Challenges and Opportunities in One Health: Google Trends Search Data.

Area of Science:

• Public Health
• Epidemiology
• Computational Biology

Background:

• Google Trends data offers insights into public interest and potential disease incidence.
• Predictive modeling using Google Trends for zoonotic diseases like Lyme disease is underutilized.
• Accurate zoonotic disease surveillance is crucial for public health interventions.

Purpose of the Study:

• To assess the utility of Google Trends data for predicting monthly state-level Lyme disease case counts in the U.S.
• To evaluate the performance of negative binomial models using various Lyme disease-related search terms.
• To explore the potential of Google Trends for broader zoonotic disease prediction and One Health research.

Main Methods:

• Collected Lyme disease case data (2010-2021) and Google Trends search data for relevant terms.
• Developed expanding window negative binomial regression models, accounting for seasonality.
• Evaluated model performance using Root Mean Squared Errors (RMSE) and visual comparisons.

Main Results:

• Models using Lyme disease-specific search terms demonstrated the highest predictive accuracy.
• Predictive performance varied significantly across different U.S. states.
• The models showed excellent predictive ability in certain regions, highlighting the specificity of search terms.

Conclusions:

• Google Trends data can be a valuable tool for predicting zoonotic disease incidence, particularly Lyme disease.
• Challenges include data availability and geographic unit mismatches, but opportunities exist for One Health integration.
• Further research should explore incorporating diverse data streams to enhance predictive performance for zoonotic diseases.