A Random Effects Sensitivity Analysis for Patient Pathways Model

Abstract

In this paper, we present a random effects approach to modelling of patient pathways with an application to the neonatal unit of a large metropolitan hospital. This approach could be used to identify pathways such as those resulting in high probabilities of death/survival, and to estimate cost of care or length of stay. Patient-specific discharge probabilities could also be predicted as a function of the random effect. We also investigate the sensitivity of our modelling results to random effects distribution assumptions.