Mixed-methods synthesis for community-level public health interventions: Some considerations from a review of skin cancer prevention

Article type
Authors
Lorenc T1, Jamal F2, O’Mara A3, Cooper C4
1London School of Hygiene & Tropical Medicine, UK
2University of East London, UK
3EPPI-Centre, Institute of Education, UK
4Peninsula Technology Assessment Group (PenTAG), Peninsula College of Medicine & Dentistry, University of Exeter, UK
Abstract
Background: The potential value of using systematic reviews of qualitative research and mixed-methods synthesis in conjunction with reviews of the effectiveness and cost-effectiveness of interventions is generally recognised. However, few studies have directly compared different synthesis methodologies.

Objectives: To explore methods for mixed-methods synthesis in the context of community-level interventions, and identify potential challenges and directions for future research.

Methods: A systematic review of the effectiveness of community-level skin cancer prevention interventions was conducted by CRD1. Our systematic review of qualitative evidence, funded by NICE, was conducted in parallel and aimed to identify barriers and facilitators to such interventions. We used thematic synthesis to identify themes in the qualitative evidence. After the completion of the NICE review, we carried out further analysis comparing two methods for cross-review synthesis: constructing a matrix of themes from the qualitative research and identifying themes addressed (or not addressed) in the intervention research; and constructing causal pathways for selected interventions and using these to identify relevant qualitative themes.

Results: Both models displayed the potential for qualitative synthesis to enhance the evidence base. However, analysis using a cross-study matrix for community-level interventions encountered some limitations, due to the lack of explicit cognitive content in the interventions, and to the limited extent of the effectiveness data. Causal pathway analysis provided a richer account of the relationships between the two forms of data, but does not enable rigorous testing of hypotheses, and requires potentially questionable interpretation of the effectiveness findings.

Conclusions: Different approaches to mixed-methods synthesis have different strengths; there may be a trade-off between rigour and informativeness. The findings suggest broader questions about the role and purpose of mixed-methods synthesis.

Reference
1. McDaid C et al. (2010). Sun protection resources and environmental changes to prevent skin cancer: A systematic review. York: Centre for Reviews and Dissemination.www.nice.org.uk/nicemedia/live/11871/ 49660/49660.pdf