Background: Maternal mortality is a significant public health problem world-wide,
as it is an important indicator for the functioning of the health system. The maternal
mortality ratio for Botswana is higher than other countries with comparable
economic growth, despite impressive access to health services. In order to develop
relevant programs and policies to reduce maternal mortality, the factors associated
with maternal mortality were studied. The study aimed to describe the maternal and
health services factors associated with maternal mortality in South East Botswana.
Methodology: A quantitative case-control study was used to retrospectively review
medical records for 71 cases of maternal deaths and 284 controls randomly selected
from mothers who delivered in the same year and at the same health facility, in
South East Botswana from 2007 to 2009. Information was collected on the maternal
and health services characteristics of the cases and controls including age, level of
education, marital status, parity, utilization of health facilities that consist of
antenatal care (ANC), type of delivery, complications during pregnancy, type of
health facility and ANC provider. Data was analyzed using Predictive Analysis
Software (PASW) Version 18. Two-sample t- test, Pearson’s Chi-square test and the
Fisher’s exact test were used to test the difference between the proportions of the
various categories of variables in cases and controls. Univariate logistic regression
analysis was applied to identify the risk factors associated with maternal deaths. A
multivariate logistic regression model was estimated to see the joint effects of the
identified risk factors for maternal mortality. Hosmer and Lemeshow test was used to
test the goodness of fit of the model.
Results: The mean age of the maternal deaths was 28.0 ± 5.3 years and they had
taken place at a hospital (100%). A large number of deaths occurred before delivery
xi
(59.0%). The causes of maternal death included both direct (73%) and indirect
causes (27%). Direct causes were the leading causes of death and they were abortion
(22.5%) and haemorrhage (18.3%). The maternal characteristics associated with
maternal mortality were having complications at delivery (OR=20.91), not receiving
ANC (OR=6.31) and delivering by caesarean section (OR= 2.66). The health facility
characteristics associated with maternal mortality were delivering outside the health
facility (OR=14.78), having been referred from another facility (OR=8.62) and
delivering at a general hospital (OR=5.91). The data produced a model with good fit
that included one maternal risk factor and three health facility risk factors. These
were being admitted with preterm labour, delivering at a general hospital or before
arrival at the health facility and having been referred from another health facility.
Conclusion: Maternal mortality was associated with both maternal and health
facility risk factors. The model developed may be used to identify and manage highrisk
women to reduce the number of maternal deaths. It was recommended that, the
current system should continue to be monitored and evaluated through the Maternal
Mortality Monitoring System (MMMS). Furthermore, the referral and management
of complications needs to be strengthened through a multi-sectoral approach.