Key messages
- Strategies to encourage adolescents to change their diet or activity levels (or both), to prevent them becoming overweight or developing obesity make no or very little difference to their body mass index (BMI; an estimate of the amount of body fat according to height and weight).
- Based on the very little information available about serious adverse events, there appears to be little or no effect of dietary or activity strategies, or both, to results in serious harms (e.g. injuries).
- Due to a lack of evidence, future research should focus on community settings (e.g. in youth clubs) and research involving adolescents with disabilities.
Why is preventing obesity in children and young people important?
More adolescents are developing overweight and obesity worldwide. Being overweight as an adolescent can cause health problems, and people may be affected psychologically and in their social life. Puberty and moving into adulthood is a challenging time, and many struggle with their mental health. Overweight adolescents are likely to be overweight or obese as adults and continue to experience poor physical and mental health.
What did we want to find out?
We wanted to find out if strategies to help adolescents modify their diet or activity (or both) were effective at preventing obesity. We also wanted to find out if these strategies were associated with any serious harms.
What did we do?
We searched many scientific databases to find studies that looked at ways of preventing obesity in children aged 12 to 18 years. We excluded studies only aimed at adolescents who were already overweight or living with obesity. However, we included studies where the sample was a whole group (e.g. a school), which may have included those living with overweight or obesity. We only included studies if the methods they used aimed to change the children's diet, their level of activity (i.e. increasing physical activity or reducing inactive time), or both. We looked only for studies that randomly placed children into groups receiving different strategies (which may include changing nothing). We assessed the rigour of the studies to get a sense of how confident we were in their results. We grouped studies together for analysis depending on whether they aimed to improve diet, activity, or both.
What did we find?
We found 74 studies that involved 83,407 children and young people. Sixty studies were based in high-income countries (e.g. USA and in Europe). In 57 studies, the strategies were tried in schools, although 12 were based in the home or other places. We found five studies based in community settings, such as youth groups. Fifty-one strategies were implemented for fewer than nine months, with the shortest intervention conducted over one visit and the longest over 28 months. Sixty-two studies declared non-industry funding; five studies were funded in part by industry (food suppliers, a PlayStation manufacturer, a gym equipment supplier, a healthcare device manufacturer and a private healthcare facility).
Our statistical analyses included results from 54 studies of 46,358 adolescents (20 studies did not report their results in a way that we could include them in our analyses). We found that adolescents who were helped with a strategy to change their diet or activity levels (or both) either did not reduce their BMI, or any reduction was meagre, compared to adolescents who were not given a strategy.
Only a few studies reported any possible harms of the interventions, and no serious harms were identified in these.
What are the limitations of the evidence?
Our confidence in the evidence is very low. It is difficult to be confident that funding more studies, at least more school-based studies, would produce a much higher level of confidence in the results. Four main factors reduced our confidence in the evidence.
1. Results were very inconsistent across the different studies.
2. A lot of the studies had limitations in how they were done (e.g. in some studies, the methods used to randomly place people into groups were not adequate or the results of some of the studies were not analysed correctly).
3. There were not enough studies reporting particular types of outcomes, such as BMI (an estimate of the amount of body fat according to height and weight) or zBMI (average BMI scores for a country) for a particular duration of follow-up to be certain about the results for some comparisons. Also, certain settings (e.g. community settings) were under-represented.
4. Results from some studies were not reported in a way that we could include them in our analyses (e.g. without any detail of the difference in change between the intervention and control groups) and this may have an impact on the results of our analyses.
This review does not provide sufficient information to be able to assess how well strategies work for adolescents with disabilities, or whether those implemented in community settings are effective.
How up to date is this evidence?
This review supersedes our previous review (Brown 2011). The evidence is up-to-date until February 2023.
The evidence demonstrates that dietary interventions may have little to no effect on obesity in adolescents. There is low-certainty evidence that activity interventions may have a small beneficial effect on BMI at medium- and long-term follow-up. Diet plus activity interventions may result in little to no difference. Importantly, this updated review also suggests that interventions to prevent obesity in this age group may result in little to no difference in serious adverse effects. Limitations of the evidence include inconsistent results across studies, lack of methodological rigour in some studies and small sample sizes.
Further research is justified to investigate the effects of diet and activity interventions to prevent childhood obesity in community settings, and in young people with disabilities, since very few ongoing studies are likely to address these. Further randomised trials to address the remaining uncertainty about the effects of diet, activity interventions, or both, to prevent childhood obesity in schools (ideally with zBMI as the measured outcome) would need to have larger samples.
Prevention of obesity in adolescents is an international public health priority. The prevalence of overweight and obesity is over 25% in North and South America, Australia, most of Europe, and the Gulf region. Interventions that aim to prevent obesity involve strategies that promote healthy diets or ‘activity’ levels (physical activity, sedentary behaviour and/or sleep) or both, and work by reducing energy intake and/or increasing energy expenditure, respectively. There is uncertainty over which approaches are more effective, and numerous new studies have been published over the last five years since the previous version of this Cochrane Review.
To assess the effects of interventions that aim to prevent obesity in adolescents by modifying dietary intake or ‘activity’ levels, or a combination of both, on changes in BMI, zBMI score and serious adverse events.
We used standard, extensive Cochrane search methods. The latest search date was February 2023.
Randomised controlled trials in adolescents (mean age 12 years and above but less than 19 years), comparing diet or 'activity' interventions (or both) to prevent obesity with no intervention, usual care, or with another eligible intervention, in any setting. Studies had to measure outcomes at a minimum of 12 weeks post baseline. We excluded interventions designed primarily to improve sporting performance.
We used standard Cochrane methods. Our outcomes were BMI, zBMI score and serious adverse events, assessed at short- (12 weeks to < 9 months from baseline), medium- (9 months to < 15 months) and long-term (≥ 15 months) follow-up. We used GRADE to assess the certainty of the evidence for each outcome.
This review includes 74 studies (83,407 participants); 54 studies (46,358 participants) were included in meta-analyses. Sixty studies were based in high-income countries. The main setting for intervention delivery was schools (57 studies), followed by home (nine studies), the community (five studies) and a primary care setting (three studies). Fifty-one interventions were implemented for less than nine months; the shortest was conducted over one visit and the longest over 28 months. Sixty-two studies declared non-industry funding; five were funded in part by industry.
Dietary interventions versus control
The evidence is very uncertain about the effects of dietary interventions on body mass index (BMI) at short-term follow-up (mean difference (MD) -0.18, 95% confidence interval (CI) -0.41 to 0.06; 3 studies, 605 participants), medium-term follow-up (MD -0.65, 95% CI -1.18 to -0.11; 3 studies, 900 participants), and standardised BMI (zBMI) at long-term follow-up (MD -0.14, 95% CI -0.38 to 0.10; 2 studies, 1089 participants); all very low-certainty evidence. Compared with control, dietary interventions may have little to no effect on BMI at long-term follow-up (MD -0.30, 95% CI -1.67 to 1.07; 1 study, 44 participants); zBMI at short-term (MD -0.06, 95% CI -0.12 to 0.01; 5 studies, 3154 participants); and zBMI at medium-term (MD 0.02, 95% CI -0.17 to 0.21; 1 study, 112 participants) follow-up; all low-certainty evidence.
Dietary interventions may have little to no effect on serious adverse events (two studies, 377 participants; low-certainty evidence).
Activity interventions versus control
Compared with control, activity interventions do not reduce BMI at short-term follow-up (MD -0.64, 95% CI -1.86 to 0.58; 6 studies, 1780 participants; low-certainty evidence) and probably do not reduce zBMI at medium- (MD 0, 95% CI -0.04 to 0.05; 6 studies, 5335 participants) or long-term (MD -0.05, 95% CI -0.12 to 0.02; 1 study, 985 participants) follow-up; both moderate-certainty evidence. Activity interventions do not reduce zBMI at short-term follow-up (MD 0.02, 95% CI -0.01 to 0.05; 7 studies, 4718 participants; high-certainty evidence), but may reduce BMI slightly at medium-term (MD -0.32, 95% CI -0.53 to -0.11; 3 studies, 2143 participants) and long-term (MD -0.28, 95% CI -0.51 to -0.05; 1 study, 985 participants) follow-up; both low-certainty evidence.
Seven studies (5428 participants; low-certainty evidence) reported data on serious adverse events: two reported injuries relating to the exercise component of the intervention and five reported no effect of intervention on reported serious adverse events.
Dietary and activity interventions versus control
Dietary and activity interventions, compared with control, do not reduce BMI at short-term follow-up (MD 0.03, 95% CI -0.07 to 0.13; 11 studies, 3429 participants; high-certainty evidence), and probably do not reduce BMI at medium-term (MD 0.01, 95% CI -0.09 to 0.11; 8 studies, 5612 participants; moderate-certainty evidence) or long-term (MD 0.06, 95% CI -0.04 to 0.16; 6 studies, 8736 participants; moderate-certainty evidence) follow-up. They may have little to no effect on zBMI in the short term, but the evidence is very uncertain (MD -0.09, 95% CI -0.2 to 0.02; 3 studies, 515 participants; very low-certainty evidence), and they may not reduce zBMI at medium-term (MD -0.05, 95% CI -0.1 to 0.01; 6 studies, 3511 participants; low-certainty evidence) or long-term (MD -0.02, 95% CI -0.05 to 0.01; 7 studies, 8430 participants; low-certainty evidence) follow-up.
Four studies (2394 participants) reported data on serious adverse events (very low-certainty evidence): one reported an increase in weight concern in a few adolescents and three reported no effect.