Sensory activities, such as repeated tasting, comparing varieties of fruits and vegetables, and discussing flavor, texture, and appearance, were used to support acceptance of healthy foods [46, 52, 61]. Gardening and farm visits further extended these experiences, engaging pupils in planting, growing, harvesting, and composting, while linking these activities back to cooking and shared meals [47, 50].

Label reading and food information literacy

Several studies included label reading and the use of food information as part of the intervention. These activities aimed to help students read nutrition labels, identify foods high in fat, sugar, or salt, and use this information when making food choices [43, 49, 53, 54, 62]. Common activities included food cards, supermarket tasks, and calorie worksheets, which helped students relate label information to recommended intake and daily energy needs [49, 62].

Media literacy and media-based components

Media-related components were addressed in different ways across the studies. One study explicitly incorporated media literacy by addressing resistance to food advertising as part of a broader food and nutrition literacy program [43]. In contrast, two other studies used media primarily to communicate and reinforce intervention messages through newsletters, school-wide announcements, posters, poster contests, and quizzes, rather than to develop students’ critical evaluation of media content [44, 63].

Healthy behavior and decision-making

Much of the content focused on everyday food-related decisions rather than knowledge alone. Lessons covered choosing healthier snacks, reducing sugary drink intake, increasing fruit and vegetable consumption, and making appropriate choices across common settings, such as school, home, and when eating out [43, 53–56, 64]. Some curricula also included refusal skills and guidance to help children make healthier food choices during shopping with parents [43]. Tools such as goal-setting worksheets, self-monitoring logs, and home tasks were used to help children monitor and adjust their behaviors over time [52, 56, 63].

Broader Food Systems, Sustainability, and Equity: Several curricula place individual eating habits within a broader food-systems context by showing how food moves “from farm to table,” how processing changes foods, and how packaging and energy use create environmental impacts [44, 48, 64]. The Food Education and Sustainability Training (FEAST) program, for example, combines lessons on food waste, fridge and fruit-bowl audits, and the role of a food-rescue organization (OzHarvest) with recipe design using commonly wasted fruits and vegetables [48]. Gardening and composting activities give pupils direct experience of planning, planting, growing, and harvesting, and encourage reflection on soil health and waste reduction [50]. Equity issues also appear, for example, in lessons about eating healthfully in “food desert” neighborhoods and in activities linking food equity and community service, which highlight how access to fresh produce is uneven across communities [64].

The Components affected by intervention (NL, HL)

The 21 reviewed studies demonstrated that school-based nutrition interventions influenced a range of components within nutrition literacy (NL) and health literacy (HL). Most interventions enhanced cognitive literacy, including knowledge of food groups, nutrient awareness, and label comprehension [43, 55]. Decision-making skills were improved through activities such as meal planning, cooking, and choosing healthier snacks, e.g., [56, 59]. Many programs supported self-management by fostering goal-setting, healthy eating behaviors, and the exploration of new foods [53, 64]. Communication skills and social norms were targeted through peer, parent, and teacher involvement [52, 61], while media literacy—though addressed less frequently—focused on interpreting advertisements and understanding environmental impacts of food choices [43, 44]. Overall, the interventions took a multi-dimensional approach to building critical competencies in children’s food and health decision-making. (Supplementary files 2; Table 4 the components affected by intervention (NL, HL)).

Meta-analysis

Among the 21 studies in this systematic review, 18 studies were included in the meta-analysis. Includes studies reporting results based on nutritional knowledge (10 studies) [43, 49, 53–55, 57, 58, 61–63] and nutritional behavior (10 studies) [44, 46, 48, 50, 52, 53, 56, 59, 61, 64], as seen in Fig. 1. An insufficient number of studies reported effect sizes consistently to facilitate subsequent meta-analytical procedures. The details are as follows:

Four randomized controlled trials (RCTs) with a total of 2,573 participants reported outcomes in a consistent format suitable for meta-analysis of nutrition-related interventions. The meta-analysis showed no significant overall effect of the interventions on the pooled nutritional behavior outcome, with a Cohen’s d of − 0.02 (95% CI: − 0.08, 0.04), p = 0.60. Given the absence of heterogeneity (I² = 0%), the consistently null results across studies indicate that, with the intensity and format used, these programs did not lead to a meaningful improvement in healthy eating behavior compared with usual practice. (Fig. 3)

Fig. 3.

A forest plot presenting data on the effects of school-based nutrition program (RCT) (Outcome: nutritional behavior)

Five quasi-experimental studies with a combined total of 3,188 participants were included in the meta-analysis of nutrition interventions reporting nutritional behavior outcomes in a consistent format. The pooled analysis showed a small, non-significant overall effect size (Cohen’s d = 0.04, 95% CI: − 0.08 to 0.15, p = 0.55). There was moderate heterogeneity across studies (I² = 63.25%), suggesting variation in intervention effects. Meta-regression with a binary country variable (USA [50, 52, 56] vs. other [46, 61]) yielded β ≈ −0.08 (95% CI: −0.47 to 0.31; p ≈ 0.56), indicating that the effects on nutritional behavior did not differ in any straightforward way between US and non-US trials. Overall, the evidence suggests limited and inconsistent effects across trials, (Fig. 4).

Fig. 4.

A forest plot presenting data on the effects of school-based nutrition program (Quasi experimental) (Outcome: nutritional behavior)

Three RCTs with a total of 1,559 participants were included in this meta-analysis, focusing on nutrition interventions that reported nutritional knowledge outcomes in a format allowing effect size calculation. The overall pooled effect was statistically significant, with a large effect size (Cohen’s d = 2.33; 95% CI: 0.05 to 4.60; p = 0.04). However, heterogeneity was very high (I² = 99.39%). The three studies differed in several areas, including the type of intervention, participant characteristics, outcome measures, intervention duration, and follow-up period. Ahmadpour [43] also reported a much larger effect size than Lakshman [49] and Rosati [57]. These differences may have contributed to the high heterogeneity. Since only three studies were included in this analysis, subgroup analysis and meta-regression were not conducted. Therefore, the pooled result should be interpreted with caution (Fig. 5).

Fig. 5.

A forest plot presenting data on the effects of school-based nutrition program (RCT) (Outcome: nutritional knowledge)

Six quasi-experimental studies with a total of 1,379 participants were included in this meta-analysis assessing the impact of nutrition interventions on knowledge and cognitive outcomes. The pooled effect size was statistically significant (Cohen’s d = 1.16, 95% CI: 0.56 to 1.77, p < 0.001), indicating a significant overall impact. However, substantial heterogeneity was observed across studies (I² = 96.16%), reflecting variability in intervention effects. The most substantial effects were observed, showing Cohen’s d > 2.0 [58, 62]. Meta-regression using region/ethnicity (East Asian [53–55] vs. other [58, 62, 63]) yielded β = −1.06 (95% CI − 2.24 to 0.11; p = 0.15) [58, 62]; East Asian trials showed smaller gains in nutritional knowledge, but the difference was not statistically significant. These findings support the effectiveness of school-based nutrition education in improving nutrition-related knowledge and understanding (Fig. 6).

Fig. 6.

A forest plot presenting data on the effects of school-based nutrition program (Quasi-experimental) (Outcome: nutritional knowledge)

Discussion

This systematic review and meta-analysis examined the effectiveness of interventions to promote nutrition literacy among school-age students to prevent obesity. They described their strategies, activity components, and implementation methods. The findings showed that these interventions improved nutrition literacy outcomes, particularly knowledge and related competencies, whereas effects on healthy eating behavior were smaller and less consistent. The pooled analysis showed clear improvements in food and nutrition literacy, but no significant effect on healthy eating behavior. Considerable heterogeneity was observed across literacy outcomes. More effective interventions generally combine classroom teaching with practical activities and support from the school or family environment.

These findings are broadly consistent with earlier reviews of school-based nutrition interventions, which have generally shown clearer effects on knowledge and selected dietary outcomes than on BMI or other anthropometric measures [38, 39]. Previous reviews have mainly examined school-based nutrition programs in relation to dietary intake, healthy eating behavior, or general nutrition outcomes, rather than nutrition literacy as the main target [22, 39]. Although the review focused more directly on food and nutrition literacy among school-age students, most interventions were still centered on functional literacy and were delivered primarily by teachers [40]. The present review adds to this literature by focusing specifically on nutrition literacy for obesity prevention and by examining intervention content and delivery in greater detail. Across the included studies, nutrition literacy was addressed through components such as label reading, food choice, decision-making, food preparation, and, in a few cases, media-related skills. The overall pattern suggests that literacy outcomes may improve more readily than behavior or weight-related outcomes. This is consistent with broader evidence showing that school-based programs can strengthen knowledge and some health behaviors, but changes in BMI are often less consistent. They may require a longer duration, greater intensity, and support from the family and school environment [22, 38].

The intervention characteristics identified in this review help explain the observed pattern of findings. Most programs were classroom-based and teacher-delivered, but stronger results were generally observed when classroom teaching was combined with practical activities and support beyond the classroom, a pattern consistent with previous reviews of experiential nutrition programs in school-age students. Study reported that programs were more likely to be successful when they included multiple strategies, parental involvement, and hands-on activities [65]. At the same time, experiential learning approaches were found to be associated with improvements in knowledge, attitudes, and healthy eating behaviors [66]. Such approaches have also been linked to gains in nutrition knowledge, self-efficacy, food preferences, and dietary behaviors in school-age students [65, 66]. Family involvement may further strengthen program effects, although evidence at the household level remains less consistent than evidence for child outcomes, as noted in a systematic review of school-based nutrition interventions involving parents and other family members [67].

The present review also highlights an important conceptual distinction between media literacy and media-based delivery. In the included studies, explicit media literacy was uncommon. Most interventions used posters, newsletters, quizzes, applications, or videos to deliver or reinforce nutrition messages. In contrast, only a small number explicitly included critical appraisal of food advertising or media content as a learning objective. This distinction is important because the use of media as a teaching tool does not necessarily indicate that media literacy is being developed as a competency. This interpretation is supported by media literacy interventions that explicitly targeted food advertising and aimed to strengthen children’s and parents’ ability to analyze persuasive messages, as well as by family-based nutrition media literacy programs that improved parent–child discussions about food marketing and aspects of the home food environment [68, 69].

Several factors may explain why behavioral and anthropometric outcomes were less consistent than nutrition literacy gains. Many interventions were relatively short and may not have been intensive enough to produce sustained change, particularly in outcomes such as BMI. Their effects may also have been limited by practical constraints in school settings, including restricted teaching time, variation in delivery, and weak alignment between classroom content and the wider food environment. In addition, children’s food choices are shaped by influences beyond the classroom, such as family practices, food availability, peer norms, and food marketing. For this reason, programs based mainly on information delivery were less likely to produce lasting behavioral change than those that also included skills practice, family involvement, or environmental support [22, 39, 65, 67]. These findings suggest that school-based nutrition literacy programs should move beyond factual knowledge alone. Stronger programs are likely to combine repeated skill practice with support from the home and school environment. Future research should include longer follow-up, clearer reporting of intervention content and implementation, and wider use of validated multidimensional measures of nutrition literacy.

Strengths and limitations

Strengths

This review synthesized a broad range of programs aimed at improving nutrition literacy as an approach to obesity prevention. Multiple outcomes were included, such as BMI and BMI z-scores, nutrition knowledge, and changes in eating behaviors, reducing reliance on a single outcome. The programs were delivered in different formats, including classroom-based instruction, game-based activities, and family or community involvement, indicating that nutrition literacy can be implemented across varied educational settings. The inclusion of a meta-analysis allowed conclusions to be supported by pooled quantitative evidence rather than narrative findings alone.

Limitations

The included studies differed in their intervention designs, participant characteristics, and outcome assessments. First, these differences may have contributed to the high heterogeneity observed in the meta-analysis. The pooled estimates should therefore be interpreted with caution, and the findings may not be generalizable to all school-based nutrition literacy interventions or school-age student populations. Second, Publication bias could not be formally assessed for some outcomes because only a small number of studies were included in the meta-analysis. Therefore, publication bias cannot be ruled out. This should be considered when interpreting the pooled findings. Third, most studies also had short follow-up periods, which limited the assessment of longer-term effects. Although family involvement was included in several interventions, the effects were inconsistent. This may reflect the difficulty of linking school-based activities with food-related practices at home. In addition, knowledge of nutrition was assessed more often than actual dietary behavior. As a result, it remains difficult to determine whether the interventions changed students’ everyday eating practices beyond improvements in knowledge. Finally, this review included only studies published in English or Thai. This criterion enabled the review team to screen, extract, and interpret the included studies accurately. However, relevant studies published in other languages may have been missed, potentially introducing language bias.

Conclusions

This systematic review and meta-analysis showed that school-based interventions can improve nutrition literacy among school-age students. However, improvements in literacy outcomes were more consistent than changes in healthy eating behavior or anthropometric outcomes. The findings suggest that the design of interventions is important. Programs were more effective when they combined classroom teaching with practical activities and support from the family or school environment. Approaches that included repeated skill practice, such as label reading, food preparation, tasting, and food-related decision-making, appeared more promising than those based solely on information delivery.

These findings have implications for both practice and research. In practice, nutrition literacy programs should go beyond classroom-based knowledge transfer and include opportunities for students to apply their skills in everyday food contexts. In research, greater attention is needed to longer follow-up, clearer reporting of intervention content and implementation, and the use of validated multidimensional measures alongside behavioral indicators. Such steps help clarify which intervention components are most effective and whether improvements in nutrition literacy can lead to sustained changes in eating behavior among school-age students.

Authors' contributions

SP and AP conceptualized and designed the study, formulated the search strategy, and conducted the article search and selection. SP and SW performed the data analyses and contributed to the discussion and interpretation of the findings. SP, AP and SWB were the primary authors of the manuscript, with SL and RCh offering critical revisions. All authors reviewed and approved the final manuscript.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was supported by Praboromarajchanok Institute, Thailand, and partially supported by the Faculty of Graduate Studies and Graduate Studies of Mahidol University Alumni Association, Mahidol University, Bangkok, Thailand.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.