Modeling hierarchical relationships in epidemiological studies: a Bayesian networks approach

Nguefack-Tsague, Georges and Zucchini, Walter (2011): Modeling hierarchical relationships in epidemiological studies: a Bayesian networks approach.

Abstract

Hierarchical relationships between risk factors are seldom taken into account in epidemiological studies though some authors stressed the importance of doing so, and proposed a conceptual framework in which each level of the hierarchy is modeled separately. The objective of this paper was to implement a simple version of their framework, and to propose an alternative procedure based on a Bayesian Network (BN). These approaches were illustrated in modeling the risk of diarrhea infection for 2740 children aged 0 to 59 months in Cameroon. The authors implemented a (naïve) logistic regression, a step-level logistic regression and also a BN. While the first approach is inadequate, the two others approaches both account for the hierarchical structure but to different estimates and interpretations. BN implementation showed that a child in a family in the poorest group has respectively 89%, 40% and 18% probabilities of having poor sanitation, being malnourished and having diarrhea. An advantage of the latter approach is that it enables one to determine the probability that a risk factor (and/or the outcome) is in a given state, given the states of the others. Although the BN considered here is very simple, the method can deal with more complicated models.

• Item Type: MPRA Paper
• Language: English
• Keywords: Bayesian networks; hierarchical model; diarrhea infection; disease determinants; logistic regression
• Subjects: I - Health, Education, and Welfare > I1 - Health > I12 - Health Behavior; C - Mathematical and Quantitative Methods > C4 - Econometric and Statistical Methods: Special Topics > C45 - Neural Networks and Related Topics; C - Mathematical and Quantitative Methods > C1 - Econometric and Statistical Methods and Methodology: General > C11 - Bayesian Analysis: General

References

1. Victora CG, Huttly SR, Fuchs SC, et al. The role of conceptual frameworks in epidemiological analysis: a hierarchical approach. Int J Epidemiol. 1997;26:224-27.

2. Victora CG, Fuchs SC, Flores JA, et al. Pneumonia among Brazilian children: a hierarchical analysis. Pediatr. 1994;93:977-85.

3. Fonseca W, Kirkwood BR, Victora CG, et al. Risk factors for childhood pneumonia among the urban poor in Fortaleza, Brasil: a case-control study. Bull World Health Organ.1996;74:199-208.

4. de Onis M, Monteiro C, Akre J, et al. The worldwide magnitude of protein-energy malnutrition: an overview from the WHO Global Database on Child Growth. Bull World Health Organ. 1993;71:703-12.

5. Takyi EEK. Nutritional status and nutrient intake of preschool children in Northern Ghana. East Afr Med J. 1999;76: 510-15.

6. Akaike H. Information theory and an extension of the maximum likelihood principle. In: Petrov B, Csáki F (eds). Second International Symposium on Information Theory. Budapest : Akadémiai Kiadó, 1973. 267-281.

7. R Development Core Team. R: A language and environment for statistical computing. Vienna : R Foundation for Statistical Computing, 2010.

8. Lauritzen SL, Spiegelhalter DJ. Local computations with probabilities on graphical structures and their applications to expert systems. In: Proceedings of the Royal Statistical Society 1988;50:154-227.

9. Nikovski D. Constructing bayesian networks for medical diagnosis from incomplete and partially correct statistics. IEEE Transactions on Knowledge and Data Engineering. 2000;12(4):509-16.

10. Long W. Medical diagnosis using a probabilistic causal network. Applied Artificial Intelligence. 1989;3:367-83.

11. Jensen FV, Skaanning C, Kjaerulff U. The sacso system for troubleshooting of printing systems. Technical report 2000, University of Aalborg, Denmark.

12. van der Gaag LC, Renooij S, Witteman C. Probabilities for a probabilistic network: a case study in oesophageal cancer. Artificial Intelligence Med. 2002;25(2):123-48.

13. Jensen FV. An Introduction to Bayesian Networks. New York, NY: Springer, 1996.

14. Jensen FV. Bayesian networks and decision graphs. New York, NY: Springer, 2001.

15. Jensen FV, Lauritzen SL, Olesen KG. Bayesian updating in causal probabilistic networks by local computations. Computational Statistics Quarterly. 1990;4:269-82.

16. Olesen KG, Lauritzen SL, Jensen FV. aHugin: A system creating adaptive causal probabilistic networks. In: Wellman M P, D'Ambrosio B, Smets P, Dubois D. (eds). Proceedings of the Eighth Conference on Uncertainty in Artificial Intelligence. Stanford, California, 1992. 223-29.

17. WHO Working Group. Use and interpretation of anthropometric indicators of nutritional status. Bull World Health Organ. 1986;64:929-41.

18. Charles River Analytics, Inc . About Bayesian Belief Networks. Cambridge : Charles River Analytics, Inc., 2004.

19. Kjaerulff U. Reduction of computational complexity in bayesian networks through removal of weak dependencies. In: Proceedings of the Tenth Conference on Uncertainty in Artificial Intelligence, 1994.

20. van Engelen RA. Approximating bayesian networks by arc removal. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1997;19:916-20.

21. Heckermann D, Breese JS. Causal independence for probability assessment and inference using bayesian networks. IEEE Transactions on Systems, Man and Cybernetics. 1996;26:826-31.

22. Henrion M. Some practical issues in constructing belief networks. In Uncertainty in Artificial Intelligence 3. 1989.

23. Pearl J. Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. San Mateo, CA : Morgan Kaufmann, 1988.

24. van der Gaag LC, Renooij S, Witteman CLM, et al. How to elicit many probabilities. In: Proceedings of the Fifteenth Annual Conference on Uncertainty in Artificial Intelligence, 1999.

25. Adusei-Poku K, Van den Brink G, Zucchini W. Implementing a Bayesian network for foreign exchange settlement: a case study in operational risk management. Journal of Operational Risk. 2007 ; 2 :1-6

26. Agresti, A. Categorical Data Analysis. 2nd ed. New York, NY: Wiley, 2002.

27. Agresti, A. An Introduction to Categorical Data Analysis. 2nd ed. New York, NY: Wiley, 2007.

28. MEASURE DHS, ICF Macro, Calverton (USA) http://www.measuredhs.com (accessed February 17 2009)

29. Evans AS. Causation and disease: a chronological journey. The Thomas Parran Lecture. Am J Epidemiol.1978;108:249-258.

30. Hill, AB. The environment and disease: association or causation? Proceedings Royal Society Medicine. 1965;58:295-300.

31. Lave LB, Seskin EP. Epidemiology, causality and public policy. American Scientist. 1979; 67:178-180.

32. Rothman KJ, Greenland S. Modern Epidemiology 2nd ed, Philadelphia: Lippincott-Raven. 1998.

33. Schlesselman JJ. "Proof" of cause and effect in epidemiologic studies. Preventive Medicine. 1987; 16:195-210.

34. Susser M. Judgment and causal inference. Criteria in epidemiologic studies. Am J Epidemiol. 1973; 105:1-15.

35. Susser M. What is a cause and how do we know one? Am J Epidemiol.1991; 133:635-648.

36. Weed DL. On the logic of causal inference. Am J Epidemiol. 1986; 123:965-979

37. Weiss NS. Inferring causal relationships. Am J Epidemiol.1981; 113:487