Login / Signup

A Predictive Model Has Identified Tick-Borne Encephalitis High-Risk Areas in Regions Where No Cases Were Reported Previously, Poland, 1999-2012.

Paweł StefanoffBarbara RubikowskaJakub BratkowskiZbigniew UstrnulSophie O VanwambekeMagdalena Rosinska
Published in: International journal of environmental research and public health (2018)
During 1999–2012, 77% of the cases of tick-borne encephalitis (TBE) were recorded in two out of 16 Polish provinces. However, historical data, mostly from national serosurveys, suggest that the disease could be undetected in many areas. The aim of this study was to identify which routinely-measured meteorological, environmental, and socio-economic factors are associated to TBE human risk across Poland, with a particular focus on areas reporting few cases, but where serosurveys suggest higher incidence. We fitted a zero-inflated Poisson model using data on TBE incidence recorded in 108 NUTS-5 administrative units in high-risk areas over the period 1999–2012. Subsequently we applied the best fitting model to all Polish municipalities. Keeping the remaining variables constant, the predicted rate increased with the increase of air temperature over the previous 10–20 days, precipitation over the previous 20–30 days, in forestation, forest edge density, forest road density, and unemployment. The predicted rate decreased with increasing distance from forests. The map of predicted rates was consistent with the established risk areas. It predicted, however, high rates in provinces considered TBE-free. We recommend raising awareness among physicians working in the predicted high-risk areas and considering routine use of household animal surveys for risk mapping.
Keyphrases
  • climate change
  • electronic health record
  • endothelial cells
  • emergency department
  • air pollution
  • risk assessment
  • high resolution
  • cross sectional
  • machine learning
  • artificial intelligence
  • breast cancer risk