甲状腺癌患者术后焦虑风险预测模型的建立及验证  

Establishment and validation of a predictive model for postoperative anxiety risk in thyroid cancer patients

作　　者：谈爱蓉 许月娥[1] TAN Airong;XU Yue'e(Department of Thyroid and Breast Surgery,the Second People's Hospital of Huai'an,Huai'an,Jiangsu,223002,China)

机构地区：[1]淮安市第二人民医院甲乳外科,江苏淮安223002

出　　处：《中国耳鼻咽喉头颈外科》2024年第7期421-425,共5页Chinese Archives of Otolaryngology-Head and Neck Surgery

摘　　要：目的分析甲状腺癌患者术后焦虑的影响因素,并以此为基础建立风险预测列线图模型。方法将2019年1月~2023年6月于淮安市第二人民医院行手术治疗的457例甲状腺癌患者采用随机数字表法分为建模组(320例)和验证组(137例),比例约为7∶3。根据甲状腺癌患者术后是否出现焦虑情绪分为焦虑组和未焦虑组。收集患者临床随访资料,采用多因素Logistic回归分析甲状腺癌患者术后焦虑的影响因素,采用R软件,rms程序包建立甲状腺癌患者术后焦虑的列线图模型;采用校准曲线、Hosmer-Lemeshow(H-L)检验评估列线图模型拟合程度。采用受试者工作特征(ROC)曲线评估并验证预测模型效能。结果多因素Logistic回归分析,显示家庭月收入≤4000元、婚姻状况为无配偶、工作状况为在职、无健康教育、疾病复发危险性为中高危、手术方式为全切是甲状腺癌患者术后出现焦虑的危险因素(P<0.05)。建模组中,列线图预测甲状腺癌患者术后焦虑的AUC为0.842(95%CI:0.797~0.880),校准曲线贴近于理想曲线,H-L检验为χ^(2)=8.323,P=0.403,校准度良好,该列线图模型预测甲状腺癌患者术后焦虑风险与实际风险基本一致。验证组中,预测甲状腺癌患者术后焦虑的AUC为0.865(95%CI:0.796~0.917);校准曲线与理想曲线较贴近,H-L检验为χ^(2)=5.618,P=0.585,该列线图模型对甲状腺癌患者术后焦虑风险的预测与实际结果一致性较好。结论家庭月收入≤4000元、无配偶、在职工作、无健康教育、疾病复发危险性为中高危、手术方式为全切是甲状腺癌患者术后出现焦虑的危险因素,基于此建立的列线图模型预测区分度较好,可为甲状腺癌患者术后焦虑的早期筛查及预防提供参考。OBJECTIVE To analyze the influencing factors of postoperative anxiety in patients with thyroid cancer(TC),and to establish a risk prediction nomogram model.METHODS From January 2019 to June 2023,457 TC patients who underwent surgical treatment at our hospital were randomly separated into a modeling group(320 cases)and a validation group(137 cases)using a random number table method,with a ratio of approximately 7∶3.TC patients were grouped into anxiety group and non anxiety group based on whether they experience anxiety after surgery.Clinical follow-up data of patients were collected,multivariate Logistic regression was applied to analyze the influencing factors of postoperative anxiety in TC patients,R software and RMS package were applied to establish a nomogram model of postoperative anxiety in TC patients.The calibration curve and Hosmer-Leme show(H-L)test were applied to evaluate the fitting degree of the nomogram model.ROC curve was applied to evaluate and validate the effectiveness of predictive models.RESULTS Multivariate Logistic regression analysis showed that family monthly income≤4000 yuan,marital status as no spouse,on-the-job work,no health education,moderate to high risk of disease recurrence,and surgical method as total thyroidectomy were risk factors for postoperative anxiety in TC patients(P<0.05).In the modeling group,the AUC predicted by the nomogram for postoperative anxiety in TC patients was 0.842(95%CI:0.797-0.880),and the calibration curve was close to the ideal curve,the H-L test wasχ^(2)=8.323,P=0.403,and the calibration was good,the risk of postoperative anxiety in TC patients predicted by the nomogram model was basically consistent with the actual risk.In the validation group,the AUC in predicting postoperative anxiety in TC patients was 0.865(95%CI:0.796-0.917);the calibration curve was relatively close to the ideal curve,H-L test showedχ^(2)=5.618 and P=0.585,the nomogram model has good consistency with predicting postoperative anxiety risk in TC patients with actual results.CO

关 键 词：甲状腺肿瘤 焦虑 列线图 模型 影响因素 

分 类 号：R736.1[医药卫生—肿瘤]