Abstract
The length of hospital stay and its implications have a significant economic and human impact. As a consequence, the prediction of that key parameter has been subject to previous research in recent years. Most previous work has analysed length of stay in particular hospital departments within specific study groups, which has resulted in successful prediction rates, but only occasionally reporting predictive patterns. In this work we report a predictive model for length of stay (LOS) together with a study of trends and patterns that support a better understanding on how LOS varies across different hospital departments and specialties. We also analyse in which hospital departments the prediction of LOS from patient data is more insightful. After estimating predictions rates, several patterns were found; those patterns allowed, for instance, to determine how to increase prediction accuracy in women admitted to the emergency room for enteritis problems. Overall, concerning these recognised patterns, the results are up to 21.61% better than the results with baseline machine learning algorithms in terms of error rate calculation, and up to 23.83% in terms of success rate in the number of predicted which is useful to guide the decision on where to focus attention in predicting LOS.
Highlights
Extracting knowledge from databases is essential for organizations, in both private enterprises and in government agencies
If the number of samples is compared to other authors, it may seem a similar number in an analysis of this kind, but in this article, the research is performed across all hospital departments, so the number of samples is significantly lower
The same applies to those using Radial Basis Function kernels (RBF) kernel, such as the Support Vector Machines (RBF Kernel) and Support Vector Classifiers (NonLinear Kernel) algorithms
Summary
Extracting knowledge from databases is essential for organizations, in both private enterprises and in government agencies. This study departs from readily available administrative data to assess resource use in hospital systems. In [9], they describe the factors (e.g. age, type of admission, hospital type) with the most substantial influence in predicting LOS. Both the research of Freitas et al and the one reported in this article share the concern of drawing consequences to improve management resources. Researchers seem to use classification techniques predominantly, methodologies using decision trees with generic hospital data have yielded better results than others in previous studies, as it is stated in [1], a study where decision trees outstand. Other outstanding classification models are Support Vector Machine algorithms, which offer remarkable results ([10]) or KNN ([11])
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
