A Strategy for the Evaluation of Activities To Reduce Maternal Mortality in Developing Countries

Abstract
The expense and effort involved in collecting the data necessary for calculating maternal mor tality rates and ratios often discourages efforts to evaluate projects. The use of a set of care fully selected process indicators is presented as a practical alternative. The evaluation strategy proposed is based upon a recently published conceptual framework of maternal mortality. The strategy includes four stages: (1) outlining the causal pathway through which an intervention is hypothesized to work; (2) determining how the pathway will be affected by external factors; (3) developing indicators at each step of the pathway; and (4) selecting a combination of indicators that together will provide sufficient evidence that the intervention was successful. This process is illustrated for three program interventions: providing safe abortion services, increasing knowledge of obstetric complications, and improving quality of medical care.