A Comparison of Methods for Testing and Implementing Community Health Interventions in Childhood: A Realist Review.

Background: Innovative methods to test healthcare interventions have recently emerged to help provide more targeted, effective, and scalable interventions. Given the importance of the early years for children's development, improved interventions for vulnerable children and families have become public health imperatives. Traditional randomized control trials (RCTs), considered the gold standards, have serious limitations due to high costs, time demands, and issues with the generalizability of the results. Indeed, new accelerated methods are being considered to improve the efficiency of RCTs. Thus, we compared innovative methods with RCTs in their ability to test and implement interventions. We also provided recommendations for best practices in the child-health research. Methods: A realist review was undertaken to identify and make recommendations on what works for whom and under what circumstances. This realist meta-review was conducted as an umbrella review of reviews, supplemented by a synthesis of the targeted grey-literature, to report both peer-reviewed and practice-based evidence on evaluation methods for community child-health interventions. We searched electronic databases, including MEDLINE, PubMed, EMBASE, PsycINFO, CINAHL, and the grey literature, and provided references. We identified, selected, and appraised sources if they were (1) written in English, (2) answered our research question, (3) described/criticized a method for intervention evaluation, and (4) focused on community-based health interventions. Results: For our final analysis, out of 5167 identified documents, we selected those that criticized or reviewed RCTs (n = 13) and innovative methods (n = 31). Following Pawson's recommendations, we developed an extraction tool to promote a consistent approach and assessed to what degree each method enabled evaluation, was theory driven, offered clear guidelines, provided clear methods or tools, fostered innovation, was fast and generalizable, worked for who and under what circumstances, and focused on children and child-related research. Conclusions: Innovative and accelerated methods offer promising alternatives to the traditional RCTs for evaluating community-based child health interventions. Among these, the Innovate, Develop, Evaluate, Adapt, and Scale (IDEAS) method emerged as the most integrative and context-sensitive approach to evaluate early interventions in a variety of settings. Other innovative methods were not well-developed, compromising the internal validity of studies focused on promoting children's health in community settings. Graphical abstract synthesizes the phases of RCTs and contrasts them with IDEAS.