Global Detection of Respiratory Illness Outbreaks in Travelers: A Statistical Approach using GeoSentinel Data, 2015–2024
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Novel respiratory pathogens have pandemic potential, making epidemiologic surveillance of acute lower respiratory tract infections (acute LRTI) a global public health priority. Monitoring acute LRTI among international travelers provides an underutilized opportunity to complement existing surveillance systems, though reliable denominator data on travel volume are often unavailable.
Aim
We aimed to develop and validate a framework for detecting acute LRTI outbreaks among international travelers in the absence of reliable denominator data.
Methods
Using syndromic and etiologic GeoSentinel data from 2015 to 2019, we modeled baseline LRTI epidemiology in travelers by comparing generalized linear mixed models (GLMMs) and selecting the preferred model using out-of-sample metrics. A robust Shewhart control-chart framework, accounting for increases in travel volume under non-epidemic conditions, was applied to detect deviations from expected trends and retrospectively to 2020 data from 64 countries to identify early COVID-19 signals.
Results
The preferred hybrid autoregressive GLMM, incorporating country-specific fixed effects, random seasonal effects, and a latent temporal autocorrelation structure, demonstrated adequate goodness-of-fit across pre-pandemic and post-pandemic (2023 to 2024) periods. The framework detected an early syndromic signal in China under the conservative assumption of up to a threefold increase in travel volume, consistent with COVID-19 emergence; a signal was also detected in Italy, driven primarily by influenza rather than novel syndromic cases.
Conclusion
Combining traveler surveillance with this statistical framework may support early detection of acute LRTI outbreaks despite absent denominator data, positioning GeoSentinel as a valuable complementary network for global health security.
