Objective: To compare maternal morbidity and mortality in two urban populations with contrasting availability of health care, and to test the hypothesis that differences in maternal outcome result mainly from the management of delivery in health facilities.
Design: A population-based study of a cohort of pregnant women which was part of a multicentre study of maternal morbidity in six countries of western Africa (MOMA).
Setting: Two different urban areas of Senegal (Saint-Louis and Kaolack).
Population: 3,777 pregnant women who were followed up throughout pregnancy, delivery and puerperium.
Main outcome measures: Maternal morbidity and mortality: morbidity was assessed from women's recall at each visit by the investigator and from obstetric complications diagnosed by the birth attendant within health facilities.
Results: Maternal mortality was higher in the Kaolack area where women gave birth mainly in district health care centres, usually assisted by traditional birth attendants, than in Saint-Louis where women giving birth in health facilities went principally to the regional hospital and were usually assisted by midwives (874 and 151 maternal deaths per 100,000 live births, respectively, P < 0 x 01). Maternal morbidity, however, was higher in Saint-Louis than in Kaolack area, especially for births in health facilities (9 x 50 and 4 x 84 episodes of obstetric complications per 100 live births, respectively, P < 0 x 01). Univariate and multivariate analyses showed that morbidity was mainly associated with the training of the birth attendant in facility deliveries and that antenatal care had no effect.
Conclusion: Midwives in health facilities appear to detect more obstetric complications than traditional birth attendants. Immediate detection leads to immediate care and to low fatality rates. This could explain differences in maternal outcome between two urban centres with contrasting health care availability. These results suggest that one of the strongest weapons in the fight against maternal mortality is the employment of the most qualified personnel possible for monitoring labour.