T. Kulmala et al., The relationship between antenatal risk characteristics, place of deliveryand adverse delivery outcome in rural Malawi, ACT OBST SC, 79(11), 2000, pp. 984-990
Objective. To describe and compare the frequency of antenatally identified
maternal 'risk' characteristics, place of delivery and occurrence of delive
ry complications.
Methods. A prospective cohort study of 780 pregnant women completing antena
tal followup at a rural health center in Malawi.
Results. Three-quarters of the subjects had at least one commonly accepted
risk characteristic. Only 30% of these women, and 22% of those with no risk
characteristics, delivered in a modern health facility. Four women died, 1
27 experienced other delivery complications and there were 52 perinatal dea
ths. The 'at-risk' classification had over 80% sensitivity but less than 30
% specificity to predict delivery complications or perinatal deaths. The po
sitive predictive values were as low as 20% for delivery complications and
7% for perinatal mortality. Most individual 'risk' characteristics were not
associated with adverse delivery outcomes, even when adjusted for the plac
e of delivery.
Conclusions. Antenatal risk identification failed to promote safe deliverie
s because of a poor predictive value of the 'risk' variables and the failur
e of the identified 'at-risk' individuals to deliver in modern health facil
ities.