School-based obesity prevention programs: an evidence-based review

Journal article citation:
Obesity, 16(5), 2008, pp.1009-1018.
Nature Publishing Group

A systematic review of experimental and quasi-experimental school-based interventions to prevent obesity is presented. Of the 14 eligible studies, 11 combined nutrition and physical activity components, two focused on physical activity and one focused on nutrition. Ten were judged to provide weak (grade 2) evidence. Twelve studies reported significant improvements in at least one measure of dietary intake, physical activity and/or sedentary behaviour. The ability to draw strong conclusions was limited by the small number of studies and by methodological concerns such as small sample sizes. Qualitative analysis suggests that programmes grounded in social learning may be more appropriate for girls, while boys respond more readily to structural and environmental interventions enabling physical activity. Future research needs to embody high quality evaluation protocols.

Extended abstract:

KROPSKI Jonathan A.; KECKLEY Paul H.; JENSEN Gordon L.;


School-based obesity prevention programs: an evidence-based review.

Journal citation/publication details

Obesity, 16(5), 2008, pp.1009-1018.


Fourteen studies of long term, school-based programmes designed to prevent obesity and overweight are reviewed. The quality of the evidence base is poor, confirming the findings of earlier reviews, and this prevents the drawing of strong conclusions about efficacy. There are, however, indications that girls and boys respond better to different types of intervention. Further, better designed and powered studies are needed to provide a firm basis for policy and practice.


Childhood overweight and obesity is a significant problem in the USA, and many states have legislation in place offering guidance to schools on nutrition and physical activity. Three reviews, and a set of UK (NICE) guidelines, have recently been published on school-based obesity prevention programmes but none offer conclusive evidence on whether they are effective. This review focuses on studies of long term, school-based programmes with the aim of offering some guidance on how best future evaluations should be conducted.


What sources were used?
PubMed, Biological Abstracts and Education Abstracts were searched. In addition, the references of previous reviews and studies selected for this review were checked, and unnamed researchers in the field of childhood obesity were contacted.

What search terms/strategies were used?
Searches covered the period 1 January 1990 to 31 December 2005, and terms ‘included’: school; obesity; overweight; weight; prevention; nutrition; physical activity; trial; BMI; blood pressure; cholesterol; environment; child; adolescent; food choices; metabolic syndrome.

What criteria were used to decide on which studies to include?
Eligible studies were: experimental or quasi-experimental; addressed school-based curricular and/or environmental interventions delivered to both normal weight and overweight children; reported primary outcomes as BMI (Body Mass Index) values; and reported outcomes at least six months post-baseline.

Who decided on their relevance and quality?
No details are given of the initial results of the searches or their subsequent refinement. Methodological quality was assessed independently by two authors using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) working group criteria, which are summarised in Tables 1 and 2. Disagreements were resolved through discussion.

How many studies were included and where were they from?
Fourteen studies were reviewed, summarised in Table 3 which gives information on study design, sample size, age range, duration, intervention, outcome measures and attrition rates. No details of geographical setting are given but the text confirms that studies from the UK (3), Thailand (1), Germany (1), Chile (1) and Australia (1) were included. It is assumed that the remainder were from North America.

How were the study findings combined?
The findings are presented in narrative form and summarised in Table 4, which covers the effect on overweight, the effect on health behaviours, methodological limitations, and evidence grade.

Findings of the review

Of the 14 studies, 11 included both nutritional and exercise components, two focused solely on exercise and one solely on nutrition. Ten of the studies were classed as Grade 2 (low quality), two as Grade 3 (moderate quality) and one as Grade 4 (high quality).

Nutritional interventions
One small scale, moderate quality randomised controlled trial from the UK evaluated a programme to reduce the intake of fizzy drinks. Compared to baseline, there was a significant reduction in consumption but BMI outcomes were not reported.

Physical activity interventions
One low quality US study showed significant improvement in moderate-to-vigorous physical activity (MVPA) time during physical education classes but there was no effect for measures of overweight. Another moderate quality study of pre-school children in Thailand found a significant reduction in the odds ratio for increased BMI among girls, but no other measures of overweight achieved statistical significance.

Combined nutritional/physical activity interventions
One US study in this category was graded high quality and showed that a combination of educational approaches to reduce TV viewing, increase MVPA, reduce consumption of high fat foods and increase consumption of fruit and vegetables to five servings a day produced a significant reduction in the odds of girls (but not boys) being overweight.

Three studies were classed as of moderate quality, with two evaluating the Child and Adolescent Trial for Cardiovascular Health programme, which combined reductions in the fat content of school meals with initiatives to increase the daily amount of MVPA. One found a significant positive changes in both these areas, but no impact on overweight. The other showed a significant relative reduction in overweight prevalence of 11% in girls and 8% in boys. The third moderate quality study evaluated an environmental intervention. It found a significant relative reduction in BMI for boys and a significant increase in their MVPA, but no significant effect for girls.

The remaining studies in this category were graded as low quality. Three, including two from the UK, reported significant improvements in targeted behaviours (fruit and/or vegetable intake, and physical activity) but no significant impact on overweight. One Chilean study found significant reductions in BMI for boys only, while two studies from Germany and Australia found significant reductions in triceps skin-fold thickness, as well as improvements in targeted behaviours such as physical activity, TV viewing and dietary measures.

The equivocal results of previous reviews in the field are summarised, and the authors find that their own, more focused, review is also unable to offer unambiguous guidance. The use of the GRADE quality criteria reveals a range of methodological weaknesses in the bulk of the reviewed studies, including Type I and Type II errors that can be introduced when studies are under-powered: the first of these relates to the observation of a difference where none in fact exists; and the second to the failure to detect a difference when one does exist. Other methodological inadequacies included the use of study populations that were not demographically representative, in that they had an above average prevalence of overweight.

Authors' conclusions

‘At present, the overall quality of evidence must be considered weak’. The review does, however, reveal ‘a number of qualitative trends warranting comment’ and further investigation. First, girls and boys do not necessarily respond in the same way to an intervention, with the former appearing to do better in programmes that are based on social learning and the latter in programmes that emphasise structural and environmental changes. Second, interventions that include family components have not so far proved more effective than those that do not, despite the commonsense belief that the home environment has a significant impact on a child’s weight status. Third, changes in diet or physical activity have not ‘reproducibly led to improvements in overweight outcomes’ although this may be a result of the relatively short time frames of the reviewed studies. And, finally, programmes directed at older children (over the age of ten) may be more effective in reducing BMI or obesity prevalence than those directed at younger children.

Implications for policy or practice

Recent policy initiatives in the USA (and elsewhere) have focused on the consumption of soft drinks in schools. This responds to an issue of serious public concern, but may ‘distract public and financial attention away from other relevant health promotion efforts offering promise of lasting impact on health behaviors and outcomes.’ Attempts to reduce the consumption of soft drinks need to be carefully evaluated in well designed and adequately powered trials, as do other interventions.

Subject terms:
nutrition, obesity, physical exercise, prevention, school children, boys, girls;
Content type:
systematic review
Journal home page
ISSN online:
ISSN print:

Key to icons

  • Free resource Free resource
  • Journal article Journal article
  • Book Book
  • Digital media Digital media
  • Journal Journal

Give us your feedback

Social Care Online continues to be developed in response to user feedback.

Contact us with your comments and for any problems using the website.

Sign up/login for more

Register/login to access resource links, advanced search and email alerts