Shifting seasonal patterns of influenza before, during and after the COVID-19 pandemic in Shanghai, China: A population-based surveillance study, 2015–2023

• Scientific question: The coronavirus disease 2019 (COVID-19) pandemic and non-pharmaceutical interventions (NPIs) significantly disrupted seasonal influenza activity. It is essential to investigate changes in the influenza seasonality to inform the development and optimization of influenza prevention and control strategies in the future. • Evidence before this study: Influenza activity decreased remarkably globally during the COVID-19 epidemic. However, few studies have explored how influenza seasonality has changed in the era of normalized COVID-19 transmission. • New findings: During the COVID-19 pandemic, nearly no influenza viruses were detected until January 2021. Only B/Victoria was circulated for about eight months, lower than the peaks of each year before the COVID-19 epidemic. In 2023, the first epidemic began one month later than the 8-year average and reached the highest peak. The second epidemic was unusually not summer-based and entered the winter-spring epidemic by approximately one month in advance. The dominant influenza strains shifted from multiple strains to a single subtype/lineage. The age preference of influenza A was changed after the relaxation of COVID-19 control measures. • Significance of the study: The seasonal patterns of influenza varied significantly across different stages of the COVID-19 pandemic, including differences in epidemic onset, peak timing, end, periodicity, and dominant strains. Stability of influenza seasonal changes requires longer-term surveillance, as well as further virologic and epidemiologic studies. Seasonal influenza activity was disrupted due to the emergence of coronavirus disease 2019 (COVID-19) and the implementation of non-pharmaceutical interventions. In this study, we analyzed weekly influenza-like illness (ILI) cases and influenza positivity rate by subtype and age group from the Shanghai influenza surveillance system from 2015 to 2023. A wavelet analysis was conducted to explore the seasonal pattern of influenza. The average weekly ILI consultation rate was 20.28 ILI cases per 1,000 consultations, with 40,778 of 173,155 specimens (22.63 %) testing positive for influenza. During the COVID-19 pandemic, nearly no influenza viruses were detected until January 2021. Only B/Victoria was circulated for about eight months and peaked in January 2022, with a positivity rate of 47.48 %, which was lower than the peaks in each year before the COVID-19 pandemic. In 2023, the first epidemic, dominated by A(H1N1), began in week 8 (one month later than the 8-year average) and peaked at 76.89 %. The second epidemic, which was unusually not summer-based and driven by A(H3N2), started in week 38 and peaked in week 48 at 75.31 %, advancing the winter-spring epidemic by approximately one month. The age preference of influenza A changed during the era of normalized COVID-19 transmission. The seasonal patterns of influenza varied significantly across different stages of the COVID-19 pandemic, including differences in epidemic onset, peak timing, duration, periodicity, and dominant strains. Stability of seasonal changes in influenza requires longer-term surveillance, as well as further virologic and epidemiologic studies.