Respiratory disease seasonality is well recognized, but its mechanistic basis remains incompletely understood. This study integrates high-resolution influenza case data with observations of human behavior in built environments across US counties. The authors conclude that seasonal influenza patterns in the United States arise primarily from behavioral adaptations to environmental conditions—such as changes in indoor mixing—rather than from environmental effects alone.
The analysis emphasizes that routine patterns of human contact and the ways people change behavior in response to weather are central to the timing and spatial spread of influenza. The study frames seasonality as an emergent property of how people use and inhabit indoor spaces under varying weather conditions combined with structural features of buildings that influence transmission.
A key empirical observation reported is persistent summer influenza circulation in counties in the southeastern United States. According to the authors, this sustained transmission is shaped by two interacting factors: weather-driven increases in indoor crowding (behavioral adaptation to local climatic conditions) and vulnerabilities inherent to the local built environment that facilitate transmission.
These southeastern reservoirs can seed epidemics as early as August. From there, infections spread across the country in a spatiotemporal invasion pattern that begins in the south and progresses nationwide. The presence of summer circulation in these areas contrasts with the classic temperate-season view that focuses on winter environmental conditions as the dominant driver of influenza seasonality.
To test mechanistic explanations, the study implements a transmission model that integrates three main elements: intercounty connectivity, school calendars, and building-modulated transmissibility. When these components are combined, the model reproduces the observed spatiotemporal invasion from the southeastern reservoirs to other regions of the United States.
The modeling approach highlights how movement between counties (connectivity) and routine social structures (for example, the opening and closing of schools) interact with transmission conditions within buildings to shape epidemic timing and geography. The authors report that incorporating building-modulated transmissibility—an explicit representation of how indoor environments alter transmission risk—was important to reproducing empirical patterns.
The authors position their findings as complementary to explanations that emphasize aerosol persistence or enhanced viral survival under specific environmental conditions. Rather than contradicting those mechanisms, the results argue that human behavior—both routine contact patterns and adaptive responses to weather—plays a central role in shaping when and where influenza circulates.
By foregrounding behavior and built environment features, the analysis expands the mechanistic toolkit for understanding seasonality beyond abiotic drivers alone. This broader framing implies multiple interacting pathways (behavioral mixing, built environment-mediated transmission, and virological/environmental effects) contribute to the observed seasonal dynamics.
The study presents practical implications for public health preparedness. Specifically, mapping and modeling connectivity from southern, summer-persistent reservoirs to northern counties can inform surveillance and early-warning strategies. The authors suggest that shifting vaccination timing locally to precede estimated epidemic onset could reduce influenza burden, particularly in areas that receive seeding infections from southern reservoirs.
These policy-relevant suggestions rely on the study's modeling framework, which links local building-mediated transmissibility and behavioral patterns to larger-scale spread. The report does not provide operational vaccination schedules but highlights the potential benefit of aligning vaccination timing with locally driven epidemic onset.
All data and code used in the project are available at the authors' public repository: https://github.com/bansallab/flu_seasonality.
Ethical review information is reported: the study was reviewed by the Georgetown University Institutional Review Board and determined to be exempt from full IRB review under specified exemption categories or approved under expedited review protocols (protocol identifiers are reported in the source). The authors confirm that relevant ethical guidelines and participant consent procedures were followed as applicable.
Funding sources declared in the preprint include the National Institutes of Health (R35GM153478) and Flu Lab. The work is posted as a medRxiv preprint and has not been certified by peer review; the authors note this status and caution that the findings should not yet be used to guide clinical practice.
All summary points above are drawn from the preprint's abstract and associated article metadata. Detailed model parameters, numerical results, and full methodological descriptions are available in the preprint and the linked code repository. Because this work is a preprint, readers should interpret conclusions as preliminary pending peer review and independent replication.