Prevention of childhood obesity is an international public health priority given the
significant impact of obesity on acute and chronic diseases, general health, development
and well‐being. The international evidence base for strategies to prevent obesity
is very large and is accumulating rapidly. This is an update of a previous review.
To determine the effectiveness of a range of interventions that include diet or physical
activity components, or both, designed to prevent obesity in children. We searched
CENTRAL, MEDLINE, Embase, PsychINFO and CINAHL in June 2015. We re‐ran the search
from June 2015 to January 2018 and included a search of trial registers. Randomised
controlled trials (RCTs) of diet or physical activity interventions, or combined diet
and physical activity interventions, for preventing overweight or obesity in children
(0‐17 years) that reported outcomes at a minimum of 12 weeks from baseline. Two authors
independently extracted data, assessed risk‐of‐bias and evaluated overall certainty
of the evidence using GRADE. We extracted data on adiposity outcomes, sociodemographic
characteristics, adverse events, intervention process and costs. We meta‐analysed
data as guided by the Cochrane Handbook for Systematic Reviews of Interventions
and presented separate meta‐analyses by age group for child 0 to 5 years, 6 to 12
years, and 13 to 18 years for zBMI and BMI. We included 153 RCTs, mostly from the
USA or Europe. Thirteen studies were based in upper‐middle‐income countries (UMIC:
Brazil, Ecuador, Lebanon, Mexico, Thailand, Turkey, US‐Mexico border), and one was
based in a lower middle‐income country (LMIC: Egypt). The majority (85) targeted children
aged 6 to 12 years.
Children aged 0‐5 years: There is moderate‐certainty evidence from 16 RCTs (n = 6261)
that diet combined with physical activity interventions, compared with control, reduced
BMI (mean difference (MD) −0.07 kg/m 2 , 95% confidence interval (CI) −0.14 to −0.01),
and had a similar effect (11 RCTs, n = 5536) on zBMI (MD −0.11, 95% CI −0.21 to 0.01).
Neither diet (moderate‐certainty evidence) nor physical activity interventions alone
(high‐certainty evidence) compared with control reduced BMI (physical activity alone:
MD −0.22 kg/m 2 , 95% CI −0.44 to 0.01) or zBMI (diet alone: MD −0.14, 95% CI −0.32
to 0.04; physical activity alone: MD 0.01, 95% CI −0.10 to 0.13) in children aged
0‐5 years.
Children aged 6 to 12 years: There is moderate‐certainty evidence from 14 RCTs (n
= 16,410) that physical activity interventions, compared with control, reduced BMI
(MD −0.10 kg/m 2 , 95% CI −0.14 to −0.05). However, there is moderate‐certainty evidence
that they had little or no effect on zBMI (MD −0.02, 95% CI −0.06 to 0.02). There
is low‐certainty evidence from 20 RCTs (n = 24,043) that diet combined with physical
activity interventions, compared with control, reduced zBMI (MD −0.05 kg/m2, 95% CI
−0.10 to −0.01). There is high‐certainty evidence that diet interventions, compared
with control, had little impact on zBMI (MD −0.03, 95% CI −0.06 to 0.01) or BMI (−0.02
kg/m2, 95% CI −0.11 to 0.06). Children aged 13 to 18 years: There is very low‐certainty
evidence that physical activity interventions, compared with control reduced BMI (MD
−1.53 kg/m2, 95% CI −2.67 to −0.39; 4 RCTs; n = 720); and low‐certainty evidence for
a reduction in zBMI (MD ‐0.2, 95% CI −0.3 to ‐0.1; 1 RCT; n = 100). There is low‐certainty
evidence from eight RCTs (n = 16,583) that diet combined with physical activity interventions,
compared with control, had no effect on BMI (MD −0.02 kg/m2, 95% CI −0.10 to 0.05);
or zBMI (MD 0.01, 95% CI −0.05 to 0.07; 6 RCTs; n = 16,543). Evidence from two RCTs
(low‐certainty evidence; n = 294) found no effect of diet interventions on BMI.
Direct comparisons of interventions: Two RCTs reported data directly comparing diet
with either physical activity or diet combined with physical activity interventions
for children aged 6 to 12 years and reported no differences.
Heterogeneity was apparent in the results from all three age groups, which could not
be entirely explained by setting or duration of the interventions. Where reported,
interventions did not appear to result in adverse effects (16 RCTs) or increase health
inequalities (gender: 30 RCTs; socioeconomic status: 18 RCTs), although relatively
few studies examined these factors. Re‐running the searches in January 2018 identified
315 records with potential relevance to this review, which will be synthesised in
the next update. Interventions that include diet combined with physical activity interventions
can reduce the risk of obesity (zBMI and BMI) in young children aged 0 to 5 years.
There is weaker evidence from a single study that dietary interventions may be beneficial.
However, interventions that focus only on physical activity do not appear to be effective
in children of this age. In contrast, interventions that only focus on physical activity
can reduce the risk of obesity (BMI) in children aged 6 to 12 years, and adolescents
aged 13 to 18 years. In these age groups, there is no evidence that interventions
that only focus on diet are effective, and some evidence that diet combined with physical
activity interventions may be effective. Importantly, this updated review also suggests
that interventions to prevent childhood obesity do not appear to result in adverse
effects or health inequalities. The review will not be updated in its current form.
To manage the growth in RCTs of child obesity prevention interventions, in future,
this review will be split into three separate reviews based on child age. Do diet
and physical activity strategies help prevent obesity in children (aged 0 to 18 years)?
Background More children are becoming overweight and obese worldwide. Being overweight
as a child can cause health problems, and children may be affected psychologically
and in their social life. Overweight children are likely also to be overweight as
adults and continue to experience poor physical and mental health. Searching for studies
We searched many scientific databases to find studies that looked at ways of preventing
obesity in children. We included studies aimed at all ages of children. We only included
studies if the methods they were using were aimed at changing children's diet, or
their level of physical activity, or both. We looked only for the studies that contained
the best information to answer this question, ‘randomised controlled trials’ or RCTs.
What we found We found 153 RCTs. The studies were based mainly in high‐income countries
such as the USA and European countries although 12% were in middle‐income countries
(Brazil, Ecuador, Egypt, Lebanon, Mexico, Thailand and Turkey). Just over half the
RCTs (56%) tried out strategies to change diet or activity levels in children aged
6 to 12 years, a quarter were for children aged 0 to 5 years and a fifth (20%) were
for teenagers aged 13 to 18. The strategies were used in different settings such as
home, preschool or school and most were targeted towards trying to change individual
behaviour. Did they work? One widely accepted way of assessing if a child is overweight
is to calculate a score based on their height and how much they weigh, and relating
this to the weight and height of many children their age in their country. This is
called the zBMI score. We found 61 RCTs involving over 60,000 children, that had reported
zBMI scores. Children aged 0 to 5, and children aged 6 to 12 who were helped with
a strategy to change their diet or activity levels reduced their zBMI score by 0.07
and 0.04 units respectively compared to children who were not given a strategy. This
means these children were able to reduce their weight. This change in zBMI, when provided
to many children across a whole population, is useful for governments in trying to
tackle the problems of obesity in children. Strategies to change diet or physical
activity, or both, given to adolescents and young adults aged 13 to 18 years, did
not successfully reduce zBMI. We looked to see if the strategies were likely to work
fairly for all children, for example girls and boys, children from wealthy or less
wealthy backgrounds, children from different racial backgrounds. Not many RCTs reported
this, but in those that did, there was no indication that the strategies increased
inequalities. However we could not find enough RCTs with this information to help
us answer this question. We also looked to see if children were harmed by any of the
strategies, for example by having injuries, losing too much weight or developing damaging
views about themselves and their weight. Not many RCTs reported this, but in those
that did, none reported any harms from children who had been given strategies to change
their diet or physical activity. We looked at how well the RCTs were done to see if
they might be biased. We decided to downgrade some information based on these assessments.
The quality of the evidence was ‘moderate’ for children aged 0 to 5 for zBMI, ‘low’
for children aged 6 to 12 and moderate for adolescents (13 to 18). Our conclusions
Strategies for changing diet or activity levels, or both, of children in order to
help prevent them becoming overweight or obese are effective in making modest reductions
in zBMI score in children aged 0 to 5 years and in children aged 6 to 12 years. This
can be useful to parents and children concerned about children becoming overweight.
It can also be useful for governments, trying to tackle a growing trend of children
who are becoming obese or overweight. We found less evidence for adolescents and young
people aged 13 to 18, and the strategies given to them did not reduce their zBMI score.