A nomogram and heat map based on LASSO‐Cox regression for predicting the risk of early‐stage severe fever with thrombocytopenia syndrome patients developing into critical illness at 7‐day and 14‐day

Hanwen Tong,Jun Wang,Naisheng Zhu,Haopeng Li,Yu Zhai,Binxia Shao,Huiying Li,Peng Xia,Yunfei Jiang,Chenxiao Jiang,Yun Liu
DOI: https://doi.org/10.1002/jmv.29921
IF: 20.693
2024-09-20
Journal of Medical Virology
Abstract:Severe fever with thrombocytopenia syndrome (SFTS) represents an emerging infectious disease characterized by a substantial mortality risk. Early identification of patients is crucial for effective risk assessment and timely interventions. In the present study, least absolute shrinkage and selection operator (LASSO)‐Cox regression analysis was conducted to identify key risk factors associated with progression to critical illness at 7‐day and 14‐day. A nomogram was constructed and subsequently assessed for its predictive accuracy through evaluation and validation processes. The risk stratification of patients was performed using X‐tile software. The performance of this risk stratification system was assessed using the Kaplan–Meier method. Additionally, a heat map was generated to visualize the results of these analyses. A total of 262 SFTS patients were included in this study, and four predictive factors were included in the nomogram, namely viral copies, aspartate aminotransferase (AST) level, C‐reactive protein (CRP), and neurological symptoms. The AUCs for 7‐day and 14‐day were 0.802 [95% confidence interval (CI): 0.707–0.897] and 0.859 (95% CI: 0.794–0.925), respectively. The nomogram demonstrated good discrimination among low, moderate, and high‐risk groups. The heat map effectively illustrated the relationships between risk groups and predictive factors, providing valuable insights with high predictive and practical significance.
virology
What problem does this paper attempt to address?