Interventions aimed at reducing obesity in early childhood

Background: Obesity is a growing problem even in very young childhood, resulting in high costs for individuals and society. As a response, numerous obesity prevention and intervention programs have been developed. Previous research has shown that early intervention programs are more effective when parents are involved, but the effectiveness of specific aspects of programs with parental involvement has not been investigated. This meta-analysis aims to investigate the features related to the effectiveness of different types of obesity intervention programs involving parents and targeting young children (0–6-year-olds).

Methods: The Web of Science, PubMed, PsycInfo, CINAHL, and ERIC databases were searched for childhood obesity prevention and intervention programs involving parents. Data were analyzed using the Comprehensive Meta-analysis (CMA) software.

Results: Fifty studies with effect sizes measured at short-term follow-up (within 3 months from the end of the intervention) and 26 studies with effect sizes measured at long-term follow-up (all reported in a total of 49 publications) were identified. The combined effect size of interventions was small but significant at short-term follow-up (d = .08, p < .01). The results suggested the presence of a potential publication bias in studies providing results at long-term follow-up, with a nonsignificant adjusted effect size (d = .02), which indicated that obesity interventions were not effective at long-term follow-up. Multivariate meta-regression analyses showed that interventions were more effective when including either interactive sessions or educational materials as opposed to those including both interactive sessions and noninteractive educational materials. No other moderators regarding sample characteristics, study design, or methodological quality were significant.

Conclusion: Interventions targeting young children that require parental involvement are effective at short-term follow-up, specifically when interventions include one mode of intervention rather than two. However, results were not retained in the long run.