Community-based surveillance of maternal, infant, and child (under-5) mortality in the health district of Tanguiéta (Benin) from 2006 through 2010.

Abstract

In Benin, maternal, infant (under 1 year) and child (under 5 years) mortality rates are measured every five years by national demographic health surveys. These provide data only at the national level, however. Because they do not reveal local disparities, they provide no information about how to target healthcare interventions. To overcome these limitations and evaluate the primary healthcare program in the Tanguieta district, the UniCredit Foundation set up a system for monitoring maternal and under-5 deaths in partnership with the local authorities. The system costs € 10,500/year, and is based on a network of 155 community healthcare workers (sentinels) responsible for counting all maternal and childhood deaths. From 2006 through 2010, the maternal mortality rate (MMR) decreased from 531 to 220 deaths per 100,000 live births; 47% of these deaths occurred at home. Mortality among children younger than 5 years (U-5) was 76 per 1,000 live births in 2006 and 77 per 1,000 live births in 2010), and 46% of the children died at home. A quality control study of our system in February 2011 showed that its sensitivity was good (95%) for MMR (232 deaths per 100,000 livebirths in 2010), but poor (48%) for U-5 mortality (155 deaths per 1,000 livebirths in 2010) and U-1 mortality (74 per 1,000 livebirths, sensitivity 47%). We conclude that the system is adequate for assessing the effect of maternal healthcare interventions but has some weaknesses in relation to childhood mortality, mainly because of local social customs and the fact that childhood deaths appear to have less resonance. Poor coverage of isolated villages may also explain this partially unsatisfactory performance. Our intervention has nonetheless contributed to improving the quality of basic data collection in the district.