Effectiveness of emergency obstetric care training at the regional level in Ukraine : a non-randomized controlled trial

Background Emergency obstetric care training, using Advances in Labour and Risk Management (ALARM) International Program (AIP) was implemented in Ukraine, a country with universal access to skilled perinatal and obstetric care but restricted resources. A total of 577 providers (65.5% of total) from 28 maternal clinics attended a 5-day training session focused on the five main causes of maternal mortality, with hands-on skill workshops, pre- and post- tests, and an objective structured clinical examination. The effects of this emergency obstetric care training on maternal outcomes is the subject of this paper. Methods A non-randomized controlled trial was conducted. The pilot areas where the training was implemented consisted of 64 maternity clinics of which 28 were considered as cases and 36 non-participating clinics were the referents. Data on maternal outcomes were collected for a 2-year span (2004-2005) prior to the trainings, which took place 2006-2007 and again after implementation of the trainings, from 2008 to 2009. Information was collected from 189,852 deliveries. Outcomes for the study were incidences of operative delivery and postpartum hemorrhage. Non-parametric statistics, meta-analyses, and difference in difference (DID) estimation were used to assess the effect of the AIP on maternal indices. Results DID analysis showed that after the training, compared to the referents, the cases had significant reduction of blood transfusions (OR: 0.56; 95%CI: 0.48-0.65), plasma transfusions (OR: 0.70; 95%CI: 0.63-0.78), and uterus explorations (OR: 0.64; 95%CI: 0.59-0.69). We observed a non-significant reduction of postpartum hemorrhage >= 1000 ml (OR: 0.92; 95%CI: 0.81-1.04; P = 0.103). Utilization of vacuum extraction for vaginal delivery increased (OR: 2.86; 95%CI: 1.80-4.57), as well as forceps assisted delivery (OR: 1.80; 95%CI: 1.00-3.25) and cesarean section (OR: 1.11; 95%CI: 1.06-1.17). There was no change in the occurrence of postpartum hysterectomy and maternal mortality. Conclusions After one week ...