Incidence, mortality, and factors associated with primary postpartum haemorrhage following in-hospital births in northwest Ethiopia

Background Of the 1010 reported maternal deaths in 2018, just over 65% occurred in hospitals in Ethiopia. However, there is a lack of standardised data about the contributing factors. This study aimed to investigate the incidence, mortality, and factors associated with primary postpartum haemorrhage following in-hospital births in northwest Ethiopia. Methods A retrospective cohort design was used; an audit of 1060 maternity care logbooks of adult women post-partum at Felege Hiwot Referral Hospital and University of Gondar Comprehensive Specialized Hospital. The data were abstracted between December 2018 and May 2019 using a systematic random sampling technique. We used the Facility Based Maternal Death Abstraction Form containing sociodemographic characteristics, women’s medical history, and partographs. Primary postpartum haemorrhage was defined as the estimated blood loss recorded by the staff greater or equal to 500 ml for vaginal births and 1000 ml for caesarean section births, or the medical doctor diagnosis and recording of the woman as having primary postpartum haemorrhage. The data analysis was undertaken using Stata version 15. Variables with P 0.10 for significance were selected to run multivariable logistic analyses. Variables that had associations with primary postpartum haemorrhage were identified based on the odds ratio, with 95% confidence interval (CI) and P-value less than 0.05. Results The incidence of primary postpartum haemorrhage in the hospitals was 8.8% (95% CI: 7.2, 10.6). Of these, there were 7.4% (95% CI: 2.1, 13.3) maternal deaths. Eight predictor variables were found to be independently associated with primary postpartum haemorrhage, including age 35 years (AOR: 2.20; 95% CI: 1.08, 4.46; P = 0.03), longer than 24 hours duration of labour (AOR: 7.18; 95% CI: 2.73, 18.90; P = 0.01), vaginal or cervical lacerations (AOR: 4.95; 95% CI: 2.49, 9.86; P = 0.01), instrumental (forceps or vacuum)-assisted birth (AOR: 2.92; 95% CI: 1.25, 6.81; P = 0.01), retained placenta (AOR: 21.83; 95% CI: 6.33, 75.20; P = 0.01), antepartum haemorrhage in recent pregnancy (AOR: 6.90; 95% CI: 3.43, 13. 84; p = 0.01), women in labour referred from primary health centres (AOR: 2.48; 95% CI: 1.39, 4.42; P = 0.02), and births managed by medical interns (AOR: 2.90; 95% CI: 1.55, 5.37; P = 0.01). Conclusion We found that while the incidence of primary postpartum haemorrhage appeared to be lower than in other studies in Africa the associated maternal mortality was higher. Although most factors associated with primary postpartum haemorrhage were consistent with those identified in the literature, two additional specific factors, were found to be prevalent among women in Ethiopia; the factors were referred women in labour from primary health facilities and births managed by medical interns. Maternal healthcare providers in these hospitals require training on the management of a birthing emergency.