Background
To combat the high prevalence of physical inactivity among children, there is an urgent need to develop and implement real-world interventions and policies that promote physical activity (PA) and reduce sedentary behaviour (SB). To inform policy makers, the current body of evidence for children’s PA/SB interventions needs to be translated.
Objectives
The current systematic review and meta-analysis aimed to identify modifiable determinants of device-measured PA and SB targeted in available intervention studies with randomized controlled trial (RCT) and controlled trial (CT) designs in children and early adolescents (5–12 years) and to quantify the effects of the interventions within their respective settings on the determinants of PA/SB and the outcomes PA and SB.
Methods
A systematic search was conducted in MEDLINE, PsycINFO, Web of Science, SPORTDiscus and CENTRAL. Studies were considered if they were randomized controlled trials (RCTs) or controlled trials (CTs), included children and/or early adolescents (5–12 years; henceforth termed children), measured PA and/or SB using device-based methods and measured PA and/or SB and determinants of PA/SB at least at two timepoints. Risk of bias was assessed using the Cochrane Risk of Bias Tool for Randomised Trials (RoB2) for RCTs and Risk of Bias in Non-randomised Studies of Interventions (ROBINS-I) for CTs. The quality of the generated evidence was assessed using Grading of Recommendations Assessment, Development and Evaluation (GRADE). Robust Bayesian meta-analysis was conducted to quantify the effects of the interventions on the determinants of PA/SB, and the outcomes PA and SB, stratifying by study design, duration of PA/SB measurement, intervention setting and duration of follow-up measurement. Study characteristics and interventions were summarized.
Results
Thirty-eight studies were included with a total sample size of n = 14,258 (67% girls). Settings identified were school, family/home, community and combinations of these. The review identified 38 modifiable determinants, spanning seven categories on individual, interpersonal and physical environmental levels, with 66% of determinants on the individual level. Overall, the results indicated trivial-to-moderate effects of the interventions on the determinants of PA and SB, with mostly trivial level of evidence for the presence of an effect (as indicated by a small Bayes factor; BF10 < 3.00). The exceptions were moderate effects on parental PA modelling in the family/home setting and SB measured during specific parts of the school day. Higher quality of evidence was found in the family/home setting compared with other settings.
Discussion
Overall, the results indicated that interventions have neither been effective in modifying the determinants of PA/SB, nor changing the PA/SB outcomes in children. In general, the approach in the current review revealed the breadth of methodological variability in children’s PA interventions. Research is needed to address novel approaches to children’s PA research and to identify potential determinants to inform policy and future interventions.
Registration
International prospective register of systematic reviews (PROSPERO): CRD42021282874.