Abstract
BACKGROUND: A persistently high maternal mortality ratio (MMR) from direct pregnancy causes of maternal deaths in a district of the Gauteng Province prompted a detailed audit to identify clinical factors associated with severe maternal outcomes (SMOs). METHODS: A cross-sectional matched case control study design was used to identify clinical factors associated with severe maternal outcomes in two regional hospitals in an urban district in Gauteng. 175 sequential SMO cases (near misses and maternal deaths) were matched to two different controls each. The first control was of women with complications while the second control was of women without complications. Data on individual and clinical obstetric characteristics of cases and controls was extracted from clinical case files. A Firth penalised conditional logistic regression was used to identify the clinical factors associated with severe maternal outcomes. RESULTS: The overall SMO incidence ratio and MMR were 34.7 per 1000 live births and 130.8 per 100 000 live births respectively. Haemorrhage (48.8%) and hypertension (46.9%) were the most common underlying causes of SMOs. The SMO incidence ratio was 39.0 and 30.2 per 1000 live births for Hospital 1 and 2 respectively. Not receiving antenatal care was the most significant risk for SMOs (p < 0.001). In women with complications, not initiating ANC attendance (AOR = 11.0; CI = 2.6-46.0) or having less than 4 ANC visits (AOR = 2.2; CI = 1.2-4.2) had the highest risk for SMOs (p < 0.001). In women without complications, anaemia (AORs = 3.0; CI = 1.4-6.6; p < 0.01), and hypertension (AORs = 71; CI = 4.2-1,196.2), significantly increased the odds of SMOs. CONCLUSIONS: The severe maternal incidence was high in both hospitals. Obstetric haemorrhage and hypertension were significantly associated with most SMOs. Not attending ANC and having few ANC visits were important factors that showed significant association with poor maternal outcomes.