Quantitative assessment of Public Health and Social Measures implementation and relaxation on influenza transmission during COVID-19 in China: SEIABR and GBDT models.

Abstract

Background: Since 2019, China has implemented Public Health and Social Measures (PHSMs) to manage the coronavirus disease 2019 (COVID-19) outbreak. As the threat from SARS-CoV-2 diminished, these measures were relaxed, leading to increased respiratory infections and strained health care resources by mid-2023.

Methods: The study utilised WHO's FluNet and Oxford's COVID-19 Government Response Tracker to assess how policy shifts have affected influenza. It examined changes in influenza incidence, subtype prevalence, and epidemic cycles over three periods: pre-COVID-19 and pre-PHSMs, during COVID-19 and PHSMs, and post-COVID-19 and post-PHSMs. The SEIABR model estimated the transmission probability () and real-time reproduction number () across these periods, while a gradient boosting decision tree (GBDT) analysed the effects of PHSM indicators on influenza transmission.

Results: Results indicate that before PHSMs, the average incidence was 4.87 per 100 000, with a β-value of (7.95 ± 1.27) × 10^-10 and R_t-value of 1.21 ± 0.16. During PHSMs, incidence dropped to 2.55 per 100 000, and β decreased to (3.17 ± 0.75) × 10^-10 (R_t-value of 0.86 ± 0.20). Post-PHSMs, the incidence surged to 17.00 per 100 000, with β rising to 8.36 × 10^-10 (R_t-value of 2.25). The GBDT model identified testing policies, public information campaigns, and workplace closures as the most impactful PHSM indicators.

Conclusions: PHSMs effectively mitigated the spread of influenza, providing a foundation for future policy development to prevent respiratory diseases.