Comparing models for quantitative risk assessment: an application to the European Registry of foreign body injuries in children

Abstract

Risk Assessment is the systematic study of decisions subject to uncertain consequences. An increasing interest has been focused on modeling techniques like Bayesian Networks since their capability of (1) combining in the probabilistic framework different type of evidence including both expert judgments and objective data; (2) overturning previous beliefs in the light of the new information being received and (3) making predictions even with incomplete data. In this work, we proposed a comparison among Bayesian Networks and other classical Quantitative Risk Assessment techniques such as Neural Networks, Classification Trees, Random Forests and Logistic Regression models. Hybrid approaches, combining both Classification Trees and Bayesian Networks, were also considered. Among Bayesian Networks, a clear distinction between purely data-driven approach and combination of expert knowledge with objective data is made. The aim of this paper consists in evaluating among this models which best can be applied, in the framework of Quantitative Risk Assessment, to assess the safety of children who are exposed to the risk of inhalation/insertion/aspiration of consumer products. The issue of preventing injuries in children is of paramount importance, in particular where product design is involved: quantifying the risk associated to product characteristics can be of great usefulness in addressing the product safety design regulation. Data of the European Registry of Foreign Bodies Injuries formed the starting evidence for risk assessment. Results showed that Bayesian Networks appeared to have both the ease of interpretability and accuracy in making prediction, even if simpler models like logistic regression still performed well.

Keywords

Bayesian Network children classification trees foreign body injuries quantitative risk assessment

Get full access to this article

View all access options for this article.

References

The Royal Society. Risk: analysis, perception and management: Report of a Royal Society Study Group. London, 1992.

Cox

. Risk analysis foundations, models and methods. ed. I.S.i.O.R.M. Science. Boston: Kluwer Academic Publisher, 2002.

Aven

Kristensen

. Perspectives on risk: review and discussion of the basis for establishing a unified and holistic approach. Reliability Eng Syst Saf 2005; 90(1): 1–14.

Vose

. The application of quantitative risk assessment to microbial food safety. J Food Protect 1998; 61(5): 640–648.

Wreathall

Nemeth

. Assessing risk: the role of probabilistic risk assessment (PRA) in patient safety improvement. Qual Saf Health Care 2004; 13(3): 206–212.

Roy

Bhatt

Rajagopal

. Quantitative risk assessment for accidental release of titanium tetrachloride in a titanium sponge production plant. J Hazard Mater 2003; 102(2-3): 167–186.

Clark

Setzer

Barton

. Framework for evaluation of physiologically-based pharmacokinetic models for use in safety or risk assessment. Risk analysis 2004; 24(6): 1697–1717.

Floerl

Inglis

Hayden

. A risk-based predictive tool to prevent accidental introductions of nonindigenous marine species. Environ Manage 2005; 35(6): 765–778.

Levy

Wolff

Evans

. A regression-based approach for estimating primary and secondary particulate matter intake fractions. Risk Analysis 2002; 22: 895–904.

10.

Fenton

Krause

Neil

. Software Measurement: Uncertainty and Causal Modelling. IEEE Software 2002; 10(4): 116–122.

11.

Heckerman

Mamdani

Wellman

. Real-world applications of Bayesian networks. Comm ACM 1995; 38(3): 25–26.

12.

Gregori

. Foreign body causing and prolonging hospitalization in children 0-14 age: early results from the ESFBI Study. In: Healthcare System and Patient Safety, Florence, Italy: Taylor & Francis, 2005.

13.

Rider

. Quantitative risk analysis. Injury Contr Saf Promotion 2000; 7(2): 115–133.

14.

Rimell

. Characteristics of objects that cause choking in children. J Am Med Assoc 1996; 274(22): 1763–1766.

15.

Stool

Rider

Welling

. Human factors project: development of computer models of anatomy as an aid to risk management. Int J Pediatr Otorhinolaryngol 1998; 43(3): 217–227.

16.

Gregori

. The Susy Safe Project A web-based registry of foreign bodies injuries in children. Int J Pediatr Otorhinolaryngol 2006; 70(9): 1663–1664.

17.

DTI. Choking risk to children under four from toys and other objects, London: DTI, 1999.

18.

Kohavi R. A study of cross-validation and bootstrap for accuracy estimation and model selection. Proceedings of Fourteenth International Joint Conference on Artificial Intelligence (IJCAI), 1995. Montreal, Quebec, Canada: Morgan Kaufman Publishing, pp. 1137–1143.

19.

Harrell

JFE

. Regression modeling strategies, New York: Springer, 2001.

20.

R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria, 2012.

21.

Cover

Thomas

. Elements of information theory 2006; xxiii, 2nd edn, Hoboken, NJ: Wiley-Interscience, pp. 748–748.

22.

Pearl

. Probabilistic reasoning in intelligent systems: networks of plausible inference, San Mateo, California: Morgan Kaufmann Publisher, 1991.

23.

Law

. McGraw-Hill series in industrial engineering and management science. Simulation modeling and analysis 2007; xix, 4th edn, Boston; London: McGraw-Hill, pp. 768–768.

24.

Priness

Maimon

Ben-Gal

. Evaluation of gene-expression clustering via mutual information distance measure. BMC Bioinformatics 2007; 8: 111–111.

25.

Jensen FV. Bayesian networks and decision graphs. Springer Verlag, 2001.

26.

Pearl

. Causality: models, reasoning and inference, Cambridge: CU Press, 2000.

27.

Joseph

Fenton

Neil

. Predicting football results using Bayesian nets and other machine learning techniques. Know-Based Syst 2006; 19(7): 544–553.

28.

Matheny

. In: Routh

(ed.) Accidental injuries, in Handbook of pediatric psychology, New York: Guilford, 1988.

29.

Rivara

Mueller

. The epidemiology and causes of childhood injuries. J Soc Issues 1987; 43(2): 13–21.

30.

Health

BMo

. A report on the health of British Columbians. Provincial health officer's annual report. Feature report: The health and well-being of British Columbia's children.. B.M.o. Health, Editor: Victoria, BC, 1997.

31.

Hackerman D. A tutorial on learning with Bayesian networks. In: Jordan MI (ed.) Learning Graphical models. Dordrecht, Netherlands: Kluwer, 1998, pp. 301–354.

32.

Decision Systems Laboratory – University of Pittsburgh. GeNIe 2.0, http://www.sis.pitt.edu/~genie/ (2006, assessed 3 September 2012).

33.

Norsys Software Corporation. Netica v3.18, http://www.norsys.com/netica.html (2006, assessed 3 September 2012).

34.

Ripley

. Pattern recognition and neural networks xi,Cambridge: Cambridge University Press, 1996, pp. 403–403.

35.

Breiman

. Classification and regression trees, New York: Chapman and Hall, 1984.

36.

Berchialla

. Predicting severity of foreign body injuries in children in upper airways: an approach based on regression trees. Risk Anal 2007; 27(5): 1255–1263.

37.

Venables

Ripley

. Modern applied statistics with S, 4th edn. New York: Springer, 2002.

38.

Cox

. Detecting causal nonlinear exposure-response relations in epidemiological data. Dose Response 2006; 4(2): 119–132.

39.

McCullagh P and Nelder JA. Generalized linear models. 2nd edn. Monographs on statistics and applied probability; 37. London: Chapman and Hall. xix, 1989, pp. 511.

40.

Breiman

. Random forests. Mach Learning 2001; 45(1): 5–32.

41.

Rimell

. Characteristics of objects that cause choking in children. JAMA 1995; 274(22): 1763–1766.

42.

Gregori

. Using data mining techniques in monitoring diabetes care. The simpler the better? J Med Syst 2011; 35(2): 277–281.

43.

Bartfay

Mackillop

Pater

. Comparing the predictive value of neural network models to logistic regression models on the risk of death for small-cell lung cancer patients. Eur J Cancer Care (Engl) 2006; 15(2): 115–124.