Unravelling infectious disease eco-epidemiology using Bayesian networks and scenario analysis: A case study of leptospirosis in Fiji

• Published in: Environmental modelling & software : with environment data news 2017-11, Vol.97, p.271-286
• Main Authors: Lau, Colleen L., Mayfield, Helen J., Lowry, John H., Watson, Conall H., Kama, Mike, Nilles, Eric J., Smith, Carl S.
• Format: Article
• Language: English
• Subjects: Animal species; Bayesian analysis; Bayesian networks; Case studies; Complexity; Disease control; Disease transmission; Environmental health; Epidemiology; Infectious diseases; Infectious diseases epidemiology; Leptospirosis; Mathematical models; Minority & ethnic groups; Modelling; Public health; Regression analysis; Regression models; Subgroups; Zoonoses
• ISSN: 1364-8152; 1873-6726
• DOI: 10.1016/j.envsoft.2017.08.004

Regression models are the standard approaches used in infectious disease epidemiology, but have limited ability to represent causality or complexity. We explore Bayesian networks (BNs) as an alternative approach for modelling infectious disease transmission, using leptospirosis as an example. Data were obtained from a leptospirosis study in Fiji in 2013. We compared the performance of naïve versus expert-structured BNs for modelling the relative importance of animal species in disease transmission in different ethnic groups and residential settings. For BNs of animal exposures at the individual/household level, R2 for predicted versus observed infection rates were 0.59 for naïve and 0.75–0.93 for structured models of ethnic groups; and 0.54 for naïve and 0.93–1.00 for structured models of residential settings. BNs provide a promising approach for modelling infectious disease transmission under complex scenarios. The relative importance of animal species varied between subgroups, with important implications for more targeted public health control strategies. •Bayesian networks are a promising approach for modelling infectious diseases .•Models that represent causality perform better.•Model that account for dependencies between predictor variables perform better.•Bayesian networks are useful for predicting outcomes under complex scenarios .•Causal models provide important insights into eco-epidemiology.