Systematic review
Effectiveness of interventions for overweight or obesity in children and adolescents
Systematic review
|Published
We have assessed the effectiveness of interventions to reduce body mass index (BMI) and/or body mass index standard deviation score (BMI z-score) in children and adolescents with overweight or obesity.
Download
Key message
Approximately 14 % of children and adolescents in Norway are overweight or obese. Overweight and obesity can affect physical and psychosocial health. Treatment may provide several health benefits.
We have assessed the effectiveness of interventions to reduce body mass index (BMI) and/or body mass index standard deviation score (BMI z-score) in children and adolescents with overweight or obesity and conclude that:
- Combined lifestyle interventions involving structured strategies for change in diet and physical activity may reduce BMI and BMI z-score compared with little or no treatment. We have moderate confidence in the documentation.
- Increased physical activity may lead to decrease in BMI compared with little or no treatment. We have moderate confidence in the documentation.
- We do not know if low glycemic index or load diet may reduce BMI z-score more than low-fat diet and diet with high glycemic index. We were unable to conclude on other dietary interventions.
- Metformin, in addition to lifestyle interventions, may reduce BMI z-scores more than placebo in obese adolescents with impaired glucose tolerance. We have low confidence in the documentation.
Surgical treatment may reduce BMI and BMI z-scores more than lifestyle interventions in adolescents with obesity. We have moderate confidence in the documentation.
Summary
Overweight and obesity in children and adolescents has increased significantly worldwide during the past 30 years. In Norway, about 14 % of children and adolescents are overweight or obese, defined as age and gender-adjusted body mass index (ISO BMI) over 25. Overweight and obesity can affect both physical and psychosocial health. In children with severe obesity, an accumulation of cardiovascular risk factors has been seen. Childhood obesity is a risk factor for obesity in adulthood. There is also a connection between obesity in adolescence and morbidity and mortality in adulthood.
We assessed the effects of lifestyle interventions, medical interventions and surgery in children and adolescents with overweight or obesity.
Method
We searched systematically for literature in nine databases (Cochrane Database of Systematic Reviews, Cochrane Central Register of Controlled Trials Register (CENTRAL), Medline (Ovid), Embase (Ovid), CINAHL via EBSCOhost, PsycINFO, ISI Web of Science, DARE (Database of abstracts of reviews of effects) and HTA). To be included the studies had to be systematic reviews or RCTs, include children and adolescents with overweight or obesity, assess effectiveness of lifestyle interventions (combined lifestyle interventions (changes in diet, physical activity, behavior changes, minimum two components), physical activity interventions, and dietary interventions), medical interventions and surgery. The interventions had to be compared with standard, minimal or no treatment, and measure body mass index (BMI) and/or BMI standard deviation score (BMI z-score, indicates how many standard deviations a child's BMI is above or below average BMI value for their age group and gender in a given reference population). Lifestyle interventions could be aimed at children and adolescents with or without their family, and take place at school, in primary care, hospitals or other health institutions.
Two researchers assessed abstracts and full text articles independently. We summarized data in text and tables. Where possible we used meta-analyses and calculated mean difference between groups. GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) was used to assess the quality of the overall documentation expressed as high, moderate, low and very low confidence in the effect estimate.
Results
We included 81 randomized controlled trials on a) lifestyle interventions (n=74), including combined lifestyle interventions (n=55), physical activity interventions (n=7), dietary interventions (n=12), b) medical interventions (n=6) and c) surgical intervention (n=1).
Combined Lifestyle Intervention: Complex lifestyle intervention aimed at children and adolescents, or whole families may give a significant reduction in BMI after six months (mean difference (MD) -0.99 [95 % CI -1.36 to -0.61]), 12 months (MD -0.67 [95 % CI -1.01 to -0.32]) and 24 months (MD -0.96 [95 % CI -1.63 to -0.29]) and BMI z-score after six months (MD -0.12 [95 % CI -0.17 to -0.06]), 12 months (MD -0.16 [95 % CI -0.21 to -0.10]) and 24 months (MD -0.16 [95 % CI -0.21 to -0.10]) compared to standard, minimal or no treatment. We have moderate confidence in the documentation, except for the results of BMI at 24 months where we have low confidence and the results of BMI z-score where we have high confidence.
Physical activity interventions: We found that increased physical activity may give a significant reduction in BMI (mean difference (MD) -0.72 [95 % CI -1.18 to -0.26]), but not of BMI z-score (MD -0.03 [95 % CI -0.08 to 0.01]), compared with minimal or no treatment after six months. We have moderate confidence in the documentation.
Dietary interventions: We found that it is unclear whether diets with low glycemic index or load may reduce BMI z- score (MD -0.10 [95 % CI -0.18 to -0.02]), and not BMI (MD -0.80 [95 % CI -2.27 to 0.67]), more than low-fat diet and diet with high glycemic index after six months. We have very low confidence in the documentation. We were unable to conclude on other dietary interventions.
Medical intervention: Treatment with orlistat in combination with lifestyle intervention may possibly reduce BMI (we were unable to estimate effect due to incomplete reporting of data in studies) after six months for adolescents with obesity. BMI z-scores were not reported. Metformin combined with lifestyle interventions may possibly lead to a reduction in BMI z-score (MD -0.08 [95 % CI -0.12 to -0.04]), but not BMI (MD -0.45 [95 % CI -1.88 to 0.98]), compared with placebo in obese adolescents with impaired glucose tolerance after six months. We have moderate confidence in the documentation on BMI z-score for metformin, and very low confidence for BMI. The treatments had some side effects in the gastrointestinal tract, but none of them serious.
Surgical intervention: One study on surgery compared with intensive lifestyle intervention found significant reduction in BMI (MD -11.40 [95 % CI -13.48 to -9.32]) and BMI z-score (MD -0.85 [95 % CI -1.12 to -0.58]) after two years. We have moderate confidence in the documentation. The treatment caused some complications related to the surgery, and some re-operations.
Discussion
There were variations in interventions, comparisons, intervention length and intensity, follow-up, number of participants in each study and the number of participants who completed the studies. Several studies found a reduction in BMI and/or BMI z-score of the intervention from start to end. The quality of the documentation ranged from moderate to very low, and was mainly downgraded due to risk of bias, few participants and high statistical heterogeneity.
Limitations in our review is that the litterature search was last updated in February 2015 and that we only used the BMI and BMI z-scores as outcome measures. By exclusively using these outcome measures, we may have missed other important effects of the interventions.
Conclusion
Based on our results we conclude that:
- Combined lifestyle interventions involving structured strategies for change in diet and physical activity may reduce BMI and BMI z-score compared with little or no treatment.
- Increased physical activity may lead to decrease in BMI compared with little or no treatment.
- It is unclear whether low glycemic index or load diet may reduce BMI z- score more than low-fat diet and diet with high glycemic index. We were unable to conclude on other dietary interventions.
- Metformin, in addition to lifestyle interventions, may reduce BMI z-scores more than placebo in obese adolescents with impaired glucose tolerance.
- Surgical treatment may reduce BMI and BMI z-scores more than lifestyle interventions in adolescents with obesity.
Further research should focus on long-term effects of lifestyle interventions (beyond two years), safety and long-term effects of medical interventions versus lifestyle interventions and long-term effects of bariatric surgery versus lifestyle interventions, and other outcomes than BMI and BMI z-scores.