Leveraging perturbations to infer the population dynamics of human rhinovirus and interaction of influenza A virus

Abstract Many respiratory pathogens co-circulate within human populations. Yet, how pathogen community structure shapes the dynamics of infectious diseases remains poorly understood. At the population level, investigating polymicrobial dynamics, with potential underlying competitive or cooperative interactions, is challenging, because of confounding factors such as differing seasonality. This is particularly true for endemic pathogens which typically exhibit stable periodic dynamics. Their disruption due to the implementation of non-pharmaceutical interventions during the COVID-19 pandemic thus represents a unique large-scale natural experiment that can be leveraged to provide valuable insights into the complex interplay between respiratory pathogens. Here, we focus on the population dynamics of human rhinovirus (common cold) and on the potential viral interference of influenza A virus (flu A), which is hypothesized to account for their asynchronous circulation. Using a Bayesian framework, we first show based on simulations that exogenous perturbations can be a powerful tool to disentangle the contribution of pathogen interaction from other epidemiological factors. We then apply our framework to long surveillance time series from the US and Canada spanning the COVID-19 pandemic. We estimate key parameters of rhinovirus but find no conclusive support for an influence of influenza A virus at the population level.