Pathways for the precise prevention and improvement of mental health among university freshmen: a network analysis and simulated intervention study based on the biopsychosocial model

Objective: Given the high prevalence of mental health problems among university freshmen and the limited explanatory capacity of traditional unidimensional models, this study adopts a biopsychosocial (BPS) framework. Network analysis combined with simulation techniques based on the Node Identify via Recursive Graphs (NIRA) algorithm was applied to explore system-level interactions and potential targets within the network. Methods: A total of 3,116 first-year university students were recruited. The network comprised biological-related functional indicators (e.g., TCM constitution types such as Qi stagnation), psychological symptoms (depression, anxiety, suicide risk), psychological traits (resilience, emotion regulation, insight), and social factors (perceived stress, childhood trauma, and social support). An Ising network model was estimated, and centrality and bridge indices were calculated. Simulation analyses were conducted by manipulating node activation probabilities to examine potential changes in overall network activation. Results: Psychological resilience emerged as the central hub node, while perceived stress acted as the strongest bridge node linking social and psychological domains. Simulation analyses suggested that reductions in stress-related nodes and improvements in Qi stagnation-related indicators were associated with decreases in overall psychological network activation. Conclusion: These findings support the utility of a network-based BPS framework for understanding freshmen's mental health. Psychological resilience and perceived stress may represent important components within the system. However, simulation findings should be interpreted cautiously, as they do not imply causal intervention effects.