新型冠状病毒肺炎(coronavirus disease 2019，COVID-19)患者的重症化风险预测对优化临床干预和资源配置至关重要。本研究通过回顾性分析2022年12月—2023年1月复旦大学附属华山医院收治的305例确诊COVID-19住院患者的临床数据，探讨与重症/危重症患者相关的危险因素及预测模型效能。本研究收集了患者人口学特征、基础疾病、实验室指标(包括炎症标志物、凝血功能及心脏生物标志物)等资料，并基于《新型冠状病毒感染诊疗方案(试行第十版)》进行临床分型。多因素Logistic回归分析表明，年龄增长、淋巴细胞计数降低、血红蛋白水平下降、C反应蛋白(C-reactive protein, CRP)、白细胞介素-6(interleukin-6, IL-6)、D-二聚体(D-dimer)及脑钠肽(brain natriuretic peptide, BNP)水平升高是患者被分类为重症/危重症的独立危险因素。基于上述指标构建的联合预测模型曲线下面积(area under the curve, AUC)达0.748，显著优于单一指标的预测效能(AUC范围：0.625～0.681)。研究结果提示，综合年龄、炎症反应、凝血功能及心脏损伤标志物，可有效识别COVID-19重症化高危人群，可为早期分层管理、精准治疗和公共卫生决策提供科学依据，并为应对未来类似突发传染病事件积累重要的循证证据。

Abstract：Predicting the risk of progression to severe/critical coronavirus disease 2019(COVID-19) is crucial for optimizing clinical interventions and healthcare resource allocation. This retrospective study analyzed clinical data from 305 hospitalized COVID-19 patients admitted to Huashan Hospital between December 2022 and January 2023, aiming to identify risk factors and evaluate the predictive performance of a multifactorial model for classifying patients into severe/critical categories. Demographic characteristics, comorbidities, and laboratory parameters (including inflammatory markers, coagulation profiles, and cardiac biomarkers) were collected, with disease severity classified according to the Diagnosis and Treatment Protocol for Novel Coronavirus Infection (Trial Version 10). Multivariable logistic regression revealed that advanced age, decreased lymphocyte count, reduced hemoglobin levels, and elevated levels of C-reactive protein (CRP), interleukin-6 (IL-6), D-dimer, and brain natriuretic peptide (BNP) were independent risk factors for severe/critical classification. A combined predictive model incorporating these indicators achieved an area under the curve (AUC) of 0.748, which significantly outperformed individual predictors (AUC range: 0.625-0.681). The findings suggest that integrating age, inflammatory response, coagulation dysfunction, and cardiac injury biomarkers can effectively identify high-risk COVID-19 patients, providing a scientific foundation for early stratified management, precision treatment, and public health policymaking. This study also contributes evidence-based insights for addressing future emerging infectious disease outbreaks.