Data extracted included sample size in both groups, basic information of participants, duration of intervention, description for control group and intervention group, out come measures, results (short- or long-term), significance and relevance.
Data such as p-value, mean, standard deviation or standardized mean differences (SMDs) and their 95% confidence interval (CI) were also extracted and analyzed by the Comprehensive Meta-Analysis (CMA) software. Data were entered in the CMA software where studies reported changes baseline to post-intervention for each group. Three studies [18-20] reported the sample size both control group and intervention group, and independent group p-value; one study  reported mean, sample size and independent groups p-value.
Random effects model was used because four studies were measuring not the same treatment effects, variations between studies should be accounted.
The results of this meta-analysis are disappointing, with dietary interventions had some positive effects on the weight loss in both obesity and general children, but the effects were not significant. Studies by Jenson et al.  and Davis et al. reported significant results of BMI reduction in obesity children, but the sample size of these two studies were not large enough to compared with Rappaport et al., so these two papers can not strongly influence the finial results of the meta-analysis.
The intervention reported by Cunha et al. focus on change the diet habits, to increase the intake of fruit and vegetables of children, even the results of BMI was not significant different between the intervention arm and control arm, the intervention successfully in changed the fruit and vegetable intake of the intervention arm. Cunha et al. also mentioned the long-term (after 1-year follow-up) results, shown that the intervention group had lower percentage of obesity increase than the control group (5% to 8% in intervention group compared with 13% to 16% in control group). Intervention reported by Cunha et al. was successful in promote children or adolescents’ eating habit and lower their risk of obesity in adulthood .
Intervention in Cunha et...