Group based approaches to supporting breastfeeding in primary care

This thesis describes two primary care intervention studies which aimed to improve breastfeeding initiation, duration and satisfaction in Scotland. I review the evidence for breastfeeding interventions in group settings and propose a framework for their design and evaluation. A clinical overview of breastfeeding was commissioned by The British Medical Journal. Three papers describe a preliminary action research controlled intervention study which significantly increased breastfeeding at 2 weeks after birth. A qualitative paper provides insights into why women preferred groups to one-to-one peer support. The intervention was not uniformly effective in the four intervention areas and mixed method data suggests that the quality of inter- and intra- health professional team relationships may have contributed to effectiveness. This preliminary study informed the design of a cluster randomised controlled trial of a policy to provide breastfeeding groups for pregnant and breastfeeding mothers in primary care. A paper describes the processes, barriers and facilitators to recruiting primary care organisations to participate and recommends using a qualitative ethnographic approach. The breastfeeding group intervention had no effect and a paper describes the outcomes, the amount of intervention delivered and the costs. A model of health service attributes necessary for a locality to successfully deliver the policy emerged from the prospectively analysed mixed method case studies, and appears to explain breastfeeding outcomes. A policy to provide breastfeeding groups can be effective if the health service context is favourable as it was in the preliminary study. However in the trial, the negative aspects of the current health service system outweighed any positive effects from the intervention. I propose embedding randomised controlled trials of complex interventions within qualitative research to evaluate context.