Mask wearing induces multiple transitions of respiratory infectious disease spreading in metropolitan populations

The outbreak of respiratory infectious diseases in metropolitan areas is often accompanied by the widespread adoption of mask wearing behavior. However, the dynamic feedback mechanism between mask wearing and disease spreading remains insufficiently understood. In this study, we first construct an age-structured metropolitan population using census data and describe its interpersonal contact network using age-specific contact matrices. Subsequently, We propose a coupled spreading dynamics model that accounts for the asymmetric interaction between mask wearing and disease spreading, where mask wearing behavior is influenced by both local and global information. A theoretical analysis framework is developed by extending the Microscopic Markov Chain Approach, and the basic reproduction number, $R_0$, is computed using the next-generation matrix method. Finally, we conduct numerical simulations to explore the coevolution of mask wearing behavior and respiratory disease spreading in metropolitan populations. The introduction of mask wearing induces multiple transitions in the system. Based on the degree of disease responsiveness to mask wearing, we identify three categories of diseases: Mask-sensitive diseases, Mask-resistant diseases, and Mask-evading diseases. The evolution of mask wearing behavior with increasing disease spreading probability exhibits three distinct phases: Non-mask phase, Growth phase, and Decline phase. While mask wearing effectively reduces the steady state infection density and the peak infection density, it simultaneously prolongs the time required to reach these states. Prolonging the duration of mask wearing increases the average mask wearing rate, whereas intensifying public campaigns reduces it. Additionally, mask wearing increases the infection risk among younger populations and within school settings.