Role of research-practice-policy partnerships in optimizing prevention science and the use of research evidence: Optimizing Implementation at Scale in Low-Resource Settings Using Routine and Trial Data from the Parenting for Lifelong Health Interventions

This abstract was presented at the 2018 Society for Prevention Research Annual Meeting which was held May 29 – June 1, 2018 in Washington, DC, US.

Inge Wessels Clowns Without Borders South Africa

Jamie Lachman University of Oxford; Lucie Cluver University of Oxford; Catherine Ward University of Cape Town; Frances Gardner University of Oxford; Mark Tomlinson Stellenbosch University

In high-income countries, parenting programs have been shown to reduce the risk of child maltreatment. Although the evidence base for these interventions within low- and middle-income countries (LMIC) is growing, there is still little programmatic evidence to address questions about the most effective delivery approaches, impacts of combined programs and add-on modules, whether different subgroups respond differently to interventions, and program scalability. These issues are all critical for real-world implementation. The large-scale dissemination of the Parenting for Lifelong Health (PLH) preventive programs for young children and adolescents provides an unprecedented opportunity to address these questions, and explore translation science from evidence to practice, by merging anonymized monitoring and evaluation (M&E) data as well as trial data from our multiple implementation sites in LMICs. The programs are currently being delivered by partner organizations in seven countries, reaching over 110 000 beneficiaries. Anonymized raw process and outcome data will be collected from these sites, cleaned, and merged within a securely stored combined dataset. Effect sizes will be compared and adjusted for across randomized and quasi-experimental studies. Hierarchical linear modelling will be used to account for nested data. Analyses of moderators, mediators, and mediated moderation will be handled in structural equation models. The study takes a collaborative approach and will work with staff from partner organizations in the analysis and publishing of findings - through this process, the study aims to develop local M&E capacity. This presentation will discuss the methods used in the collection and pooling of data, an approach that has not yet been used in the context of scaled-up preventive interventions in LMIC. It will also discuss the lessons learned to date in developing collaborations with partners and strengthening routine M&E systems in low-resource settings when partners implement on a large scale. Preliminary implementation data from the various delivery sites will be presented.