Risk-adjusted multi-stage bayesian performance monitoring model for maternal healthcare

Abstract

Most statistical process control programs in healthcare focus on surveillance of outcomes at the final stage of a procedure, such as mortality or failure rates. Such an approach ignores the multi-stage nature of these procedures, in which a patient progresses through several stages prior to the final stage. In this study, a Bayesian network and a multivariate binary logistic regression predictive model have been formulated considering different aspects of antepartum period and some new outcome variables and risk factors. The model formulation is based on the combination of an extensive study of previous researches, expert opinions and empirical evidences. Based on the model, data have been simulated for monitoring by the multi-stage exponentially weighted moving average control charts. The formulated models and control charts demonstrate that different variables of antepartum period and other new variables incorporated in this study are crucial in evaluating the risk of the pregnant mothers and infants. The predictive model with control charts not only benefits the patients, but also gives the healthcare management a vital competitive edge by enhancing efficiency and accuracy of performance and better utilization of different resources.