Aspects of information management and resource allocation in hospitals with special reference to Accident and Emergency.

Abstract

The management and control process in an Accident and Emergency (A/E) department of a District General Hospital is investigated and the functional relationship between the A/E department and the inpatient hospital service is discussed. Attention is focused on resource allocation and methods are proposed towards reconciling levels of service and resource utilisation. Within the framework of control problems inside the A/E department, a computerised patient record system has been designed and implemented, on an experimental basis, to allow easy access to patient-related information for performance evaluation. Established statistical techniques are employed to demonstrate how such information can be utilised in medium-term management activities in the A/E department and to provide a sound basis for defining areas where specific problems arise. A method is developed, which uses patient data to the extent that they are routinely available through the patient record system, for allocating physicians to weekly shifts in a way which takes account of the fixed number of physician hours per week; of physician preferences with regard to shifts; and of the patient assessment of the service provided. With regard to the role of the A/E department as an essential link between the community at large and the hospital service, a simulation model is developed for determining the number of beds in hospital inpatient departments on the basis of expected demand and according to a pre-specified set of measures of hospital efficiency. The measures used are the rapid admission of emergency patients; high occupancy rates; and short lengths of waiting lists. A further study on bed capacity planning concerns the contemplated development of an observation ward in the A/E department. Owing to the increased uncertainty in planning for prospective units, approximation is accepted for the sake of procedural simplicity and an analytic infinite server queueing model is employed to evaluate various numbers of beds for the unit interms of the average occupancy rates and of hourly and daily service levels.