Modeling and Exploring Stillbirth Risks in Northern Pakistan

<b>Background:</b> The World Health Organization (WHO) defines stillbirth as the loss of a fetus after 28 weeks of gestation. Annually, approximately 2 million stillbirths occur worldwide. Projections indicate that by 2030, this figure could rise to nearly 15.9 million, with half of these stillbirths expected to take place in Sub-Saharan Africa. In the global literature, causes include placental complications, birth defects, and maternal health issues, though often the cause is unknown. Stillbirths have significant emotional and financial impacts on families. <b>Methods:</b> The process involves using chi-square tests to identify candidate covariates for model building. The relative risk (RR) measures the association between variables using the sample data of 1435 mothers collected retrospectively. Since these tests are independent, covariates might be interrelated. The unadjusted RR from the bivariate analysis is then refined using stepwise logistic regression, guided by the Akaike Information Criterion (AIC), to select the best subset of covariates among the candidate variables. The logistic model's regression coefficients provide the adjusted RR (aRR), indicating the strength of the association between a factor and stillbirth. <b>Results:</b> The model fit results reveal that heavy bleeding in the second or third trimester increases stillbirth risk by 4.69 times. Other factors, such as water breaking early in the third trimester (aRR = 3.22), severe back pain (aRR = 2.61), and conditions like anemia (aRR = 2.45) and malaria (aRR = 2.74), also heightened the risk. Further, mothers with a history of hypertension faced a 3.89-times-greater risk, while multifetal pregnancies increased risk by over 6 times. Conversely, proper mental and physical relaxation could reduce stillbirth risk by over 60%. Additionally, mothers aged 20 to 35 had a 40% lower risk than younger or older mothers. <b>Conclusions:</b> This research study identifies the significant predictors for forecasting stillbirth in pregnant women, and the results could help in the development of health monitoring strategies during pregnancy to reduce stillbirth risks. The research findings further support the importance of targeted interventions for high-risk groups.