Efficacy of Exercise Intervention for Weight Loss in Overweight and Obese Adolescents: Meta-Analysis and Implications

Lee Stoner, David Rowlands, Ariel Morrison, Daniel Credeur, Michael Hamlin, Kim Gaffney, Danielle Lambrick, Anna Matheson

The global rise in obesity prevalence among children and adolescents has been linked to modifiable lifestyle factors, including lack of physical activity. However, no known meta-analysis has been conducted on the effects of exercise intervention on body composition and cardiometabolic risk factors in overweight and obese adolescents. The aim of this study was to (1) estimate whether exercise intervention meaningfully improves body composition and cardiometabolic risk factors in overweight and obese adolescents; and (2) discuss the implications of the findings in terms of primary healthcare provision and public health policy, using New Zealand as an exemplar context. Electronic databases (PubMed, Web of Science, SPORTDiscus, Google Scholar) from inception to May 2015. The reference lists of eligible articles and relevant reviews were also checked. Inclusion criteria were (1) randomized controlled trial; (2) structured exercise intervention, alone or combined with any other kind of intervention; (3) control group received no structured exercise or behavioural modification designed to increase physical activity; (4) participants overweight or obese (body mass index [BMI] ≥85th percentile); and (5) participants aged between 10 and 19 years. Initially, 1667 articles were identified. After evaluation of study characteristics, quality and validity, data from 13 articles (15 trials) involving 556 participants (176 male, 193 female, 187 unknown) were extracted for meta-analysis. Meta-analyses were completed on five body composition parameters and ten cardiometabolic parameters. Effect sizes (ESs) were calculated as mean differences, as well as standardized mean differences in order to determine effect magnitude. Exercise intervention reduced BMI (mean 2.0 kg/m(2), 95 % CI 1.5-2.5; ES moderate), body weight (mean 3.7 kg, 95 % CI 1.7-5.8; ES small), body fat percentage (3.1 %, 95 % CI 2.2-4.1; ES small), waist circumference (3.0 cm, 95 % CI 1.3-4.8; ES small), but the increase (improvement) in lean mass was trivial (mean 1.6 kg, 95 % CI 0.5-2.6). The response to an oral glucose tolerance test following exercise intervention was for a decrease in the area under the curve for insulin (mean 162 μU/μl, 95 % CI 93-231; ES large) and blood glucose (mean 39 mg/dl, 95 % CI 9.4-69; ES moderate). Improvements in the homeostatic model assessment were also noted (mean 1.0, 95 % CI 0.7-1.4; ES moderate) and systolic blood pressure (mean 7.1 mmHg, 95 % CI 3.5-10.7; ES moderate). The effects of exercise on total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, fasting insulin and fasting blood glucose were inconclusive. Most of the included trials were short term (6-36 weeks) and 13 had methodological limitations. Additionally, the meta-analyses for some of the secondary outcomes had a small number of participants or substantial statistical heterogeneity. The current evidence suggests that exercise intervention in overweight and obese adolescents improves body composition, particularly by lowering body fat. The limited available evidence further indicates that exercise intervention may improve some cardiometabolic risk factors.