Twenty-five studies (evaluating 53,118 participants) were included in the review. All of the studies were controlled trials, with 18 incorporating some form of randomisation.
ssues relating to the quality of the intervention and implementation, including the exploration of differences and success mechanisms between effective and ineffective interventions, were reported in the paper. It was unclear what intervention components would succeed on a larger scale.
Seventeen studies (evaluating 42,301 participants) were deemed effective in terms of statistically significant differences between the intervention and comparison groups in measures of BMI, skin-folds, or both. Many of these included a physical activity component as part of the regular school curriculum, with others targeting reductions in television viewing. Four studies were effective for both outcome measures, with two studies deemed effective only for girls, and two studies equally effective for boys and girls. Another study was effective for girls but for BMI only. Three of the eight studies deemed ineffective reported small relative reductions in BMI or triceps skin-fold measures compared with controls. A further study was effective after 1 year, but this was not sustained at a 3-year follow-up.
Adverse outcomes were reported as those relating to unhealthy consequences of the intervention. One study showed a significant increase in underweight prevalence combined with a significant reduction in overweight and obesity prevalence. Three studies suggested a statistically significant increase in overweight prevalence as a result of the intervention.