Investigation of the prevalence of tularaemia under the aspect of climate change.

Introduction Wild animal habitats are greatly influenced by geographical location, structure, climate, fauna and flora and may extend deep into human settlement areas, depending on animal species. While this may be in accordance with ecological requirements, it also poses the danger of animal-human disease transmission. This research project aims at investigating the impact of climate and weather on the prevalence of tularaemia (pathogen: Francisella tularensis) in hare populations in the lowlands of eastern Austria. The fundamental relationship between bacterial infectious diseases, such as tularaemia, and climate parameters can be identified by the ambient conditions required by the pathogen. Bacteria reproduce at moderate, rising temperatures, are destroyed at elevated temperatures and are resistant to cold. In addition, the population density of host animals and the abundance of potential disease vectors (ticks, gnats) also play a decisive role. The infection is transmitted to animals and humans by direct contact with infected animals and vectors, by inhalation of pathogens or the consumption of insufficiently cooked hare meat. Several cases are recorded in Austria each year, but the number of unreported cases is likely to be far higher. The causative agent of tularaemia is also considered as a potential biological weapon due to its low infection dose. Material and methods A total of 271 cases of tularaemia in hares were recorded in the area under investigation (Lower Austria, Burgen-land, Styria) in the period from 1994 to 2005 and georeferenced according to sender postcode. Temperature and precipitation data for the selected region were available from 30 weather stations of the Central Institute for Meteorology and Geodynamics. These data provided the basis for calculating an altitude dependent temperature distribution for suitable monthly means and period sums. The areal distribution of precipitation was calculated using the geostatistical universal kriging method without taking the influence of altitude into account. Those data were used for a 2 step analysis. The first step led to boundary values for the spatial distribution of tularaemia. These boundaries were used to estimate the distribution of tularaemia at the year 2035. The second step explained the different annual incidences using actual climate data. Results First step - finding spatial boundary values and estimate the areal distribution of tularaemia till 2035 A high incidence probability, based on the local isoline encircling the study area, was obtained for annual precipitation totals below 720 mm, summer precipitation below 180 mm, winter temperatures above 0.5 degrees C and mean May temperatures above 14 degrees C (Fig. 1). These limit values allowed a calculation of the diseases spatial distribution for current and future conditions. A climate change induced warming of 2 to 4 degrees C was assumed for predicting the distribution area of the disease by 2035, with warming expected to be more pronounced at higher altitudes than in lowlands. Fig. 2 shows the possible spatial distribution of tularaemia in 2035 following a rise in mean annual temperatures. Precipitation was not taken into account due to the lack of a suitable scenario. Under these conditions, tularaemia will slowly spread from the eastern lowlands via the Danube valley to the west and via southern Styria further to the south. Additional incidents of the disease could also occur in inner Alpine areas providing favourable climatic conditions. Second step - explaining coherences between climate parameter and incidence Inside of tularaemia zones a clear correlation was established between the 2 climate parameters and local disease incidence, which can be represented by the following linear regression model: Number of cases per year = 52.12 + 4.08 x (average of monthly mean temperatures for December, January and February) - 3.46 x (monthly mean temperature for May) + 0.26 x (precipitation total for June and July) This formula does not allow absolute incidence in nature to be calculated, since it is based on sample data of one specific region. Of special note, however, is the significant (p<0.05) influence of the parameters selected on the incidence rate of the disease and the coefficient of determination obtained (R(2) = 74.6%). It becomes clear that about 3/4 of inter-year differences can be explained by temperature and precipitation conditions: higher winter temperatures result in an increase in incidence, while higher May temperatures lead to a decrease; high precipitation in summer again has an increasing effect. The ideal conditions for the spread of the disease are thus warm winters combined with low May temperatures and high precipitation in summer. The result represents a feasible development of hares. Warm winters increase the hare population. Low May temperatures and wet summers degrade leverets. So, tularaemia finds better conditions to increase. This correlation was derived from observations and of course does not apply to arbitrary temperature and precipitation values. Conclusion In summary it can be stated that climate is the key parameter for explaining the distribution of tularaemia in the past and that limit values for the individual climate parameters can be identified. The expected warming could result in a massive expansion of the potential tularaemia distribution area. It would therefore be important to inform risk groups (hunters, foresters, farmers, laboratory staff, taxidermists, housewives etc.) beyond current prevalence regions and recommend taking preventative measures of work hygiene (protective gloves, moistening the fur when skinning hares, insect protection, face masks in the lab) when handling hares and rodents and following good kitchen hygiene practices when preparing and cooking hares.