Infectious diseases, particularly those that are transmitted by mosquitoes, are of special interest for public health officials because of the possible consequences of climate change on the incidence of such diseases. There are many different diseases that are of concern, such as dengue and malaria, but the one of focus for this paper is Barmah Forest Virus (BFV). On the coast of Queensland, Australia, Naish and her colleagues (2013) acquired data on the number of BFV cases, climate temperatures, rainfall levels, tidal levels, and socio-economic circumstances in the area during the years 2000—2008. In combination with climate prediction grids for 2025, 2050, and 2100, the analysts modeled possible BFV distributions at both the current climate levels and under possible forecasted climate changes, finding large expected changes in the BFV distribution. Predicting future locations of BFV occurrence can facilitate prevention planning.–Posted by Sarah King
Naish S., Mengersen K., Hu W., Tong S., 2013. Forecasting the Future Risk of Barmah Forest Virus Disease under Climate Change Scenarios in Queensland, Australia. PLoS ONE.
Naish and her colleagues from the Queensland University of Technology gathered the data that they needed and the climate grids from outside sources: Queensland Health, Australian Bureau of Meteorology, Australian Bureau of Statistics, and Queensland Transport. The group divided Queensland into many blocks to determine the existing incidence of BFV in different parts of the state. Queensland is known for varying climates and weather, so each block provided insight into the possible connection between different climates and the incident of BFV. Focusing on the climate, socio-economic, and tidal variables, Naish and her team created multivariable logistic regression models to predict the chance of a BFV outbreak depending on the different levels of each variable in a given area. Once the current models were determined, the best-fit model (accuracy of 90.2%, sensitivity of 98%, specificity of 88.4%) was used to predict the outcomes in the possible future climates given by the 2025, 2050, and 2100 grids.
The model was then used to predict the regional risk of BFV under different assumptions. The first was based on varying rainfall but constant minimum temperature, the second on varying minimum temperature but constant rainfall, and the third with both variables varying. Naish and her team found that the areas around Brisbane and Cairns are at a particularly high risk of increase BFV incidence in 2025, 2050, and 2100 but that risk decreases between 2025 and 2100 in all the other areas on the coast of Queensland.
The model that Naish et al created proved to have strong predictive ability of the future risk of BFV outbreaks in different climate change situations. It uses accessible predictors, is accurate in comparison to independent data, and generates outputs that are useful in the public health decision-making process. The variables that are used consider multiple concerns and incorporate the climate, human elements, and mosquito livelihood. Each variable proved to be a significant indicator of high-risk BFV outbreaks.