A spatio-temporal causal network for multi-scale analysis of infectious respiratory diseases transmission

Understanding the spatio-temporal transmission characteristics of infectious respiratory diseases is crucial for effective control. However, most existing studies rely on correlation analysis, which obscures the true causal pathways and directionality of infectious respiratory disease transmission, preventing accurate identification of epidemic sources and sinks. To address these challenges, we proposed a novel spatio-temporal causal analysis framework. First, a spatio-temporal causal network is constructed using the Convergent Cross Mapping (CCM) model. This method effectively overcomes the limitations of traditional correlation analysis in identifying spurious correlations and determining causal direction. Subsequently, the weighted k-shell decomposition and Louvain algorithm are applied to analyze the multi-scale structural characteristics of the network, including critical paths, core nodes, and community structures, revealing the multi-scale transmission patterns of the system. We conducted a case study using influenza data from 30 provinces in mainland China from 2010 to 2018. A total of 120 directional transmission pathways were identified, primarily driven by interprovincial population mobility, showing an 83.9 % concordance with the results of the Bayesian phylogenetic analysis. Moreover, provincial importance in transmission was found to be highly correlated with the Hu Huanyong Line. This study provided new insights into the causal relationships and multi-scale structure of infectious disease transmission, offering an important reference for formulating targeted regional prevention and control strategies.