There is great demand to show that progress is being made as a result of Safe Motherhood programmes to reduce maternal mortality. However, while maternal mortality indicators are intuitive and appealing, they are costly to measure precisely. Measures of maternal morbidity are also problematic. This paper examines the scope for using health outcome indicators for measurement, and the use of both experimental and descriptive study designs. Women’s perceptions of obstetric morbidity can be measured using population-based interviews, but they agree poorly with biomedical diagnoses. Underlying (chronic) maternal morbidity, but not direct obstetric morbidity, can be measured via relatively expensive health-examination surveys. Facility-based information on maternal morbidity is a promising alternative to population-based interviews. The impracticality of using health outcome indicators leaves process indicators as the alternative in most settings, possibly supplemented by facility-based data on health outcomes. Considerably more work on process indicators, and documentation of that work, is needed. Study design is a key factor in determining whether change can be attributed to a Safe Motherhood programme intervention. Experimental approaches are definitive, but difficult and costly; hence, descriptive approaches are more likely to be used. Descriptive designs require the use of many process indicators, not just a few global ones, to build a convincing case that change is due to programme interventions. Better record-keeping and more critical use of data are recommended.
Anon. Measuring Progress in Safe Motherhood Programmes: Uses and Limitations of Health Outcome Indicators. In: Berer, M.; Sundari Ravindran, T.K. (Eds), Safe Motherhood Initiatives: Critical Issues. Blackwell Science Limited for Reproductive Health Matters, (1999) 31-42. ISBN 0 9531210 1 1