Aswi Aswi, Susanna Cramb, Earl Duncan, Wenbiao Hu, Gentry White, Kerrie Mengersen
A range of Bayesian models have been used to describe spatial and temporal patterns of disease in areal unit data. In this study, we applied two Bayesian spatio-temporal conditional autoregressive (ST CAR) models, one of which allows discontinuities in risk between neighbouring areas (creating ‘groups’), to examine dengue fever patterns. Data on annual (2002–2017) and monthly (January 2013 - December 2017) dengue cases and climatic factors over 14 geographic areas were obtained for Makassar, Indonesia. Combinations of covariates and model formulations were compared considering credible intervals, overall goodness of fit, and the grouping structure. For annual data, an ST CAR localised model incorporating average humidity provided the best fit, while for monthly data, a single-group ST CAR autoregressive model incorporating rainfall and average humidity was preferred. Using appropriate Bayesian spatio-temporal models enables identification of different groups of areas and the impact of climatic covariates which may help inform policy decisions. © 2020
ARC Centre of Excellence for Mathematical and Statistical Frontiers, Queensland University of Technology, Australia; Universitas Negeri Makassar, Indonesia; School of Public Health and Social Work, Queensland University of Technology, Australia; Science and Engineering Faculty, School of Mathematical Science, Queensland University of Technology, GPO Box 2434, Brisbane, 4001, Queensland, Australia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock