Evolution in caesarean section practices in North Kivu: Impact of caregiver training

A caesarean section is a major obstetric procedure that can save the life of mother and child. Its purpose is to protect the mother's health from the complications of childbirth and to protect the baby's health. In sub-Saharan Africa (SSA), there are major inequalities in access to caesarean sections and significant variations in practices to determine the indications for the procedure. Periodic analyses of maternal deaths have shown that more than half of maternal and new born deaths are due to suboptimal care and are therefore potentially preventable. The objective of our study is to assess the impact of health staff training under the PADISS project (to support the health system's integrated development) on the quality of CS procedures in North Kivu, by comparing two periods.

The populations compared were recruited from the referral hospitals in North Kivu, DRC (Democratic Republic of Congo). The first (group 1) was made up of patient files studied retrospectively for the period from 01/11/2013 to 01/01/2016. The second group (group 2), studied prospectively, comprised patient files from June 2019 to January 2020. Obstetric, maternal and foetal data were compared. Statistical analyses were performed using STATA/IC 15.0 for Windows. Univariate and multiple logistic regression was performed to determine which characteristics are associated with maternal and perinatal morbidity and mortality. A p value < 0.05 was considered statistically significant.

CS frequency was approximately 17% in both study periods. We observed a CS frequency of about 34% at North Kivu provincial hospital for the two populations studied. The main indications for CS were dystocia, foetal distress and scarred uterus for both populations. In the population studied prospectively, after the implementation of health staff training, there were fewer incidence rate of dystocia, foetal distress and neonatal death, a more complete patient record, shorter hospital stay, and fewer blood transfusions but more incidence rate of scarred uterus, post-operative complications and low birth weight. Intervention had no statistically significant impact on low birth weight (OR = 1.9, p = 0.13), on neonatal mortality (OR = 0.69, p = 0.21).

Our study shows a decrease in neonatal deaths, dystocia and foetal distress, but an increase in post-operative complications, maternal deaths and cases of scarred uterus and low birth weight. However, multiple logistic regression did no support the conclusion.