Welcome to The Community Guide! Let us know what you think of the website by completing this quick survey.

Obesity: Meal or Fruit and Vegetable Snack Interventions to Increase Healthier Foods and Beverages Provided by Schools


What the CPSTF Found

About The Systematic Review

In 2013, the Agency for Healthcare Research and Quality (AHRQ) conducted a meta-analysis on the effectiveness of childhood obesity prevention programs implemented in 6 intervention settings. The CPSTF finding is based on a subset of studies from the review that focused on dietary-only approaches in schools (Wang et al., 2013; 15 studies, search period through August 2012) combined with more recent evidence (36 additional studies, search period August 2012 to January 2017).

This review was conducted on behalf of the CPSTF by a team of specialists in systematic review methods, and in research, practice and policy related to obesity prevention and control. The 2016 findings about school interventions to prevent obesity update and replace the 2003 CPSTF findings on School-Based Programs Promoting Nutrition and Physical Activity pdf icon [PDF - 1.15 MB] and School-Based Programs to Prevent Obesity pdf icon [PDF - 679 kB].


Healthy eating during childhood is important (CDC , 1998; Dietary Guidelines Advisory Committee, 2010). Schools can play an important role in preventing obesity by providing nutritious and appealing foods and beverages (CDC, 2016a; CDC, 2011).

Summary of Results

More details about study results are available in the CPSTF Finding and Rationale Statement pdf icon [PDF - 1.24 MB].

The systematic review included 25 studies with 26 study arms.

Weight-Related Outcomes

  • Overweight/obesity prevalence –median decrease of 9.6 percentage points (5 studies)
  • BMI z-score – median increase of 0.01 (not significant; 3 studies)

Diet-Related Outcomes

  • Fruit or vegetable intake – 20.0% increase (9 study arms)

Summary of Economic Evidence

More details about study results are available in the CPSTF Finding and Rationale Statement pdf icon [PDF - 1.24 MB].

A systematic review of economic evidence indicates that meal interventions and fruit and vegetable snack interventions are cost-effective. All monetary values are reported in 2016 U.S. dollars.

The economic review included 7 studies (4 from the United States, 1 from Norway, 1 from the Netherlands, and 1 from Taiwan).

  • Median intervention cost per student per year for fruit and vegetable snack interventions: $50 (3 studies)
  • Incremental intervention cost per student per year to provide school meals that meet nutrition guidelines: $198 and $624 (2 studies)
  • Lifetime savings per student in healthcare costs due to two fruit and vegetable snack interventions: $28 and $17 (1 study)
  • Annual savings per student in healthcare cost due to improved nutritional content of school meals: $17 (1 study)
  • Cost per quality adjusted life year (QALY) gained was $10,800 (1 study). This estimate is less than $50,000 – a benchmark for cost-effectiveness.
  • Costs per disability adjusted life year (DALY) averted were $8,014 and $14,934 (2 studies). Both estimates are considered cost-effective based on a per capita annual income of $49,390 in the Netherlands.


Based on results for interventions in different settings and populations, findings should be applicable to the following:

  • Elementary and middle school-aged children
  • Girls and boys
  • Students from different racial and ethnic backgrounds
  • Students from different income levels
  • Urban, suburban, and rural environments

Evidence Gaps

Additional research and evaluation are needed to answer the following questions and fill existing gaps in the evidence base.

  • Which intervention activities, school breakfast or lunch, fresh fruit and vegetable programs, or combinations of activities are most effective? Which activities are critical to success?
  • What are the cumulative effects of adding intervention components? Is a single component equally effective?
  • In order to increase comparability what are the best measures for dietary intake outcomes?
  • Do children act as agents of change by discussing changes in the school environment with parents? Do parents incorporate healthier dietary habits at home?
  • How often do schools implement interventions with fidelity? What amount of training is needed for faculty?
  • Does effectiveness vary by students’ age group?
  • Are national, state, or local policies most effective?
  • For studies reporting on milk products and alternatives to dairy, what is the fat content of these foods and beverages?
  • Do interventions lead to other health benefits such as improvements in cholesterol or blood pressure?
  • What does the intervention cost to implement and what are the major drivers of cost?
  • What are long term benefits of averted healthcare cost and improved productivity in adulthood associated with interventions shown to improve nutrition and prevent obesity in childhood?

Study Characteristics

  • Study designs included group randomized controlled trials (7 studies), group non-randomized (2 studies), repeat cross-sectional with comparison (1 study), time series (1 study), single group before-after (4 studies), repeat cross-sectional (3 studies), or post-test only with comparison (7 studies).
  • Studies were conducted in the United States (12 studies), Canada (3 studies), Norway (3 studies), the United Kingdom (1 study), the Netherlands (1 study), Spain (1 study), Australia (1 study), Denmark (1 study), Greece (1 study), and one study collected data from 3 countries (Norway, the Netherlands, and Spain).
  • Studies were conducted in schools alone (23 studies) or in schools plus one or more additional settings (2 studies).
  • Studies were conducted in elementary schools (11studies), middle schools (6 studies), high school (2 studies), or a combination of elementary, middle, or high schools (6 studies).
  • Studies were set in urban (3 studies), rural (2 studies), or a combination of urban, suburban or rural (9 studies) settings.
  • About half of each study population was female (22 studies; 3 studies did not provide information).
  • Twelve studies reported race/ethnicity with study populations identifying as white (median: 58.9%; 9 studies), black (median: 15.4%; 8 studies), Hispanic (median: 18.5%; 8 studies), First Nation (100.0%; 2 studies), and other (100%; 1 study).