A Multi-attribute Utility Framework for Patients to Determine Childbirth Method Considering Uncertainties, Patient Preferences, Risk Attitudes, and Pregnancy Complications

Abstract

Abstract Despite increasing advocacy of shared decision-making (SDM) approaches to maternal healthcare and the significance of including patient preferences in choosing the childbirth method, quantitative research to help the patients to make informed decisions is still lacking. This paper proposes a multi-attribute decision framework to guide the patients toward the best childbirth method considering their preferences, risk attitudes, pregnancy complications, and uncertainties during the delivery process. Valuations of health condition, intrapartum pain, duration of the process, recovery time, and feeling of empowerment constitute the attribute set of our model. Four major pregnancy complications - malpresentation, anemia, eclampsia, and gestational diabetes have been considered to assess the patients’ pregnancy complication state. A deidentified dataset from the maternal delivery unit of the Dhaka Medical College (DMC) Hospital in Bangladesh, which was screened and validated by doctors from three different hospitals, was utilized to present a case study. For mothers with no complications, vaginal delivery (VG) is preferred regardless of the risk attitude. For mothers with single difficulty, as risk attitude changes from risk-averse to risk-seeking, the optimal strategy shifts from cesarean to VG. Cesarean delivery is preferred regardless of the risk attitude for mothers with more than one complication. Sensitivity analyses reveal that valuations of health conditions and intrapartum pain are the most sensitive attributes. The decisions obtained from the model are found consistent with the decision taken at DMC 85.25% times. This paper presents new insights to foster SDM and improve the childbirth experience.