Mixed-effects modelling for crossed and nested data: an analysis of dengue fever in the state of Goias, Brazil

Abstract

Dengue fever is a viral disease transmitted by the mosquito Aedes aegypti. In order to avoid epidemics and deaths, this transmitting vector must be controlled. This work assembles, for the first time, data from multiple governmental bodies describing the number of dengue cases reported, and meteorological conditions in 20 cities in the Goias state, Brazil, from 2008 to 2015. We then apply generalised linear mixed modelling to this novel data set to model dengue occurrences in this state, where the tropical climate favours the proliferation of the main transmitting vector of this disease. The number of reported dengue cases is estimated using meteorological variables as fixed effects, and city and year data are included in the model as random effects. The proposed models can cope with complex data structures, such as nested data, while taking into account the particularities of each year dependent on the city under analysis. The results confirm that precipitation, minimum temperature, and relative air humidity contribute to the increase of dengue cases number, while year and city location are determining factors. Public policies, based on these new results, together with joint actions involving local populations, are essential to combat the vector transmitting dengue and avoid epidemics.