Forecasting of influenza activity and estimating the impact of COVID-19 pandemic in Hong Kong

Abstract

Background: Various public health and social measures (PHSMs) have been implemented worldwide to fight against COVID-19 pandemic since early 2020. These measures, including mask wearing and social gathering restrictions, would also prevent virus transmission in general, thus affected the transmission dynamics of influenza viruses. This study was to assess the impact of COVID-19 PHSMs on the activity of influenza viruses in Hong Kong in 2020. Methods: We developed a statistical regression based forecasting framework by using surveillance and virological data on influenza virus activity, meteorological data, and school holiday/closure data in Hong Kong. We performed short-term (1-4 weeks ahead) and long-term (1-52 weeks ahead) forecasts for influenza virus activity in 2020 seasons. Cross-validation was used for model selection based on the forecast performance. We evaluated the impact of COVID-19 PHSMs on influenza activity by comparing counterfactual and observed attack rate during 2019-20 seasons. Results: Influenza activity dropped to a negligible level since February 2020 in Hong Kong. Absolute humidity and ozone concentration were found to be potential predictors. The short-term forecast could reach the coefficient of determination of >0.5. We estimated the overall attack rate of 27.2% in 2019-20 season with two peaks. The estimated reduction in attack rate was 79% for the winter peak during the period of December 2019 – March 2020, and 87.7% overall for the 2019/20 season. Conclusions: Our results suggest substantial reduction in influenza infection given the PHSMs against COVID-19 in Hong Kong. However, this may imply low level of herd immunity against influenza infection in the community and the possibility of the surge in influenza activity after PHSMs relaxation cannot be ruled out.