目的 探讨疾病诊断相关分组（Diagnosis Related Group,DRG）中低风险组患者死亡的独立危险因素,利用径向基神经网络模型构建风险预测模型,以期降低DRG低风险组患者死亡率,提高院内患者医疗安全质量。方法 选取上海市某三甲医院2023年1—8月50344条病案首页数据,根据出院情况分为治愈组、未愈组和死亡组,通过单因素分析筛选出重要风险因素,作为径向基神经网络的分析变量,构建风险预测模型。采用曲线下面积（Area Under Curve,AUC）及模型预测准确度、敏感度及特异性等评估模型的预测效能。结果 径向基神经网络模型整体预测准确度为98.69%。其中,患者出院情况为治愈的AUC=0.829,95%CI：0.826~0.832,约登指数最大值为0.5823,敏感度为76.29%,特异性为82.06%。患者出院情况未愈的AUC=0.825,95%CI：0.822~0.828,约登指数最大值为0.5779,敏感度为76.84%,特异性为80.95%。患者出院情况死亡的AUC=0.600,95%CI：0.596~0.605,约登指数最大值为0.2009,敏感度为44.99%,特异性为75.10%。结论 对于DRG低风险组患者风险客观化预测中,径向基神经网络模型中治愈模型的预测性能较优。输血反应、患者年龄、住院天数、住院总费用为影响DRG低风险组出院情况的重要独立危险因素,本研究的模型可以为DRG低风险组患者出院情况恶化和干预提供理论依据。
Abstract
ObjectiveTo investigate the independent risk factors of death in the low-risk group of diagnosis related group (DRG), and construct a risk prediction model by using the radial basis neural network model, in order to reduce the death rate of patients in the low-risk group of DRG and improve the quality of medical safety of patients in hospital. Methods A total of 50344 pieces of home page data of medical records in a grade Ⅲ-A hospital in Shanghai from January to August 2023 were selected and divided into cured group, non-cured group and death group according to discharge conditions. Important risk factors were screened out through single factor analysis and used as analysis variables of radial basis neural network to build a risk prediction model. The area under curve (AUC) and the accuracy, sensitivity and specificity of the model were used to evaluate the prediction efficiency of the model. Results The overall prediction accuracy of radial basis function neural network model was 98.69%. Among them, for cured group, the AUC 0.829, 95%CI: 0.826-0.832, the maximum Jorden index was 0.5823, the sensitivity was 76.29%, and the specificity was 82.06%. For non-cured group, the AUC was 0.825, 95%CI: 0.822-0.828, the maximum Jorden index was 0.5779, the sensitivity was 76.84%, and the specificity was 80.95%. For death group, the AUC was 0.600, 95%CI: 0.596-0.605, the maximum Jorden index was 0.2009, the sensitivity was 44.99%, and the specificity was 75.10%. Conclusion For the risk objective prediction of DRG low-risk patients, the radial basis function neural network model of the cure model has better predictive performance. Blood transfusion response, patient age, length of stay and total hospitalization cost are important independent risk factors affecting the discharge status of DRG low-risk group. The model of this study can provide theoretical basis for the deterioration of discharge status and intervention of DRG low-risk group.
关键词
径向基神经网络；疾病诊断相关分组；低风险死亡组；风险预测模型；独立危险因素；准确度；敏感度；特异性
{{custom_keyword}} /
Key words
radial basis function neural network (RBFNN); diagnosis related group (DRG); low-risk death group; risk prediction model; independent risk factors; accuracy; sensitivity; specificity
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 国家卫生健康委. 关于印发《三级医院评审标准(2022年版)》及其实施细则的通知[EB/OL]. (2022-12-06)[2024-10-28]. http://www.nhc.gov.cn/yzygj/s3585/202212/cf89d8a82a68421c bb9953ec610fb861.shtml.
[2] 谢友勉, 袁方, 许学军, 等. 基于DRG的低风险死亡住院病案 首页分析与整改措施[J]. 中国病案, 2021, 22(4): 12-14.
Xie YM, Yuan F, Xu XJ, et al. Analysis and rectification measures on homepages of low risk death medical records based on DRGs[J]. Chin Med Rec, 2021, 22(4): 12-14.
[3] 许轲, 杨剑, 邱亭林. 基于DRG的肿瘤专科医院死亡病例分析[J]. 中国医院管理, 2023, 43(1): 41-44.
Xu K, Yang J, Qiu TL. Analysis of death cases in oncology hospitals based on DRG[J]. Chin Hosp Manag, 2023, 43(1): 41-44.
[4] 任莹莹, 李田英, 张挪富, 等. 基于DRG的某医院呼吸内科医 疗服务绩效分析[J]. 中国医院管理, 2023, 43(8): 39-41.
Ren YY, Li TY, Zhang NF, et al. Performance analysis of respiratory medical service in a hospital based on DRG[J]. Chin Hosp Manag, 2023, 43(8): 39-41.
[5] 齐荣君, 刘玲, 尹英杰. 基于DRG相关指标的公立医院精细化 管理[J]. 中国病案, 2023, 24(7): 35-39.
Qi RJ, Liu L, Yin YJ. Fine management of public hospitals based on DRG-related indicators[J]. Chin Med Rec, 2023, 24(7): 35-39.
[6] Ramlawi B, Deeb GM, Yakubov SJ, et al. Mechanisms of death in low-risk patients after transcatheter or surgical aortic valve replacement[J]. Cardiovasc Revasc Med, 2022, 42: 1-5.
[7] 付宏煜, 朱志宸, 冯立, 等. 711名儿童患者输血反应回顾性分 析[J]. 中国输血杂志, 2024, 37(6): 678-683.
Fu HY, Zhu ZC, Feng L, et al. Transfusion reactions in 711 pediatric patients:a retrospective analysis[J]. Chin J Blood Transf, 2024, 37(6): 678-683.
[8] 向东, 范亮峰, 刘曦, 等. 免疫性溶血性输血反应100例分析[J]. 临床输血与检验, 2022, 24(1): 11-15.
Xiang D, Fan LF, Liu X, et al. Analysis of 100 cases of immune hemolytic transfusion reaction[J]. J Clin Transf Lab Med, 2022, 24(1): 11-15.
[9] 单明华. 226例输血反应临床分析[J]. 北京医学, 2014(11): 966-967.
[10] 龚秀全, 龚晨曦. 年龄—时期—队列视角下老年人死亡质量 差异研究[J]. 人口与经济, 2023(2): 79-93.
Gong XQ, Gong CX. Research on the difference in quality of death of the Chinese elderly from perspective of age-periodcohort[J]. Popul Econ, 2023(2): 79-93.
[11] 孙小会, 温星来, 陈美玲, 等. 低风险病种死亡原因的影响因 素分析[J]. 中国病案, 2022, 23(4): 48-51.
Sun XH, Wen XL, Chen ML, et al. Analysis of influencing factors on death causes of low-risk diseases[J]. Chin Med Rec, 2022, 23(4): 48-51.
[12] 严浚略. 慢性心力衰竭老年患者住院天数增加的风险模型 构建与验证[D]. 广州: 广州中医药大学, 2022.
Yan JL. A risk model for increased hospitalization days in elderly patients with chronic heart failure: construction and validation[D]. Guangzhou: Guangzhou University of Chinese Medicine, 2022.
[13] 田欣怡. 急性缺血性脑卒中患者住院期间死亡风险预测模 型构建与验证[D]. 衡阳: 南华大学, 2023.
Tian XY. Construction and validation of death risk prediction model for patients with acute ischemicstroke during hospitalization[D]. Hengyang: University of South China, 2023.
[14] 柏星驰, 黎艳娜, 杜芳, 等. 深圳市住院死亡患者医疗费用负 担的影响因素研究——基于病案首页数据[J]. 中国卫生政策 研究, 2023, 16(4): 51-58.
Bai XC, Li YN, Du F, et al. A study on the influencing factors of the medical expense burden of hospitalized deaths in Shenzhen based on homepage data of medical records[J]. Chin J Health Policy, 2023, 16(4): 51-58.
[15] 游颖娟, 熊娟, 刘樱. 单病种付费下阑尾炎患者费用分析[J]. 中国病案, 2022, 23(10): 55-58.
You YJ, Xiong J, Liu Y. Cost analysis of appendicitis patients under single-disease payment[J]. Chin Med Rec, 2022, 23(10): 55-58.
[16] 刘轲, 张冉, 崔志斌, 等. 云模型优化径向基函数神经网络算法 研究[J]. 河南科技大学学报(自然科学版), 2023, 44(5): 49-55.
[17] 刘育强, 魏庆生, 李浩然, 等. 基于径向基函数神经网络的 空间漂浮机械臂装配控制[J]. 哈尔滨工程大学学报, 2023, 44(5): 831-836.
Liu YQ, Wei QS, Li HR, et al. Assembly control of a space floating manipulator based on radial basis function neural network[J]. J Harbin Eng Univ, 2023, 44(5): 831-836.
[18] 赵小惠, 王凯峰, 胡胜, 等. 基于改进径向基神经网络的滚动 轴承故障诊断[J]. 组合机床与自动化加工技术, 2023(3): 66-69.
Zhao XH, Wang KF, Hu S, et al. Improved radial basis neural network rolling bearing fault diagnosis for whale algorithms[J]. Modular Mach Tool Autom Manuf Tech, 2023(3): 66-69.
[19] Tsiklidis EJ, Sinno T, Diamond SL. Predicting risk for trauma patients using static and dynamic information from the MIMIC Ⅲ database[J]. PLoS One, 2022, 17(1): e0262523.
[20] Nieto-Codesido I, Calvo-Alvarez U, Diego C, et al. Risk factors of mortality in hospitalized patients with COVID-19 applying a machine learning algorithm[J]. Open Respir Arch, 2022, 4(2): 100162.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
国家青年基金项目（82203742）；上海交通大学中国医院发展研究院医院管理建设项目（CHDI-2019-B-14）。
{{custom_fund}}
PDF(1714 KB)
京公网安备 11010502038061号