Predictors of Poor Maternal Outcomes among Women with Hypertensive Disorders of Pregnancy Attending Primary Health Centres in Rivers State, Nigeria

Abstract

Introduction: Hypertensive disorders of pregnancy are a leading cause of maternal morbidity and mortality globally, with disproportionately higher burdens in sub-Saharan Africa. In Nigeria, delays in recognising and managing these disorders contribute substantially to poor maternal outcomes. Evidence from Primary Health Centres in Rivers State, Nigeria, is sparse. This study aimed to determine the predictors of poor maternal outcomes among women with hypertensive disorders of pregnancy managed at Primary Health Centres in Rivers State, Nigeria. Methods: A retrospective cross-sectional analytic study was conducted across ten functional Primary Health Centres (five in Port Harcourt and five in Obio/Akpor Local Government Areas), for twelve months from January 2024 to December 2024. Case records of a clustered sample size of 360 women with hypertensive disorders of pregnancy were reviewed. Data were analysed using descriptive statistics and logistic regression. Both crude odds ratios and adjusted odds ratios were computed. Model performance was assessed with multicollinearity tests, Hosmer–Lemeshow goodness-of-fit, and the Receiver Operating Curve analysis. Results: The mean age of participants was 29.8 ± 6.4 years. The majority, 216 (60.0%), were multiparous (parity 1-3). Pre-eclampsia and eclampsia were the most frequent hypertensive disorders of pregnancy. Binary regression identified significant predictors of poor outcomes as pre-eclampsia, eclampsia, unbooked patients, delay at presentation >24 hours, delay in referral >6 hours, non-use of magnesium sulphate, and maternal age ≥35, while multivariate analysis confirmed pre- eclampsia, eclampsia, being unbooked, delayed presentation, delay in referral>6 hours, and non-administration of magnesium sulphate as independent predictors. The model demonstrated a good fit (Hosmer–Lemeshow χ² = 5.20, p = 0.74) and excellent discrimination (AUC = 0.82). Cut-off analysis showed that 0.50 provided balanced sensitivity and specificity, while 0.30 favoured sensitivity. Conclusion: Poor maternal outcomes among women with hypertensive disorders of pregnancy were strongly predicted by referral status, pre-eclampsia, eclampsia, and inadequate antenatal care. The logistic regression model demonstrated excellent discrimination and can support Primary Health Center -level decision-making.