Date of Award

2002

Degree Type

Thesis

Degree Name

Bachelor of Science Honours

Faculty

Faculty of Communications, Health and Science

First Advisor

James Cross

Abstract

The health care industry b a very expensive one, constituting a significant proportion of the government budgets. Princess Margaret Hospital (PMH) for children is the only tertiary paediatric centre in Western Australia. PMH has over 40,000 children aged 0-16 years of age who present to the emergency department each year. PMH is one of many hospitals funded from government sources. The emergency department is a high cost area and an area with limited ability to curtail services due to financial constraints. A busy hospital will over a period of time have a constantly changing number of people presenting to the emergency department for treatment. This may depend upon the time of year, the day of the week, the weather of the day or the presence of a holiday period. An ability to accurately predict the daily, weekly patient flow, together with seasonal fluctuations, could enable more informed decisions to be made regarding support services needed. This would in tum result in cost savings and improved medical care. The study used Time Series Analysis techniques, Regression Analysis and other statistical techniques to model the presentation rates to the PMH emergency department. The analysis also examined the factors that have the greater influences upon these rates. The resulting prediction models can be used for quantitative short-term forecasts of future ED volumes. An overall comparison was made between the differing techniques used in the analysis. Two cycles were clearly identified within the ED data series used. The dynamic methods proved to be the better at modelling and providing forecast values for the series analysed. It is hoped these types of models may provide formal criteria to assist staff in earlier recognition of extreme periods.

Share

 
COinS