Optimal exercise dose on Body Mass Index (BMI) in children and adolescents with overweight and obesity: a systematic review and bayesian model-based network meta-analysis

Abstract

Background: Exercise is widely used for obesity management, but the optimal doses of exercise for improving body mass index (BMI) in children and adolescents with overweight and obesity remain unclear. This study aimed to evaluate the dose‒response effects of various exercises on BMI in children and adolescents with overweight and obesity.

Methods: A systematic search was conducted in Web of Science (Core Collection), PubMed/MEDLINE, Embase, Scopus, and the Cochrane Library for randomized controlled trials on relevant studies, covering literature up to July 2024. Three independent reviewers assessed bias via the Cochrane risk of bias tool. The quality of evidence was assessed using Confidence in Network Meta-Analysis framework. A dose‒response network meta-analysis was used to evaluate the impact of various exercise interventions and explore dose‒response relationships. All outcomes were analyzed with the mean difference (MD) and 95% credible intervals (CrIs) calculated for combined statistics.

Results: The study included 39 publications with 1,814 participants, 47.3% female and a median age of 14 years. The intervention involved six exercise modalities. High-intensity interval training demonstrated a significant reduction in BMI (MD = -1.33, 95% CrIs - 2.01 to -0.66), followed by combined exercise (MD = -1.25, 95% CrIs - 1.93 to -0.61), moderate-intensity continuous training (MD = -1.09, 95% CrIs - 1.73 to -0.45), and mixed aerobic exercise (MD = -1.05, 95% CrIs - 1.67 to -0.42). There was an 'L'-shaped nonlinear dose-response relationship between total exercise dose and BMI, with 200 METs-min/day identified as the minimum exercise dose required to achieve a clinically meaningful reduction in BMI.

Conclusion: Low-quality evidence indicates that HIIT, CE, MAE, and MICT improve BMI in children and adolescents with overweight and obesity, with clarified exercise doses for clinical benefit. These findings are relevant for exercise prescription and public health policy.

Trial registration: CRD42024566450.

Keywords: Bayesian network meta-analysis; Children and adolescents; Dose response; Exercise; Obesity.

Fig. 1

PRISMA flow diagram of the study selection process

Fig. 2

A network plot depicting direct and indirect comparisons of network meta-analyses. b Bayesian ranking diagram. The surface under the cumulative ranking curve (SUCRA) value is used to assess the relative effectiveness of different exercise interventions. Higher SUCRA values indicate a better exercise effect. CG, control group; MICT, moderate-intensity continuous training. CE, combined exercise. RT, resistance training. MAE, mixed aerobic exercise; HIIT, high-intensity interval training

Fig. 3

Relative effects of different types of exercise on BMI compared with the control group. K, study arms; N, number of participants; SUCRA, subsurface under the cumulative ranking curve; CG, control group; MICT, moderate-intensity continuous training. CE, combined exercise. RT, resistance training. MAE, mixed aerobic exercise; HIIT, high-intensity interval training

Fig. 4

Dose‒response relationship between daily exercise and BMI in children and adolescents with overweight and obesity. MD, mean difference; METs, metabolic

Fig. 5

Dose‒response relationships between different types of exercise and BMI levels in overweight and obese children. MD, mean difference; METs, metabolic equivalents. CG, control group; MICT, moderate-intensity continuous training. CE, combined exercise. RT, resistance training. MAE, mixed aerobic exercise; HIIT, high-intensity interval training