Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Applying a realist lens to the evidence for breastfeeding peer support

Trickey, Heather and Sanders, Julia 2014. Applying a realist lens to the evidence for breastfeeding peer support. Presented at: The 1st International Conference on Realist Approaches to Evaluation and Synthesis: Successes, Challenges, and the Road Ahead., Liverpool University, Liverpool, UK, 27-30 October 2014.

Full text not available from this repository.

Abstract

Aim: To critically appraise the existing evidence base for breastfeeding peer support (BPS) interventions drawing on principles of realist review. To identify components of BPS intervention context and intervention mechanism which are linked to positive and negative outcomes. To set out implications of a realist approach for future BPS research and evaluation studies. Background: Breastfeeding peer support interventions are internationally recognised as having potential to contribute to improving breastfeeding continuation rates. WHO recommendations to implement BPS are supported by results of Cochrane review and findings from qualitative research studies. However, recent meta-regression of RCT studies found little evidence that BPS improves breastfeeding rates in developed countries, and no evidence to support commissioning in the UK. Methods: Drawing on recent reviews which summarise (i) RCT evidence for BPS, (ii) RCT and Quasi-RCT evidence for 'extra support', and (iii) a meta-synthesis of the evidence on women's experiences, we apply a realist lens to critically appraise the evidence base for BPS. Lay theories relating to effective and ineffective context-mechanism-outcome configurations are identified from the studies within these reviews. A growing literature on realist approaches to evaluation of complex interventions is used to identify approaches to evaluation research that might be likely to produce results that are meaningful to commissioners. Results/ Findings: As a basis for commissioning decisions, existing outcome data from RCTs are of limited value. Interpretation difficulties include issues of study and intervention design, implementation problems, heterogeneity, under-theorising of mechanisms, and difficulties caused by complex and emergent relationships between intervention and context. Lay theories relating to effectiveness included 'supportive relationships', 'proactive and accessible interventions', 'timely intervention', and 'supportive infrastructure'. Lay theories relating to ineffectiveness included 'difficulties in the peer supporter/ professional interface', 'other support available', 'the women who are targeted did not want support', and 'targeted mothers not motivated to breastfeed'. Case study approaches are likely to be helpful in extending our understanding of context-mechanism-outcome configurations within complex BPS interventions. Researchers should consider embedding future trials within a prospective realist evaluation framework. Conclusions: The message that BPS is unlikely to be effective in developing countries is premature. We contend that applying realist approaches to the evidence for BPS will help policy makers to understand which kinds of interventions are worth commissioning and testing, in which circumstances, where, and for whom. In view of the complexity within the existing evidence base, we argue that a theory-driven focus is needed to move our understanding forward. Continuing to commission traditional RCTs without a strong theoretical underpinning is unlikely to produce useful information.

Item Type: Conference or Workshop Item (Paper)
Date Type: Completion
Status: Unpublished
Schools: Development and Evaluation of Complex Interventions for Public Health Improvement (DECIPHer)
Healthcare Sciences
Social Sciences (Includes Criminology and Education)
Subjects: H Social Sciences > H Social Sciences (General)
R Medicine > RA Public aspects of medicine
Last Modified: 31 Oct 2017 11:07
URI: http://orca.cf.ac.uk/id/eprint/83201

Actions (repository staff only)

Edit Item Edit Item