Towards a taxonomy of logic models in systematic reviews and health technology assessments: A priori, staged, and iterative approaches

The complexity associated with how interventions result-or fail to result-in outcomes and how context matters is increasingly recognised. Logic models provide an important tool for handling complexity, with contrasting uses in programme evaluation and evidence synthesis. To reconcile these, we developed an approach that combines the strengths of both traditions, propose a taxonomy of logic models, and provide guidance on how to choose between approaches and types of logic models in systematic reviews and health technology assessments (HTA). The taxonomy distinguishes 3 approaches (a priori, staged, and iterative) and 2 types (systems-based and process-orientated) of logic models. An a priori logic model is specified at the start of the systematic review/HTA and remains unchanged. With a staged logic model, the reviewer prespecifies several points, at which major data inputs require a subsequent version. An iterative logic model is continuously modified throughout the systematic review/HTA process. System-based logic models describe the system, in which the interaction between participants, intervention, and context takes place; process-orientated models display the causal pathways leading from the intervention to multiple outcomes. The proposed taxonomy of logic models offers an improved understanding of the advantages and limitations of logic models across the spectrum from a priori to fully iterative approaches. Choice of logic model should be informed by scope of evidence synthesis, presence/absence of clearly defined population, intervention, comparison, outcome (PICO) elements, and feasibility considerations. Applications across distinct interventions and methodological approaches will deliver good practice case studies and offer further insights on the choice and implementation of logic modelling approaches.