Cargando…

Improved correlation of human Q fever incidence to modelled C. burnetii concentrations by means of an atmospheric dispersion model

BACKGROUND: Atmospheric dispersion models (ADMs) may help to assess human exposure to airborne pathogens. However, there is as yet limited quantified evidence that modelled concentrations are indeed associated to observed human incidence. METHODS: We correlated human Q fever (caused by the bacterium...

Descripción completa

Detalles Bibliográficos
Autores principales: van Leuken, Jeroen PG, van de Kassteele, Jan, Sauter, Ferd J, van der Hoek, Wim, Heederik, Dick, Havelaar, Arie H, Swart, Arno N
Formato: Online Artículo Texto
Lenguaje:English
Publicado: BioMed Central 2015
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4440286/
https://www.ncbi.nlm.nih.gov/pubmed/25888858
http://dx.doi.org/10.1186/s12942-015-0003-y
Descripción
Sumario:BACKGROUND: Atmospheric dispersion models (ADMs) may help to assess human exposure to airborne pathogens. However, there is as yet limited quantified evidence that modelled concentrations are indeed associated to observed human incidence. METHODS: We correlated human Q fever (caused by the bacterium Coxiella burnetii) incidence data in the Netherlands to modelled concentrations from three spatial exposure models: 1) a NULL model with a uniform concentration distribution, 2) a DISTANCE model with concentrations proportional to the distance between the source and residential addresses of patients, and 3) concentrations modelled by an ADM using three simple emission profiles. We used a generalized linear model to correlate the observed incidences to modelled concentrations and validated it using cross-validation. RESULTS: ADM concentrations generally correlated the best to the incidence data. The DISTANCE model always performed significantly better than the NULL model. ADM concentrations based on wind speeds exceeding threshold values of 0 and 2 m/s performed better than those based on 4 or 6 m/s. This might indicate additional exposure to bacteria originating from a contaminated environment. CONCLUSIONS: By adding meteorological information the correlation between modelled concentration and observed incidence improved, despite using three simple emission profiles. Although additional information is needed – especially regarding emission data - these results provide a basis for the use of ADMs to predict and to visualize the spread of airborne pathogens during livestock, industry and even bio-terroristic related outbreaks or releases to a surrounding human population. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12942-015-0003-y) contains supplementary material, which is available to authorized users.