Signals Before Symptoms: Integration of Digital Mobility Data and Syndromic Surveillance for Real-Time Outbreak Prediction in Europe

Matteo Romano Alnagib¹, Giulia Ferrara², Luca Moretti³, Davide Rinaldi⁴⁻⁵, Sofia Marchetti⁶

Abstract

Background: Traditional infectious disease surveillance systems often detect outbreaks after community transmission has already intensified. Digital mobility metrics and real-time syndromic surveillance may provide earlier warning signals and enhance predictive accuracy.

Methods: A multicenter European retrospective modeling study was conducted across Italy, France, Germany, and the United Kingdom from January 2022 to December 2025. Aggregated anonymized mobility datasets (n=6.8 million users) were integrated with syndromic surveillance reports and confirmed respiratory infection case counts. Time-series machine learning models were developed to compare traditional surveillance-only approaches versus hybrid digital-integrated models. Predictive performance was assessed using area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and lead-time gain.

Results: Digital mobility increases above 15% of seasonal baseline were significantly associated with outbreak onset within 7–12 days (adjusted RR 1.42; 95% CI 1.28–1.58; p<0.001). The hybrid forecasting model demonstrated superior predictive accuracy (AUC 0.93) compared to traditional surveillance models (AUC 0.79; p<0.001). Median outbreak detection lead-time improved by 12 days. Sensitivity reached 88% with specificity of 83% across regions. Model robustness persisted across seasonal and geographic variations.

Conclusion: Integration of digital mobility analytics with syndromic surveillance significantly enhances early outbreak detection in European settings. Adoption of hybrid digital-epidemiologic systems could substantially strengthen public health preparedness and response capacity.