Transparent two stage approach to identification and selection of qualitative literature for meta ethnographic synthesis

Article type
Authors
Taylor S1, Lewin S2
1Centre for Health Sciences, Institute of Health Science Education, Queen Mary University of London, Barts & The London School of Medicine and Dentistry, London, UK
2Norwegian Knowledge Centre for the Health Services, Oslo, Norway
Abstract
Background: Systematic reviews of qualitative research offer an opportunity to better understand patients’ experience of living with disease and treatment. Noblitt and Hare’s meta ethnographic synthesis method has become a popular way of synthesizing qualitative literature to generate new insights. However, where there is an extensive amount of qualitative literature this method is impractical, whilst using only selected literature in a synthesis is likely to influence conclusions. A good rationale for choosing particular studies to synthesize is required. Objectives: To combine standard systematic review methodology with ‘‘mapping’’ of literature to identify papers to include in a meta ethnographic synthesis of qualitative studies describing the lived experience of patients with chronic obstructive pulmonary disease (COPD). Methods: We searched electronic databases and identified studies for potential inclusion. Study setting, aims, theoretical underpinning, sampling approach, participant characteristics, methods, key findings and study quality were abstracted by two independent reviewers. Reviewers made a subjective assessment of the ‘‘richness’’ and novelty of the data. We generated a diagram representing the typical trajectory and disease related life events of a patient with COPD and added the studies we identified. We attempted to express the amount of data in the individual papers visually on the diagram. We used this process to identify studies to include in the meta ethnographic synthesis. Results: We identified 48 studies including PhD theses, published up to February 2007. Mapping reveled that most studies involved patients with very advanced disease and there were few studies around the time of diagnosis or when symptoms were mild. This two stage review process allowed us to identify studies to include in our more detailed meta-ethnographic synthesis and made the rationale for study selection more transparent. Conclusions: This two stage approach may prove useful to other researchers interested in the synthesis of voluminous qualitative literature.