Theory of change models: explaining the link between intervention strategies and outcomes in test-treatment studies

Article type
Authors
Naidoo S1, Ochodo E1
1Centre for Evidence-Based Health Care, Stellenbosch University
Abstract
Background:
Theory of change models (conceptual/logic models) are used to explain intervention implementation and influencing factors, in relation to hypothesized change processes, and how the intervention will activate these change pathways. The use of malaria rapid diagnostic tests has shown limited impact on late stage patient health outcomes. The problem could be due to poor links between intervention and intended outcomes.

Objectives:
We aim to investigate the extent to which malaria test-treat intervention studies have made use of change models, whether these models demonstrate the link between intervention strategies and intended outcomes, and whether interventions have activated the change pathways as intended.

Methods:
We are conducting a mixed-methods systematic review. All randomised and non-randomised impact studies will be assessed by two reviewers for the presence of a theory of change model/logic model, and whether these models explain the link between intervention strategies and intended outcomes, and whether change pathways have been activated. We will use simple descriptive statistics to assess the proportion of studies that report change models. We will appraise the logic model to determine intuitively if the model explains the link between intervention and intended outcomes. We will use the Checklist for Implementation tool (Ch-IMP) to determine to what extent the intervention activates the behaviour change pathways, by assessing how many domains in the checklist have been addressed.

Results:
Amongst the 123 articles included for data extraction there are 27 randomised studies and 22 non-randomised studies. We intend to present our findings, including a graphical classification regarding the use of logic models and whether these models adequately explain the link between the intervention strategies and expected outcomes.

Conclusions:
We hope to highlight areas where test-treat interventions can be improved to effect change pathways. Well thought out change models are essential for effective strategies that are aimed at bringing about change in practice by relevant stakeholders.

Patient or healthcare consumer involvement:
Health consumers will be interviewed in the qualitative component of this research at a later stage to provide insight into their perspectives with regard to the test-treat review findings and the possible solutions to problems identified.