Modelling Patient Flow in a Radiotherapy Department

This paper reports the development of a simulation model of the radiotherapy department of \Valsgrave Hospital and shows the results of experiments conducted using the model. In the radiotherapy department a key target is the number or days from the time that the consultant refers the patient to the department to the elate that the patient receives their first treatment. Marry factors can affect this time and this project has attempted to idcntilv and quantilv the effects or these factors. Initial discussions were held to provide Cl framework for the study These discussions were followed by a series or visits to the department to observe the way ill which it operates and to interview staff in each of' the main areas of work. Follmving this a flow diagram of the department was produced. This was used as a basis for the development of a computer simulation model which represents the way in which a patient moves through the department, Irom referral by a consultant, to the planning of the treatment, to the treatment itself, and finally discharge from the hospital. A series of experiments was used to evaluate the effects of changes to the input to the model, the number of resources in the model and the way in which the department operates. Finally, recommendations are made for future work to develop the model further and suggestions arc made for its application. Introduction The simulation models were developed in response to a request from the Walsgravc Hospital (Coventry) to help understand the effects of changes in the demand for its services and the resources available at its radiotherapy department. Onc source of concern that "vas identified by the Radiotherapy Department was the length of time from the original referral to the department to the time that the first appointment is received. Generally the date for the start of the treatment is not given to the patient until they have their session on a simulator machine. This means that there is a period of' uncertainty after the appointment with the consultant when they do not know how long they will have to wait to begin treatment. From canvassing experts in the area, '1'ay101'and Robinson (200G) note that health care is a priority application area for simulation, and that models in this area arc very complex, yet need to be developed very quickly. Eldahi et 211(20(H), however, make the point that communication is a major issue with clinicians, and persuading them that simulation modelling may be of benefit is not an easy task. In order to address this, the work was undertaken in a participatory way. The process of building the simulation model can be of benefit in itself as it can often lead to a greater understanding of the system that is being modelled. This may be a valuable outcome, even if it is not possible to build an accurate model and the discussion and identification of the problem are important steps in themselves (Robinson 200 I). Edwards et al (2005) note that wider consideration and the willingness of participants to share their knowledge is important and it is equally applicable to this case. A number of papers have described the use of Cl soft approach La