| Summary: | Spatio-temporal disease mapping models can be used to describe the geographical pattern of disease incidence across space and time. This paper discusses the development and application of spatio-temporal disease models based on generalized linear mixed models (GLMM) incorporating spatially correlated random effects, temporal effects and space–time interaction. Further, the models are fitted within a hierarchical Bayesian framework with Integrated Nested Laplace Approximation (INLA) methodology. The main objectives of this study are to choose the model that best represents the pattern of dengue incidence in Peninsular Malaysia from 2015 to 2017, to estimate the relative risk of disease based on the model selected and to visualize the risk spatial pattern and temporal trend. The models were applied to weekly dengue fever data at the district level in Peninsular Malaysia as reported to the Ministry of Health Malaysia from 2015 to 2017. In conclusion, it can be seen that there was a difference in dengue trend for every district for 2015–2017 and the models used was effective in identifying the high and low risk areas of dengue incidence. © 2022, The Author(s), under exclusive licence to Malaysian Mathematical Sciences Society and Penerbit Universiti Sains Malaysia.
|
|---|