DOI: 10.12688/f1000research.187738.1 ISSN: 2046-1402

Statistical methods, outcome structures, and research designs in hybrid type II effectiveness-implementation studies: protocol for a scoping review

Marc Parsons, Olajesutofunmi (Ola) Akinyemi, Shirin Golchi, Rachael Laritz, Rebecca Legnick-Hall, Justin D. Smith, Kathryn Hyzak, Lawrence Mbuagbaw, Guillaume Fontaine
Show PDF Cite
Background Hybrid type II effectiveness-implementation studies evaluate health intervention and implementation strategy effectiveness simultaneously. While the emphasis on each “side” varies within hybrid type II studies, the focus of this review is those that attempt a more-or-less equivalent dual emphasis. This version of type II studies spans many research designs including cluster randomized trials, sequential multiple assignment randomized trials (SMARTs), stepped wedge designs, pre-post and quasi-experimental designs, and program evaluation approaches. This bivariate structure introduces statistical challenges, including multiple testing and clustering. In addition, when outcomes are modeled as trajectories over time, dose, or implementation-strategy intensity, additional considerations arise related to effect estimation and sample-size determination. Although methodological solutions exist, the extent to which they are applied and the consistency with which authors justify them, remain unclear. This scoping review will describe the statistical methods, outcome structures, and research designs used in hybrid type II effectiveness–implementation studies. Methods This protocol follows JBI guidance (Peters et al., 2022) and will be reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) framework. We will search Embase, Medline, Scopus, and Web of Science for either: (a) hybrid type II effectiveness-implementation studies or (b) methodological guidance on their design or analysis. There will be no search limits related to publication type (e.g., protocols). Data will be extracted by two independent reviewers. The statistical methods employed will be categorised and other characteristics, such as implementation strategy used, will be extracted. The guidance of included methodological studies will be narratively summarised. Discussion This review will map the gaps between methodological availability and applications, with particular attention to designs generating trajectory outcomes. Findings will inform the planning and reporting of future hybrid type II studies and identify priorities for methodological development in this growing area of implementation science.

More from our Archive