How should we measure maternal mortality in the developing world?: a comparison of household deaths and sibling history approaches

This article, published in the Bulletin of the World Health Organization, compares two approaches to measuring the maternal mortality ratio (MMR). In the "sibling history" approach, adults are asked about sisters who have survived or who have died during or shortly after pregnancy, whereas the "household deaths" approach asks similar questions about women within each household. Using data from Bangladesh, the paper finds that the two approaches produce similar estimates of pregnancy-related mortality (deaths during or shortly after a pregnancy). However, the data also suggest that about 15 per cent of these deaths were caused by factors unrelated to the pregnancy, so that these figures overestimate true maternal mortality.

Since the two approaches produce similar estimates, the paper concludes that the decision about which to use should be based on convenience rather than accuracy. However, neither approach produces precise enough estimates to monitor progress towards the Millennium Development Goal on MMRs. The authors therefore argue that in developing countries that lack complete and accurate registration of deaths, adequate monitoring will require widespread use of census questions to identify household deaths. They also call for verbal autopsies (further questions designed to ascertain cause of death) to be used to identify maternal deaths.