Dept. of Medical Informatics and Clinical Epidemiology
Oregon Health & Science University
This thesis presents three decision models for a child-birth after cesarean decision. The first decision model used an AHP approach, the second used a decision tree approach, and the third model used a hybrid AHP-decision tree approach. The AHP model assessed medical risk subjectively while the decision tree and hybrid models assessed medical risk objectively. Decision criteria included both maternal and neonatal outcomes. Maternal outcomes included hysterectomy, numbness/pain near incision, incontinence, and placental abnormalities causing a risk to future pregnancies. Neonatal outcomes included disability and death. Data on 96 women with a prior cesarean were used from a partnering studying that used the AHP decision aid tool. Utilities for the decision tree were derived from a normalization method of AHP criteria weights. Various sensitivity analyses revealed the decision tree was sensitive to all probabilities of maternal outcomes except for hysterectomy. The decision models revealed that the mode of risk assessment plays a big role in determining the decision. Multiple decision models using subjective and objective risk assessments can play an important role in the clinician-patient shared decision making process.
School of Medicine
Sharma, Poonam, "A collection of multi-criteria decision analyses for a childbirth after cesarean decision using two decision methodologies : the analytic hierarchy process (AHP) and decision trees" (2007). Scholar Archive. 828.