Forecasting hospital bed requirements and cost of care using phase type survival trees

In our previous work we proposed a phase type survival tree method for clustering patients into clinically meaningful patient groups. We also discussed its extension which uses the survival tree based analysis for patient pathway prognostication and to examine the relationship between length of stay in hospital and destination on discharge. The current paper illustrates how phase type survival tree based clustering can be used for better forecasting of bed requirements and cost in a stroke unit by analyzing the effects of individual clusters and their interactions with the whole patient care unit. The paper then describes how this approach can be used for better estimation of patient demography and cost of care at a given time in the future. An application of the approach is illustrated using 5 years' retrospective data of patients admitted to the Belfast City Hospital with a diagnosis of stroke.