Global Detection of Respiratory Illness Outbreaks inTravelers: A Statistical Approach using GeoSentinel Data

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 important underutilized opportunity to complement existing surveillance systems, although reliable denominator data on travel volume are often unavailable. Using GeoSentinel data from 2015-2019, capturing syndromic and etiologic LRTI cases, we modeled baseline epidemiology in travelers by comparing generalized linear mixed models (GLMMs) using out-of-sample metrics. A She-whart control-chart framework, accounting for increases in travel volume under non-epidemic conditions, was applied to detect deviations from expected trends. The preferred hybrid autoregressive model incorpo-rated country-specific fixed effects, random seasonal effects, and a latent temporal autocorrelation structure, and was evaluated for goodness-of-fit in pre-pandemic (2015-2019) and post-pandemic (2023-2024) periods before retrospective application to 2020 data to identify early COVID-19 signals. The hybrid autoregressive GLMM performed best for modeling baseline epidemiology. Applied retrospectively to early 2020 data from 64 countries, 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 conservative signal was also detected in Italy, though driven primarily by influenza A and B rather than novel syndromic cases. Combining traveler surveillance with this statistical framework--integrating GLMMs for baseline modeling and Shewhart charts for outbreak detection--may support early detection of acute LRTI outbreaks despite absent denominator data, positioning GeoSentinel as a valuable complementary network for global health security and pandemic preparedness. SignificanceGlobal interconnectivity accelerates spread of pathogens, increasing pandemic potential. Travel medicine networks are underutilized for outbreak detection of respiratory diseases. Absence of denominator data, complex seasonal and autocorrelated baselines can mask early signaling of outbreaks. We used a flexible baseline model and robust control charts to signal increased reporting without denominator data. By retrospectively applying this statistical framework to surveillance data from international travelers returning from 64 countries, we identified an increase in influenza-like illness among travelers from China by week 5 of 2020, well before the WHO officially declared COVID-19 a pandemic. We demonstrated that traveler surveillance can operate as a scalable, proactive early-warning system, strengthening global health and enabling identification of threats before they escalate into international crises.