tetano
Editor, Senior Moderator
Int J Epidemiol
. 2026 Oct 14;55(6):dyag250.
doi: 10.1093/ije/dyag250.
Mojie Song 1 , Guang Li 2 , Yake Lei 3 , Yan Liu 1 , Sujin Zhang 1 , Hongtao Zhou 1 , Xiaobo Feng 4 , Yanqing Xu 1
Affiliations Expand
Background: Interactions among respiratory viruses may alter epidemic timing, healthcare demand, and immunization strategies in the postpandemic era. We analyzed multi-city surveillance data to characterize population-level viral interference among four major respiratory viruses.
Methods: From June 2024 to August 2026, 40 987 respiratory specimens were collected through sentinel surveillance across seven cities and routinely tested for multiple respiratory pathogens using nucleic-acid assays. City-specific Bayesian Distributed Lag Non-linear Models with multivariate meta-analysis were used to quantify the magnitude and lag patterns of viral interference. Age-stratified analyses assessed heterogeneity across life stages.
Results: All co-circulating viruses demonstrated reciprocal inhibitory associations [all cumulative relative risks (CRRs) < 1.00]. Respiratory syncytial virus (RSV) exerted the strongest inhibitory associations: a 10% increase in RSV positivity was associated with a 49.2% reduction in severe acute respiratory syndrome coronavirus 2 risk [CRR: 0.508 (0.402-0.640)], with less pronounced associations observed for influenza virus and human rhinovirus. Interference patterns differed substantially across virus pairs in both magnitude and temporal duration. Influenza produced immediate and sustained inhibition, whereas RSV exhibited delayed inhibitory associations peaking at weeks 5-6. Age-stratified analyses revealed substantial heterogeneity across age groups.
Conclusions: Respiratory viruses exhibit distinct pathogen-specific inhibitory associations with heterogeneous temporal dynamics, consistent with population-level viral interference. Incorporating these interactions into surveillance and forecasting models may improve epidemic prediction and support immunization planning and respiratory healthcare resource allocation in post-pandemic settings.
Keywords: Bayesian inference; distributed lag nonlinear models; respiratory viruses; viral interference.
. 2026 Oct 14;55(6):dyag250.
doi: 10.1093/ije/dyag250.
Magnitude and temporal dynamics of respiratory viral interference: a multi-city study in Central China
Mojie Song 1 , Guang Li 2 , Yake Lei 3 , Yan Liu 1 , Sujin Zhang 1 , Hongtao Zhou 1 , Xiaobo Feng 4 , Yanqing Xu 1
Affiliations Expand
- PMID: 43009322
- DOI: 10.1093/ije/dyag250
Abstract
Background: Interactions among respiratory viruses may alter epidemic timing, healthcare demand, and immunization strategies in the postpandemic era. We analyzed multi-city surveillance data to characterize population-level viral interference among four major respiratory viruses.
Methods: From June 2024 to August 2026, 40 987 respiratory specimens were collected through sentinel surveillance across seven cities and routinely tested for multiple respiratory pathogens using nucleic-acid assays. City-specific Bayesian Distributed Lag Non-linear Models with multivariate meta-analysis were used to quantify the magnitude and lag patterns of viral interference. Age-stratified analyses assessed heterogeneity across life stages.
Results: All co-circulating viruses demonstrated reciprocal inhibitory associations [all cumulative relative risks (CRRs) < 1.00]. Respiratory syncytial virus (RSV) exerted the strongest inhibitory associations: a 10% increase in RSV positivity was associated with a 49.2% reduction in severe acute respiratory syndrome coronavirus 2 risk [CRR: 0.508 (0.402-0.640)], with less pronounced associations observed for influenza virus and human rhinovirus. Interference patterns differed substantially across virus pairs in both magnitude and temporal duration. Influenza produced immediate and sustained inhibition, whereas RSV exhibited delayed inhibitory associations peaking at weeks 5-6. Age-stratified analyses revealed substantial heterogeneity across age groups.
Conclusions: Respiratory viruses exhibit distinct pathogen-specific inhibitory associations with heterogeneous temporal dynamics, consistent with population-level viral interference. Incorporating these interactions into surveillance and forecasting models may improve epidemic prediction and support immunization planning and respiratory healthcare resource allocation in post-pandemic settings.
Keywords: Bayesian inference; distributed lag nonlinear models; respiratory viruses; viral interference.