Application of Network Analysis in Evaluating the Efficacy of Antiviral Therapy in People Living with HIV/AIDS

Objective This study extended prior research on AIDS by examining the underlying relationships and their intensity among key determinants, with the aim of identifying the most critical intervention points. Methods A retrospective cohort study was conducted to analyze the factors influencing the effectiveness of antiretroviral therapy (ART) among people living with AIDS in Fucheng District, Mianyang City, Sichuan Province from January 2019 to December 2024. Incorporate statistically significant influencing factors into an undirected network; use metrics such as strength (the total weight of a factor’s connections to other factors) and expected value (a measure of a factor’s potential influence on information propagation) to evaluate the relationships among the influencing factors; and analyze the dependencies among these factors using Bayesian networks. Results A total of 438 patients were included in the study, of whom 28.77% had poor ART outcomes. The results showed that factors including age at diagnosis, marital status, educational level, and route of infection were associated with ART outcomes (P < 0.05), with correlation coefficients of 0.473, 0.265, -0.407, and -0.395, respectively. The analyses showed that age at diagnosis and route of infection were the factors with the strongest associations in the network, with strength values of 1.0 and 1.1, respectively. Marital status and diagnosis age reflected the expected treatment outcomes, with expected values of 0.9 and 0.7, respectively. The Bayesian Information Criterion (BIC) was -2 813.756, indicating that the model fit the data well. Conclusions Age at diagnosis was the most influential factor affecting the efficacy of ART, and various social factors exerted their influence on treatment outcomes through this mediating effect. Raising awareness of screening among high-risk groups could contribute to the success of antiviral therapy.