Search

Abstract Background Even though it has been shown that peer support to mothers at home helps to increase exclusive breastfeeding, little is known about the experiences of peer supporters themselves and what is required of them to fulfil their day-to-day tasks. Therefore, a community-based randomised control trial using trained "lay" women to support exclusive infant feeding at home was implemented in three different sites across South Africa. The aim of this paper is to describe the experiences of peer supporters who promote exclusive infant feeding. Methods Three focus group discussions were held, in a language of their choice, with peer supporters. These meetings focused on how the peer educators utilised their time in the process of delivering the intervention. Data from the discussions were transcribed, with both verbatim and translated transcripts being used in the analysis. Results Unlike the services provided by mainstream health care, peer supporters had to market their services. They had to negotiate entry into the mother's home and then her life. Furthermore, they had to demonstrate competence and come across as professional and trustworthy. An HIV-positive mother's fear of being stigmatised posed an added burden - subsequent disclosure of her positive status would lead to an increased workload and emotional distress. Peer supporters spent most of their time in the field and had to learn the skill of self-management. Their support-base was enhanced when supervision focused on their working conditions as well as the delivery of their tasks. Despite this, they faced other insurmountable issues, such as mothers being compelled to offer their infants mixed feeding simultaneously due to normative practices and working in the fields postpartum. Conclusion Designers of peer support interventions should consider the skills required for delivering health messages and the skills required for selling a service. Supportive supervision should be responsive both to the health care task and the challenges faced in the process of delivering it. Trial registration NCT00297150.

Statistical Analysis for Structure-Activity Relationships. Daniels, K. L., Goyert, J. C, Strand, R. H., and Farrell, M. P. (1983). 3: 350–352. The relationship between organic structure and biological activity (toxicity) is examined by statistically comparing concentration-response regression lines among structurally related compounds. Response data from acute toxicity tests are initially transformed to the logistic function while concentration data are transformed to the logarithm base 10 before being fit to a linear model using a weighted least squares analysis. A sequential approach is presented that uses statistical models for testing differences among related compounds. The approach first tests for the overall equality of regression lines (slopes and intercepts). If the regression lines are found to be different, another set of tests are conducted to determine if the slopes of the lines are equal (i.e. parallel). If the slopes are equal, multiple comparisons are made using Scheffe's procedure for determining which compounds differ in their intercepts. Relative toxicities are then estimated for various concentrations of particular compounds. [ABSTRACT FROM PUBLISHER]