A study published on the medRxiv preprint server has reconstructed the early spatial spread of pandemic respiratory viruses in the United States, including COVID-19 and A/H1N1 (H1N1pdm) influenza. The research aimed to understand why containing new pandemics is challenging due to chance events, travel hubs, and delayed detection, and to identify potential surveillance strategies.
Understanding Viral Spread Dynamics
Novel respiratory pathogens like SARS-CoV-2 (COVID-19) and H1N1pdm influenza have caused significant global morbidity, mortality, and socioeconomic disruption. Their rapid adaptation and spread across geographic boundaries highlight the importance of early detection and public health responses. Human mobility, including air travel and commuting, is crucial for viral transmission.
Limited case reporting and surveillance data hinder a full understanding of early viral spread at granular geographic scales, especially during initial pandemic stages when under-ascertainment is substantial.
Inference Framework and Methodology
Researchers developed an inference framework by combining city-level influenza-like illness (ILI) records from medical claims with estimated daily SARS-CoV-2 infections at the county level. This included both reported and unreported cases to reconstruct transmission pathways of H1N1pdm influenza and SARS-CoV-2 across U.S. Metropolitan Statistical Areas (MSAs).
MSAs, as densely populated regions with strong social and economic connections, were used as units for analyzing pathogen spread.
A process-based stochastic transmission model was employed to quantify uncertainty in early spatial dynamics. This model incorporated inter-MSA air travel, commuting patterns, and pathogen superspreading potential.
Mapping Respiratory Virus Spread
- Simulated Outbreak: A hypothetical novel respiratory virus originating in Minnesota was simulated. The simulation showed a hub-and-spoke transmission network structure, consistent with prior studies. Significant variation in transmission links was observed, with 56.9% appearing in less than 20% of simulations, indicating variability increases with lower transmissibility and greater superspreading potential.
- Validation: A prediction-based inference framework was validated using a simulated outbreak. The algorithm achieved 79.3% precision and 78.2% recall in identifying true transmission links for individual inference realizations. Aggregating results improved overall inference performance.
- SARS-CoV-2 Network: The inferred SARS-CoV-2 transmission network comprised 304 links, also exhibiting a hub-and-spoke pattern. Seattle and New York were identified as key sources of national spread via air travel, while regional areas like Chicago, Atlanta, New Orleans, and San Francisco facilitated local dissemination. Most inter-metropolitan transmission events occurred between late February and mid-March 2020.
- H1N1pdm Influenza Network: For pandemic influenza, early seeding locations were assumed to be San Diego, San Antonio, and New York. The reconstructed network for H1N1pdm influenza differed structurally from SARS-CoV-2 but also shared a hub-and-spoke pattern. Major international travel centers were often highly connected. However, some MSAs with high international travel volume (e.g., Miami, Los Angeles) were not major transmission sources, suggesting travel volume alone does not predict spatial spread. Fewer high-confidence transmission links were inferred for H1N1pdm influenza, possibly due to coarser temporal resolution and sparser surveillance data.
Conclusions and Future Strategies
The study concludes that pandemic respiratory pathogens can establish widespread local transmission rapidly, often before detection or intervention is possible. Both pandemics analyzed, despite differences, showed common transmission areas and stochastic factors complicating containment efforts.
Simulation results suggest that early detection strategies, such as airport wastewater surveillance, may be most effective when deployed across a broad set of metropolitan hubs rather than a limited number of major airports. The effectiveness of expanded surveillance depends on pairing detection with interventions that reduce onward transmission. Future research is needed to refine simulation models with more social and demographic details and to evaluate pragmatic surveillance and intervention strategies for future pandemics.
Important Note
medRxiv publishes preliminary scientific reports that are not peer-reviewed and should not be regarded as conclusive, used to guide clinical practice or health-related behavior, or treated as established information.