Abstract
Background: Health interventions often involve complex, multifaceted components and dynamic, nonlinear effects that challenge traditional experimental methods like randomized controlled trials (RCTs). Contribution Analysis (CA) offers a theory-based impact evaluation approach that shifts the focus from solely attributing outcomes to interventions to understanding their contributions within specific contexts. Despite its widespread use in fields such as international development and public administration, the application of CA in health interventions remains underexplored. To address this gap, we conducted a scoping review to examine its application in evaluating health-related interventions, including the types of interventions assessed, the methodological approaches, and the strengths and challenges encountered.
Approach: Using established scoping review methods, we searched 15 databases—including MEDLINE, Embase, and Web of Science—for peer-reviewed studies published from 1999 to 2023. We included studies that addressed the application of CA in health-related interventions with a clear link to health behaviours or outcomes. Data extraction focused on study characteristics, implementation details of CA, and reported strengths and challenges. We used descriptive analysis to summarize study characteristics and applied content analysis to synthesize data on CA implementation and its perceived effectiveness across studies. Our study is reported using the PRISMA-ScR guidelines to ensure transparency and rigour.
Results: Our search yielded 1,875 citations. The seven studies remaining for synthesis were conducted in diverse health contexts, such as health promotion programs, policies, and interventions focused on nutrition, cardiovascular disease, substance misuse, and suicide prevention. We found that CA was primarily implemented using Mayne’s six-step method, often adapted to suit the health context. Methods for developing the Theory of Change, a core component of CA, included participatory workshops, document reviews, and stakeholder consultations. We found that engaging stakeholders like program staff, policymakers, and clinicians was crucial for ensuring the evaluation's relevance and validity. Strengths of CA include the flexibility to adapt to complex situations, the potential to inform decision-making and intervention refinement, and capacity to enhance our understanding of programs and policies. CA also fosters ownership and appreciation for the evaluation process by actively involving stakeholders. However, challenges exist, including determining appropriate levels of evidence to support contribution claims and managing the resource-intensive nature of CA, such as accessing quality data, particularly in retrospective evaluations, and coordinating multiple rounds of data collection.
Implications: Our scoping review highlights the growing use of CA in assessing complex health interventions. It emphasizes CA's potential to offer detailed insights into the mechanisms behind how and why these interventions achieve their outcomes. Its flexibility makes it especially suitable for evaluating integrated care initiatives. However, the variation in study reporting suggests that CA could benefit from more standardized methods and robust stakeholder engagement strategies. To enhance its rigour and credibility, future research should focus on developing reporting guidelines, standardized methods, and tools to support wider adoption and effective implementation of CA in evaluating complex health interventions.