The impact of traditional birth attendant training on delivery complications in Ghana

Objectives: In their efforts to reduce maternal and neonatal morbidity and mortality, many national and inter-national agencies make considerable investments in training traditional birth attendants (TBAs). The value of TBA training is controversial, and plausible arguments are made both for and against. Numerous process evaluations are reported in the literature and the results are mixed, though generally positive. Outcome evaluations, however, are scarce. This article describes an outcome evaluation of TBA training conducted in two districts of Brong-Ahafo Region, Ghana, during 1996. Design and methods: Data from a random sample survey of 1961 clients of TBAs were subjected to logistic regression modelling to determine the effect of training on maternal outcomes, controlling for other independent variables. Results: Of eight outcomes modelled, three were associated with training and five were not. Three additional outcomes were not modelled, primarily due to low prevalence. Conclusions: Despite some inherent design limitations, this study found that the evidence for a beneficial impact of TBA training was not compelling. Training sponsors should consider alternative health investments and, where TBA training remains the intervention of choice, be realistic about expectations of impact.