BACKGROUND: Following the withdrawal of non-pharmaceutical interventions (NPI) implemented during the COVID-19 pandemic, the resurgence of respiratory viral infections in children deviated from established pre-pandemic seasonal patterns. AIM: To utilize the disruption in the seasonality of respiratory viral infections caused by NPI measures to identify key drivers of seasonality before and since the COVID-19 pandemic. METHODS: We applied a Bayesian model to crossover-predict the incidence of infections with Human Metapneumovirus, Influenza Viruses and Respiratory Syncytial Viruses (RSV) in children admitted to Berlin University Children's Hospital in the pre-pandemic and in the post-pandemic era (from 08/2021) after training based on data from the respective other period. We analysed the effect of environmental and viral coefficients on the accuracy of prediction. RESULTS: Incidence of infection with respiratory viruses was accurately predicted both for the pre-pandemic and the post-pandemic era. After excluding autoregression, no single factor independently predicted the incidence of infection with Metapneumovirus or RSV. Incidence of infection with Influenza Virus was predicted by a previous RSV epidemic as a single coefficient. CONCLUSION: Our data show that it takes several years before seasonal patterns of common viral respiratory infections are reestablished. However, drivers of the abnormal seasonality of respiratory virus infections in children following the lifting of NPI measures during the COVID-19 pandemic did not differ from drivers of the pre-pandemic seasonality. Both environmental factors and viral interference shape the seasonal pattern of respiratory viral infections. Incidence patterns of Influenza seem to be more prone to virus-virus interactions than those of RSV.
Terliesner et al. (Tue,) studied this question.