Abstract
BACKGROUND: Previous studies have found that meteorological conditions are associated with an increased risk of viral hepatitis. However, other environmental factors, such as greenhouse gases and air pollutants, that may influence this risk have not yet been identified. MATERIALS AND METHODS: The epidemiological characteristics of major viral hepatitis, including hepatitis B virus (HBV) and hepatitis C virus (HCV), in China were investigated. Quantile-based g-computation (Qgcomp) and Bayesian kernel machine regression (BKMR) were then applied sequentially to explore the single and mixed effects of greenhouse gases and air pollutants, as well as their susceptibility associations with HBV and HCV morbidity and mortality across different age groups. Finally, the gradient boosting regression tree (GBRT) model was used to evaluate the prediction performance for different types of hepatitis. RESULTS: Regarding the mixed effects, environmental factors that showed a positive dose-response relationship with HBV morbidity and mortality were mainly concentrated in patients aged 35-64 years and older. For HCV subtypes, there was a positive dose-response relationship between environmental factors and morbidity and mortality, except for patients aged 0-14 years. When considering single effects, NH(3) (P25-P50) at a low concentration level showed a significant positive association with HBV pathogenesis, while CO (P5-P25) at a low concentration level and non-methane volatile organic compound (NMVOC) at all levels showed a significant negative association with HBV pathogenesis. CO(2)bio (P25-P95) showed significant positive associations with HBV mortality at nearly all concentration levels. The associations with HBV mortality decreased with increasing concentration (maximum association (95%CI) =11.54 (4.49, 18.60)), while NMVOC and CH(4) showed significant negative associations with HBV mortality at higher concentration levels (P25-P75). CO(2) and NMVOC at all concentration levels and NH(3) at a moderate level showed significant positive associations with HCV incidence, while CO at all concentration levels showed significant negative associations with HCV incidence. The results based on environmental factors showed that the best predictive performance was observed for HCV (R(2)Training=0.950, R(2)Test=0.942), followed by HBV mortality. CONCLUSION: We found that most greenhouse gases and air pollutants were associated with increased morbidity and mortality of HBV and HCV in China. These findings may have important implications for the development of effective public health interventions and integrated early warning systems for air pollution and viral hepatitis.