Emergence of tick-borne diseases at northern latitudes in Europe: a comparative approach

The factors that drive the emergence of vector-borne diseases are difficult to identify due to the complexity of the pathogen-vector-host triad. We used a novel comparative approach to analyse four long-term datasets (1995–2015) on the incidence of tick-borne diseases in humans and livestock (Lyme disease, anaplasmosis and babesiosis) over a geographic area that covered the whole of Norway. This approach allowed us to separate general (shared vector) and specific (pathogen reservoir host) limiting factors of tick-borne diseases, as well as the role of exposure (shared and non-shared pathogens in different hosts). We found broadly similar patterns of emergence across the four tick-borne diseases. Following initial increases during the first decade of the time series, the numbers of cases peaked at slightly different years and then stabilized or declined in the most recent years. Contrasting spatial patterns of disease incidence were consistent with exposure to ticks being an important factor influencing disease incidence in livestock. Uncertainty regarding the reservoir host(s) of the pathogens causing anaplasmosis and babesiosis prevented a firm conclusion regarding the role of the reservoir host-pathogen distribution. Our study shows that the emergence of tick-borne diseases at northern latitudes is linked to the shared tick vector and that variation in host-pathogen distribution and exposure causes considerable variation in emergence.


Potential biases in the disease records
Processes potentially creating bias may be linked to change in 1) diagnostic criteria, 2) reporting and/or 3) treatment. Overall, these uncertainties should not affect the spatial pattern as Lyme disease records have an overall high quality 4 .

B. Babesiosis and anaplasmosis in cattle.
The national database on babesiosis and anaplasmosis in cattle covers over 90% of all dairy cattle in Norway 5 . Each individual dairy cow has a Cow Health Card which is registered on a database, where information with the veterinarian's diagnosis, description of the case and subsequent treatment are available. These data are part of a dairy recording system, which has been established nationwide since 1975 6 .

1a) Diagnostic of babesiosis.
Babesiosis is easily recognized with symptoms of bloody urine, which is pathognomonic with very few differential diagnoses. 1b) Diagnostic of anaplasmosis. It is more difficult to diagnose anaplasmosis compared to both Lyme disease and babesiosis. The bacterium A. phagocytophilum causing anaplasmosis attacks the white blood cells 7 , therefore causing more general symptoms like high fever which will occur in many diseases. In anaplasmosis regularly causes immune depression leading to secondary diseases, and it could be difficult to identify anaplasmosis as the primary disease. However, the diagnostic criteria have not changed over the time. Anaplasmosis is therefore likely underreported, but this does not necessarily imply bias in terms of change over time. 2) Reporting rates. The data quality for the major diseases, like mastitis, is considered very good with about 82% coverage in dairy cattle 8 . For disease in calves incidence is more underreported with coverage of 40 % of the true incidence 9 . However, calf disease reporting has increased and improved during the study period from 1995 till 2016. From 2008 it was possible for veterinarians to record disease both in cattle and sheep by digitally directly into the databases. During 2016 about 90% of data was entered directly, digitally. Interestingly, as data coverage in calves and young stock improved, the incidence of babesiosis declined.
Despite that the largest improvement of recording disease in calf and young stock was after 2005, these two diseases have been dropping in cattle after 2005. We believe the recording quality could not be the main reason for this initial increase and not at all for the subsequent decrease after 2005. 3) Treatment. During the period 1995 to 2016 the recorded incidence of major diseases in dairy cattle has decreased by 60-70% due to intensive prevention and breeding program 6 . However, this is unlikely to be the case for rare disease like babesiosis and anaplasmosis. The typical treatment for anaplasmosis is the antibiotic Tetracycline, which is very seldom used in Norway. A general aim of reducing antibiotic use was initiated in 1995, and the total use of Tetracycline in Norway in dairy farming has declined by 70%. However, since babesiosis is not treated this way, the largely similar pattern of changes in this disease's incidence, suggests this reduction in overall use of antibiotic has not markedly affected the specific treatment of anaplasmosis. Babesiosis is treated with imidacard, and the use of it is not registered in Norway.
C. Anaplasmosis in sheep. 1) Diagnostic criteria. As for anaplasmosis in cattle, it is difficult to diagnose anaplasmosis in sheep; nevertheless the way it is diagnosed has not changed. 2) Reporting rates. Most cases in the database likely come from the period of grazing infield during early spring, when lambs first get exposed to ticks. When sheep are grazing in the outfields during summer it is typically with limited surveillance in many cases.
Though not all infected sheep contract anaplasmosis 10 , the mortality of lambs may reach 20-30% in some flocks and survivors often suffer reduced body growth 11 . Hence, the total number of cases is likely markedly underreported. Further, there is no mandatory reporting of anaplasmosis in the sheep control in Norway. Hence, there is a higher likelihood of underreporting and bias compared to the other 3 disease records. The fact that we used the number of sheep in the database rather than total number of sheep in the municipality when calculating incidence will to some extent control for this. However, we cannot exclude the possibility of an increased level of recording of diseases as such, in the database over time.
Therefore, we have entered overall reporting of diseases as a covariate in our analysis. 3) Treatment. The way of treatment anaplasmosis in sheep has not changed over time.