Unit for Veterinary Public Health and Epidemiology, University of Veterinary Medicine Vienna, Austria, Veterinaerplatz 1, Vienna, 1210, Austria.
BMC Infect Dis. 2020 Jun 26;20(1):448. doi: 10.1186/s12879-020-05156-7.
Why human tick-borne encephalitis (TBE) cases differ from year to year, in some years more 100%, has not been clarified, yet. The cause of the increasing or decreasing trends is also controversial. Austria is the only country in Europe where a 40-year TBE time series and an official vaccine coverage time series are available to investigate these open questions.
A series of generalized linear models (GLMs) has been developed to identify demographic and environmental factors associated with the trend and the oscillations of the TBE time series. Both the observed and the predicted TBE time series were subjected to spectral analysis. The resulting power spectra indicate which predictors are responsible for the trend, the high-frequency and the low-frequency oscillations, and with which explained variance they contribute to the TBE oscillations.
The increasing trend can be associated with the demography of the increasing human population. The responsible GLM explains 12% of the variance of the TBE time series. The low-frequency oscillations (10 years) are associated with the decadal changes of the large-scale climate in Central Europe. These are well described by the so-called Scandinavian index. This 10-year oscillation cycle is reinforced by the socio-economic predictor net migration. Considering the net migration and the Scandinavian index increases the explained variance of the GLM to 44%. The high-frequency oscillations (2-3 years) are associated with fluctuations of the natural TBE transmission cycle between small mammals and ticks, which are driven by beech fructification. Considering also fructification 2 years prior explains 64% of the variance of the TBE time series. Additionally, annual sunshine duration as predictor for the human outdoor activity increases the explained variance to 70%.
The GLMs presented here provide the basis for annual TBE forecasts, which were mainly determined by beech fructification. A total of 3 of the 5 years with full fructification, resulting in high TBE case numbers 2 years later, occurred after 2010. The effects of climate change are therefore not visible through a direct correlation of the TBE cases with rising temperatures, but indirectly via the increased frequency of mast seeding.
为什么人类 tick-borne encephalitis（TBE）病例每年都有差异，有时甚至高达 100%，目前仍不清楚原因。导致趋势增减的原因也存在争议。奥地利是欧洲唯一一个有 40 年 TBE 时间序列和官方疫苗接种覆盖率时间序列的国家，可以用来调查这些悬而未决的问题。
开发了一系列广义线性模型（GLMs）来确定与 TBE 时间序列趋势和波动相关的人口统计学和环境因素。观察到的和预测的 TBE 时间序列都进行了频谱分析。所得的功率谱表明哪些预测因子是导致趋势、高频和低频波动的原因，以及它们对 TBE 波动的解释方差贡献了多少。
增加的趋势可以与人口增加的人类人口统计学相关。负责的 GLM 解释了 TBE 时间序列的 12%的方差。低频波动（10 年）与中欧大尺度气候的十年变化有关。这些变化很好地被所谓的斯堪的纳维亚指数描述。这个 10 年的波动周期被净移民等社会经济预测因子所加强。考虑到净移民和斯堪的纳维亚指数，GLM 的解释方差增加到 44%。高频波动（2-3 年）与小哺乳动物和蜱之间的自然 TBE 传播周期的波动有关，这些波动是由山毛榉果实的结果所驱动的。同时考虑到前两年的果实结果，可解释 TBE 时间序列的 64%的方差。此外，作为人类户外活动预测因子的年日照时间可将解释方差增加到 70%。
本文提出的 GLMs 为 TBE 年度预测提供了基础，主要由山毛榉果实决定。在完全结果的 3 年中，有 5 年中有 2 年之后 TBE 病例数量很高，这发生在 2010 年之后。因此，气候变化的影响并不是通过 TBE 病例与气温升高的直接相关性，而是通过增加的结实频率间接地显现出来。
