JMIR Publications

ABSTRACT

Background:

The COVID-19 pandemic significantly altered global influenza transmission patterns, highlighting the need to reassess environmental influences.

Objective:

This study explores the relationship between climatic variables and the dynamics of transmission in Northern China, focus on seasonal influenza subtypes (A/H1N1, A/H3N2, and B/Victoria) before and after the pandemic.

Methods:

Utilizing 13 years of influenza surveillance and meteorological data, we employed Extreme Gradient Boosting (XGBoost) enhanced by SHapley Additive exPlanations (SHAP), and Generalized Additive Models (GAMs) to analyze how climate variables affect influenza positivity rates.

Results:

Results indicate notable post-pandemic shifts in climatic drivers. Absolute humidity (AH) consistently influenced influenza A subtypes, contributing over 32% to positivity rates. Visibility emerged as a critical factor post-pandemic, rising from 4th to 2nd in importance. GAM analysis revealed key thresholds: AH levels below 4 g/m³ correlating with increased A/H3N2 positivity rate, and a shift from increased to decreased risk was observed at a threshold of 4.5 g/m³ (95% CI, 4.0-4.8) for A/H1N1 after pandemic. Additionally, relative humidity (RH) showed a W-shaped response for A/H3N2, with thresholds ranging from 45.1% to 69.7% pre-epidemic and 48.7% to 70.6% post-epidemic, beyond which the influence on the outcomes turned positive. In contrast, visibility demonstrated opposing effects for A/H3N2 and B/Victoria subtypes post-epidemic; visibility value below 17.1 km was associated with a decrease in A/H3N2 infections, whereas below 18.3 km correlated with an increase in B/Victoria infections.

Conclusions:

These findings highlight pandemic-driven shifts in climatic drivers and identify subtype-specific thresholds, offering insights for tailoring public health interventions under evolving conditions.