This review concluded that interventions, at school or in the community, could help prevent excessive sedentary behaviour, preventing unfavourable health outcomes, for children and adolescents. There were concerns about the trial quality, variation, and the clinical significance of some of the results of the analyses, so the conclusions of this review may not be reliable.

To evaluate the effects of interventions, at school or in the community, to prevent excessive sedentary behaviour, in children and adolescents, on their sedentary behaviour and body mass index (BMI).

PubMed, EMBASE, Web of Science, PsycINFO, and Cochrane Database of Systematic Reviews were searched for published studies from between 1990 and March 2011. Search terms were reported. Bibliographies of relevant studies and reviews were searched for additional references.

Controlled trials, with at least one intervention and one control or non-intervention group, were eligible. Trials had to describe an intervention that aimed to reduce sedentary behaviour in children or adolescents (age range zero to 18 years). They could also target other behaviour, such as physical activity or diet (multiple health behaviour trials), if they explicitly stated the intervention elements that were aimed at sedentary behaviour. Trials had to report a sedentary behaviour or a weight-related outcome, or both. Those of high-risk populations (children or adolescents who were overweight or obese) and those that only targeted physical activity, without sedentary behaviour elements, were excluded.

Most of the included trials focused on children younger than 12 years, were school-based (with or without a home component), and used a multiple health behaviour intervention. Interventions often involved parents, and they lasted from 30 minutes to four years. Sedentary behaviour outcome measures varied (primarily sedentary screen time or television viewing) and were generally self-reported by the child. Most control groups received no intervention. Most trials were conducted in the USA.

Studies were selected by two reviewers independently, with disagreements resolved by a third reviewer.

Trial quality was assessed independently, by two reviewers, using the Cochrane risk of bias tool. Disagreements were resolved by consensus.

Data on the outcomes, in minutes of sedentary behaviour per day and BMI, were extracted to calculate mean differences. Trial authors were contacted for additional data as needed.

The authors did not state whether the data were extracted by more than one person.

Overall mean differences were calculated, using a random-effects meta-analysis. Differences between the intervention and control, in their results after the intervention, as well as in the change from baseline to after the intervention, were calculated.

Subgroup analyses were performed to investigate the effects of single (only targeting sedentary behaviour) versus multiple health behaviour interventions, and the effects of age (by group) and intervention setting, on the pooled estimates. Heterogeneity was assessed using Ι². Sensitivity analyses, using a fixed-effect model, were performed.

Thirty-four controlled trials (18,142 patients; range 20 to 2,190) were included; 30 were randomised, and four were not. Randomisation methods were adequately reported in 13 trials, and allocation concealment in seven trials. Eight trials reported blinding of the outcome assessor. Drop-out rates were reported and rated acceptable in 29 trials. Fifteen reported possible baseline differences between intervention and control groups. Follow-up was undertaken up to two years after the intervention.

Interventions aimed at preventing excessive sedentary behaviour significantly reduced sedentary behaviour by around 18 minutes per day (MD -17.95, 95% CI -26.61 to -9.28; 22 trials). There was some evidence of heterogeneity between study results (Ι²=41%). The difference in change from baseline to after intervention was similar (MD -20.44, 95% CI -30.69 to -10.20; 17 trials; Ι²=71%).

There was a mean difference in BMI of -0.25kg per m² (95% CI -0.40 to -0.09; 14 trials; Ι²=0) favouring the intervention group. The change from baseline to after intervention mean difference was -0.14 kg per m² (95% CI -0.23 to -0.05; 13 trials; Ι²=33%) in favour of the intervention group.

There were no significant differences in the effects on sedentary behaviour and BMI, between single and multiple health behaviour interventions. No moderating effects of age and intervention setting were found for either outcome.

Interventions, at school or in the community, could help prevent excessive sedentary behaviour, preventing unfavourable health outcomes, for children and adolescents.

The review question and selection criteria were clearly stated. Several bibliographic sources were searched, but not sources of unpublished studies. Steps were taken to minimise reviewer error and bias, when selecting the studies and assessing their quality. It was unclear if similar attempts were made during data extraction.

The quality assessment indicated several gaps in the quality of the trials. It raised notable concerns about selection, performance and detection biases, which could have overestimated the intervention effects. A relatively large number of trials and patients were included. The methods of analysis were described in detail and generally seem to have been appropriate. There was evidence of significant heterogeneity for some of the estimates. Potential sources of heterogeneity were explored, but the source remained unclear. The estimates for the reduction in BMI were only just statistically significant, and the clinical relevance of the pooled estimates was not discussed.

Given these concerns, the conclusions about the link between the prevention of sedentary behaviour and health outcomes may not be sufficiently cautious.

Practice: The authors did not state any implications for practice.

Research: The authors stated that studies with long-term follow-up were needed to evaluate the sustainability of the intervention effects. These studies should provide details on the intervention, and the types of outcome measures, to explore effective intervention elements. This should include the health behaviours targeted and the alternatives provided for sedentary behaviour. Mediation analyses could explore the relationship between sedentary behaviours and weight-related outcomes.

Funded by ZonMw, the Netherlands Organisation for Health Research and Development.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.