Interventions with children and parents to improve physical activity and body mass index: a meta-analysis.

OBJECTIVE Examine the effect of interventions with parents and children on children's physical activity and body mass index (BMI). DATA SOURCE Computerized searches for intervention studies published between 1990 and 2011 used multiple ProQuest databases, including unpublished dissertations and theses to minimize publication bias. STUDY INCLUSION AND EXCLUSION CRITERIA English-language, intervention-testing studies of children, parents, or families with outcomes of physical activity or BMI were retrieved from peer-reviewed journals, dissertations, and theses. Eliminated studies had no control or comparison group; had no continuous outcome variable; had no physical activity/exercise and/or BMI as outcomes; or had incomplete statistics necessary for meta-analysis (means, standard deviations, or confidence intervals). DATA EXTRACTION Twenty-one studies met inclusion criteria. Quality criteria were control group, objective outcome variable measure, clarity of variable definitions, and number and reason for subject withdrawal. DATA SYNTHESIS Meta-analysis on the raw difference of means estimated mean weighted effect size (MWES) assessed dispersion of effects and computed a summary effect. RESULTS MWES for interventions with parents and children on physical activity (Z = 2.92; confidence interval [CI] = .09 to .48; p = .002) and on BMI for interventions with children alone (Z = -2.10; CI = -.16 to -.01; p = .02) was significant. CONCLUSION A significant effect on physical activity but not on BMI was found when interventions included both parents and their children.