基于不良事件管理的医疗纠纷列线图预测模型构建及评价  

Construction and Evaluation of a Nomogram Prediction Model for Medical Disputes Based on Adverse Events Management

作　　者：王州 林海涛[1] 周瑾[1] 唐琴 沈乐冲 王志刚 谢印库 WANG Tongzhou;LIN Haitao;ZHOU Jin(Jiangsu Cancer Hospital(Jiangsu Province Hospital Affiliated to Nanjing Medical University/Jiangsu Cancer Center),Nanjing,Jiangsu,210000,China;不详)

机构地区：[1]江苏省肿瘤医院(南京医科大学附属肿瘤医院/江苏省癌症中心),江苏南京210000

出　　处：《中国卫生质量管理》2025年第4期74-78,共5页Chinese Health Quality Management

基　　金：江苏省老年健康科研项目(编号:LKM2023009);江苏省肿瘤医院卓越管理计划(编号:ZG202206)。

摘　　要：目的基于医疗质量安全不良事件分析医院医疗纠纷的影响因素,构建医疗纠纷风险列线图预测模型并进行评价。方法选取2020年1月—2023年12月江苏省某三甲医院503例不良事件作为研究对象,采用单因素和多因素logistic回归分析医疗纠纷发生的影响因素,并建立医疗纠纷风险的列线图预测模型,采用受试者工作特征曲线和校准曲线评价列线图模型的诊断效能和校准度,采用决策曲线评估其收益性。结果医务人员性别男性、患者年龄段≤40岁、患者职业为农民、患病程度非危重、一~三级不良事件是医疗纠纷的独立危险因素(P均<0.05),列线图预测模型的受试者工作特征曲线下面积为0.888,H-L检验P=0.059,校准曲线内部验证的C-Index=0.897,平均绝对误差为0.021。决策曲线的阈值概率为0.02~0.80时净收益较高。结论基于医疗不良事件管理构建的医疗纠纷风险列线图预测模型具有较好的预测效能和临床适用性,可为医疗质量安全管理提供思路。Objective To analyze the influencing factors of medical disputes in hospitals based on medical quality and safety adverse events,construct a nomogram prediction model for medical disputes risk,and evaluate its performance.Methods A total of 503 adverse events from a tertiary hospital in Jiangsu Province between January 2020 and December 2023 were selected as the study subjects.Univariate and multivariate logistic regression analyses were used to identify the influencing factors of medical disputes.A nomogram prediction model for medical disputes risk was established.The diagnostic performance and calibration of the nomogram model were evaluated using receiver operating characteristic(ROC)curve and calibration curve,while decision curve analysis was employed to assess its utility.Results Male healthcare workers,patient age≤40 years,patient occupation as farmer,non-critical illness severity,and adverse events of levels one to three were independent risk factors for medical disputes(all P<0.05).The area under the ROC curve for the nomogram prediction model was 0.888,with an H-L test P-value of 0.059.The C-Index for internal validation of the calibration curve was 0.897,and the mean absolute error was 0.021.The net benefit of the decision curve was high when the threshold probability ranged from 0.02 to 0.80.Conclusion The nomogram prediction model for medical disputes risk constructed based on medical adverse events management demonstrates good predictive performance and clinical applicability,providing insights for medical quality and safety management.

关 键 词：医疗纠纷 不良事件 列线图 预测模型 

分 类 号：R197.323[医药卫生—卫生事业管理]