The new paper published in ScienceDirect discusses the potential of using citizen science tick reports to predict tick–human interactions and TBE risk, alongside socioeconomic and environmental data. Although tick-borne encephalitis (TBE) is a major public health risk in many countries, systematic surveillance is limited due to the cost and effort required for field monitoring.
In this study, statistical and machine learning models were applied to citizen science reports from Sweden, combined with socioeconomic and environmental data, to identify key drivers of TBE and assess predictive performance.
Over the past decade, climate change and land use patterns have led to significant changes in factors such as temperature and population density, creating more suitable habitats for ticks while also increasing human–tick encounters. Consequently, there is a growing need for improved methods of surveillance and prediction of tick-borne diseases.
The study used citizen science data submitted to the Swedish Veterinary Agency (SVA) through the online platform Rapportera Fästing, where users uploaded tick photos and provided details such as the location and host. These data were used to identify the factors that best explain where and when people encounter ticks. The results were then used to predict the number of tick-borne encephalitis (TBE) cases at the municipality level, alongside environmental, social, economic and ecological data.
The climate and environmental data included variables such as air temperature, soil temperature, precipitation, minimum and maximum temperature, and dew point temperature. Socioeconomic data included dog registrations, urban and rural population distribution, recreational land use, total land use, the number of summer houses (as an indicator of seasonal human presence in nature), and the Gini index, which measures income or wealth inequality.
Among the range of modelling strategies evaluated, XGBoost was found to be the best performing model which was used in the study. Citizen science data was one of the important predictors, after temperature and rural population size, for identifying TBE case counts, suggesting its potential for early warning systems for tick-borne disease risk.
Read the paper here:https://www.sciencedirect.com/science/article/pii/S2589004226022571?via%3Dihub
Reference: Liu, Y., Guo, J., Fransson, P., Widgren, S., Omazic, A., & Rocklöv, J. (2026). Exploring prediction of tick and tick-borne encephalitis cases in Sweden using citizen science data. iScience, 29, 116879. https://doi.org/10.1016/j.isci.2026.116879
