Developing a Bayesian belief network for the management of geriatric hospital care

Health Care Management Science - 2001
A.H. Marshall1, S.I. McClean1, C.M. Shapcott1, I.R. Hastie2, P.H. Millard2
1School of Information and Software Engineering, Faculty of Informatics, University of Ulster, Northern Ireland
2Department of Geriatric Medicine, St. George's Hospital, London, UK

Tóm tắt

Resource management is an essential feature of hospital management. This is especially true for geriatric services, as older people often have complex medical and social needs. Hospital management should benefit from an explanatory model that provides predictions of duration of stay and destination on discharge. We describe how a Bayesian belief network models the behaviour of geriatric patients using predictive variables: personal details, admission reasons and dependency levels. This approach is illustrated using data on 4722 patients admitted to geriatric medicine at St. George's Hospital, London; distributions of the patient outcome given typical values of the predictive variables are provided.

Từ khóa


Tài liệu tham khảo

F.V. Jensen, An Introduction to Bayesian Networks (UCL Press Ltd., 1996).

E.J. Dunstan, K. Amar, A. Watt and D.G. Seymour, First steps in building ACME – an admission case-mix system for the elderly, Age & Ageing 25 (1996) 102–108.

B. Bertozzi, P. Barbisoni, S. Franzoni, R. Rozzini, G.B. Frisoni and M. Trabucchi, Factors related to length of stay in a geriatric evaluation and rehabilitation unit, Aging (Milano) 8 (1996) 170–175.

K. Kafetz, J. O'Farrell, A. Parry, V. Wijesuriya, G. McElligott, B. Rossiter and M. Lugon, Age-related geriatric medicine: relevance of special skills of geriatric medicine to elderly people admitted to hospital as medical emergencies, Journal of the Royal Society of Medicine 88 (1995) 629–633.

P.R. Cox, Demography (Cambridge University Press, 1976).

F.I. Mahony and D.W. Barthel, Functional evaluation: the Barthel index, Maryland State Medical Journal 14 (1965) 61–65.

D. Cox and N. Wermuth, Multivariate Dependencies (Chapman and Hall, London, 1996).

W. Buntine, A guide to the literature on learning probabilistic networks from data, IEEE Transactions on Knowledge and Data Engineering 8 (1996) 195–210.

D. Heckerman, Bayesian Networks for Knowledge Discovery. Advances in Knowledge Discovery, eds. U.M. Fayyad et al. (AAAI Press/MIT Press, 1996).

S.L. Lauritzen and N. Wermuth, Graphical models for associations between variables, some of which are qualitative and some quantitative, The Annals of Statistics 17 (1989) 31–57.

J.H. Badsberg, A Guide to CoCo – An Environment for Graphical Models (Institute of Electronic Systems, Department of Mathematics and Computer Science, Aalborg University, Denmark, 1992).

P.M. Sagar, M.N. Hartley, J. MacFie, B.A. Taylor and G.P. Copeland, Comparison of individual surgeon's performance. Risk-adjusted analysis with POSSUMscoring system, Diseases of the Colon & Rectum 39 (1996) 654–658.

A.H. Marshall, S.I. McClean, C.M. Shapcott and P.H. Millard, Using Bayesian belief networks to predict the survival of stroke patients, in: Proceedings of the IX International Symposium on Applied Stochastic Models and Data Analysis (University of Lisbon, Portugal, 1999) pp. 112–117.

A.H. Marshall, S.I. McClean, C.M. Shapcott and P.H. Millard, Learning dynamic Bayesian belief networks using conditional phase-type distributions, in: Lecture Notes in Artificial Intelligence, Vol. 1910, eds. D.A. Zighed, J. Komorowski and J. Zytkow (Springer, 2000).

Thông tin
Thông tin xuất bản

Health Care Management Science

Thông tin tác giả