Modelling patient flow and outcomes in community healthcare − a fluid approximation of a stochastic queueing system

An ambition of UK healthcare policy in recent decades has been to deliver more care in the community. However, questions remain over the impact of shifting services from acute settings closer to patient homes. This is complicated by a lack of comparable measures, nationally and locally, for evaluating quality across differing community services. In this project we develop a novel patient flow model to aid the evaluation of community services which incorporates patient outcomes. The model includes patient flow dynamics common to community care such as the use of multiple services and possible re-use of services. We represent outcomes as groups which patients may move between during a course of care, which are used to model differentiated service. The model provides insight into the performance of an interrelated healthcare services. We extend a first order fluid approximation of a stochastic queueing system with service reuse to include multiple patient groups. In considering differentiated service, we implement a novel method for dynamically allocating servers across parallel queues, overcoming problems of server inactivity. Furthermore, we develop a new measure of “the flow of outcomes” to evaluate how individual services contribute to the output of outcomes from a system of care over time