Epidemiology of respiratory infections and their association with air pollution and meteorological factors in China, 2004-2018

Background Long-term epidemiological trends of respiratory infectious diseases (RIDs) and their associations with meteorological factors and air pollutants remain understudied in recent 10 years in China. This study investigates temporal patterns of seven major RIDs during 2004–2018 and quantifies their relationships with meteorological variables and ambient pollutants, with the aim of informing evidence-based health policy and environmental interventions. Methods Data on seven RIDs were collected from the National Notifiable Infectious Disease Surveillance System, meteorological and air pollutants data were obtained from the meteorological monitoring stations and national air quality monitoring stations, respectively. Descriptive analyses were used to present trends, and joinpoint regression models were used to examine changes in incidence and mortality for each respiratory infectious disease and to estimate average annual percentage changes. A Distributed Lag Non-Linear Model (DLNM) with relative risk was applied to analyze the impact of meteorological conditions and air pollutants on RIDs. We also applied a time-series decomposition approach based on locally weighted regression to present the seasonality of seven RIDs. Results A total of 23,444,640 cases and 45,291 deaths caused by seven respiratory infectious diseases were recorded in China, and the national mean age-standardized incidence and mortality were 115.87/100,000 and 0.23/100,000, respectively; the change of incidence and mortality differed by age groups. sulfur dioxide (SO 2 ) and particulate matter <10 μm (PM 10 ) in air pollutants and relative humidity and sunshine hours in climatic factors had significant effects on most respiratory diseases in this study. Additionally, meteorological factors had a stronger impact on RIDs with an acute and short-term lag effect compared with air pollutants. Conclusions The prevention and control strategies for RIDs need to be formulated based on their own characteristics and shift from a single biomedical intervention model to an integrated approach encompassing medicine, environmental science, climate science, and socioeconomics.