Prediction Markets as a Medical Forecasting Tool: Demand for Hospital Services
By Rajakovich, David; Vladimirov, Vladimir
Volume 3, Number 2, August 2009 , pp. 78-106(29)Background
This paper presents the outcome of a study conducted at the Royal Devon and Exeter Hospital in which a prediction market was established in order to forecast demand for services. To the researcher's knowledge, it does not appear that prediction markets have been previously utilized in a healthcare environment.
The purpose of this study is to provide evidence for the effective use of prediction markets in a healthcare environment.
Methodology and Approach
The study was conducted over a period of one week, and involved sixty-five participants. Each was asked to provide an estimate for demand for services at the Royal Devon and Exeter Hospital. Characteristics gathered for each participant included level of education, occupation, directorate, number of years worked for the hospital, and number of years worked for the National Health Service.
The study confirms the effectiveness of prediction markets to forecast future events as overall hospital demand was forecasted with an error of only 0.3%. The prediction market was less successful in predicting demand for services for each department, which the researcher attributes to the small sample size and lack of diversity of participants. Additionally, only a very small percentage of the characteristics captured registered a statistically significant correlation with the accuracy of the estimate. Further studies should focus on different characteristics and/or use a larger sample size to either confirm or refute the existence of such characteristics.
The findings of this work could potentially be used as an innovative way to augment the forecasting function for a wide range of healthcare facilities. With the preliminary success of this study to forecast demand, further research in the field is warranted.
Thursday, August 20, 2009
Prediction Markets in Demand For Hospital Services
Interesting new study in the Journal of Prediction Markets: