Childhood Obesity Evidence Base: A pilot test of a novel meta-analytic method for evidence aggregation

To make better use of evidence from a wide variety of obesity prevention studies and initiatives, NCCOR is pilot testing a taxonomic meta-analytic approach for evidence aggregation. Commonly accepted meta-analytic methods for clinical trial results such as Cochrane and CONSORT have strict study inclusion such as meeting specific trial design and evaluation criteria to calculate an aggregated effect size. These systematic reviews for aggregating evidence may include studies of varying types; however, comparability of evidence, study components, and design types may or may not be examined.

The NCCOR Childhood Obesity Evidence Base (COEB) taxonomic meta-analytic approach categorizes intervention approach and components, recipients, and context, and weighs the rigor of the methods, evaluation strategy, and evidence. This approach allows comparison of the evidence from reports with varying levels of rigor and specificity. As well, it supports examination of the success of specific intervention components in targeted populations and circumstances and will provide a comparison to evidence generated by well-accepted meta-analytic methods.

In 2018, NCCOR members formed a workgroup and constituted an external expert panel (EEP) that includes experts in the field of childhood obesity with expertise in social determinants of health, health economics, the development of eating habits and food preference early in life, and the effects of obesity policy on youth and their families. The workgroup and EEP held two in-person meetings in September 2018 and February 2019 where they selected a topic focus—the prevention of overweight and obesity in children aged two to five years—and finalized the taxonomies for the analytic examination of identified evidence sources.

This project will result in:

• Searchable public database of studies and initiatives reviewed using the taxonomic structure and contextual factors, and an accompanying manual of procedure
  • As this is a pilot study, the database will not be exhaustive but rather representative of prevention efforts from the first 200 papers identified (representing approximately 50 studies from 2010 to 2019).
• Comparison and methods papers describing the meta-analytic process and rationale for its use
• Results paper of the findings from this taxonomic approach and comparison to evidence generated by well-accepted meta-analytic approaches

This pilot study using a novel approach to evidence aggregation has the potential to inform future research, initiatives, program development, and policy implementation to enhance the field’s impact on prevention of childhood obesity.