Predicting forensic admission among the mentally ill in a multinational setting: A Bayesian modelling approach

Authors:
Erkki J. O. Soini;Tarja Rissanen;Jari Tiihonen;Sheilagh Hodgins;Markku Eronen;Olli-Pekka Ryynänen
Affiliations:
Department of Health Policy and Management, University of Kuopio, Kuopio, Finland and Department of Social Pharmacy (Centre for Pharmaceutical Policy and Economics), University of Kuopio, Kuopio, ...;Department of Health Policy and Management, University of Kuopio, Kuopio, Finland and Niuvanniemi Hospital, Kuopio, Finland;Niuvanniemi Hospital, Kuopio, Finland and Department of Forensic Psychiatry, University of Kuopio, Finland;Department of Forensic Mental Health Science, Institute of Psychiatry, London, UK;Vanha Vaasa Hospital, Vaasa, Finland;General Practice Unit, Kuopio University Hospital, Kuopio, Finland and Department of Public Health and Clinical Nutrition, University of Kuopio, Kuopio, Finland
Venue:
Data & Knowledge Engineering
Year:
2009

Citing 4
Cited 0

Expert systems in psychiatry: A review

Journal of Medical Systems
A Simple Generalisation of the Area Under the ROC Curve for Multiple Class Classification Problems

Machine Learning
Using AUC and Accuracy in Evaluating Learning Algorithms

IEEE Transactions on Knowledge and Data Engineering
On Discriminative Bayesian Network Classifiers and Logistic Regression

Machine Learning

Quantified Score

Hi-index	0.00

Visualization

Abstract

Our objective was to explore protective and risk factors for a forensic admission among the mentally ill. The most influential risk factors included violent crime prior to the crime that leads to the index hospitalization, conviction of the biological father, and no use of psychotropic medications before the age of 18. The main contributions of this study in comparison to previous studies in this domain included the use of multinational setting, greedy Bayesian algorithm, generalized country-independent factors, a merger model with high predictive performance, various measures of predictive performance including relative predictive value, and independent learning and test sets.