目的 探究甲状腺乳头状癌(PTC)患者创伤后应激障碍(PTSD)发生的关键因素,构建风险预测模型。方法 选取2020年1月至2025年6月新疆医科大学第一附属医院收治的1022例PTC患者,按7∶3分为建模组(n=715)和验证组(n=307),建模组根据创伤后应激障碍检查量表-平民版(PCL-C)评分及精神科医师临床确诊分为PTSD组(n=171)、非PTSD组(n=544),差异指标纳入KNN、XGBoost、Logistic回归中评估最优模型,基于最优模型筛选变量纳入CatBoost模型对变量进行重要性排序,并构建列线图预测模型,通过Bootstrap法内部验证。结果 两组年龄、性别、婚姻状况、教育程度、月收入、患病时间、癌症分级、侧颈部淋巴转移对比差异显著(P＜0.05)。较XGBoost、Logistic回归相比,KNN模型性能最优,基于此筛选得到:年龄、婚姻状况、教育程度、月收入、患病时间、癌症分级6项重要因素,CatBoost模型分析因素重要性排序为婚姻状况＞癌症分级＞年龄＞患病时间＞教育程度＞月收入。构建列线图,Hosmer-Lemeshow检验为χ2=6.811,P=0.557,建模组及验证组受试者操作特征曲线(ROC曲线)曲线下面积(AUC)分别为0.778(95%CI=0.733~0.822)、0.698(95%CI=0.611~0.784),校准曲线显示与实际风险一致性良好。结论 PTSD的发生与年龄、婚姻状况、教育程度、月收入、患病时间、癌症分级相关,基于此构建的列线图模型区分度与校准度良好。
Abstract
Objective To explore the key factors contributing to post traumatic stress disorder (PTSD) in patients with papillary thyroid carcinoma (PTC) and to construct a risk prediction model. Methods A total of 1022 PTC patients admitted to the First Affiliated Hospital of Xinjiang Medical University from January 2020 to June 2025 were selected and divided into a modeling group (n=715) and a validation group (n=307) at a ratio of 7∶3. The modeling group was further divided into the PTSD group (n=171) and the non-PTSD group (n=544) based on the civilian version of the PTSD Checklist (PCL-C) score and clinical diagnosis by psychiatrists. The difference indicators were included in KNN, XGBoost, and Logistic regression to evaluate the optimal model. Based on the optimal model, variables were selected and included in the CatBoost model to rank the importance of the variables, and a nomogram prediction model was constructed. Internal validation was conducted using the Bootstrap method. Results There were significant differences in age, gender, marital status, education level, monthly income, duration of illness, cancer grade, and lateral cervical lymph node metastasis between the two groups (P<0.05). Compared with XGBoost and Logistic regression, the KNN model had the best performance. Based on this, the following six important factors were selected: age, marital status, education level, monthly income, duration of illness, and cancer grade. The CatBoost model ranked the importance of the factors as marital status > cancer grade > age > duration of illness > education level > monthly income. A nomogram was constructed, and the Hosmer-Lemeshow test showed χ2=6.811, P=0.557. The area under the receiver operating characteristic (ROC) curve (AUC) for the modeling group and validation group was 0.778 (95%CI=0.733-0.822) and 0.698 (95%CI=0.611-0.784), respectively. The calibration curve showed good consistency with the actual risk. Conclusion The occurrence of PTSD is related to age, marital status, education level, monthly income, duration of illness, and cancer grade. The constructed nomogram model has good discrimination and calibration.
关键词
甲状腺乳头状癌 /
创伤后应激障碍 /
最小绝对收缩和选择算法 /
影响因素 /
受试者操作特征曲线 /
列线图 /
预测模型 /
内部验证
Key words
Papillary thyroid carcinoma /
Post-traumatic stress disorder /
Machine learning algorithm /
Influencing factors /
Receiver operating characteristic curve /
Nomogram /
Prediction model /
Internal validation
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] MOHANTY A,AFKHAMI M,RRYES A,et al. Exploring markers of immunoresponsiveness in papillary thyroid carcinoma and future treatment strategies[J]. J Immunother Cancer,2024,12(7):e008505.
[2] 李萍,陈文娟,樊玲,等. 基于增强CT联合超声的列线图模型预测甲状腺乳头状癌中央区淋巴结转移的价值[J]. 中国耳鼻咽喉头颈外科,2025,32(9):595-599.
[3] GAO X,YANG Y,WANG Y,et al. Efficacy and safety of ultrasound-guided radiofrequency,microwave and laser ablation for the treatment of T1N0M0 papillary thyroid carcinoma on a large scale:a systematic review and meta-analysis[J]. Int J Hyperthermia,2023 ,40(1):2244713.
[4] 胡荣毅,屈倩倩,杨尧禹,等. 甲状腺乳头状癌颈淋巴结转移高危因素及预测模型构建研究进展[J]. 国际耳鼻咽喉头颈外科杂志,2025,49(1):36-40.
[5] 樊金芳,周伟,王龚. 基于甲状腺乳头状癌颈侧组淋巴结的超声特征及弹性成像列线图预测淋巴结良恶性的研究[J]. 中国超声医学杂志,2025,41(2):129-132.
[6] ANDERSON D,JONESONES V. Psychological interventions for cancer-related post-traumatic stress disorder:narrative review[J]. BJPsych Bull,2024,48(2):100-109.
[7] LOW C E,TAN S Y P,LOH A,et al. Post-traumatic stress disorder and symptoms in paediatric cancer survivors and their family nucleus:systematic review,meta-analysis and meta-regression[J]. BJPsych Open,2024,10(6):e207.
[8] 王海凤,方继红,王利敏,等. 肿瘤化疗患者照顾者经济毒性的潜在剖面分析及与创伤后应激障碍的关系[J]. 中华现代护理杂志,2025,31(6):771-778.
[9] 高梦琪,王文超,伍新春. 共情对住院肿瘤患者创伤后应激障碍和创伤后成长的影响:肿瘤复发恐惧的中介作用[J]. 中国临床心理学杂志,2022,30(6):1403-1407.
[10] 中华医学会内分泌学分会,中华医学会外科学分会甲状腺及代谢外科学组,中国抗癌协会头颈肿瘤专业委员会,等. 甲状腺结节和分化型甲状腺癌诊治指南(第二版)[J]. 中华内分泌代谢杂志,2023,39(3):181-226.
[11] 杨晓云,杨宏爱,刘启贵,等. 创伤后应激检查量表平民版的效度、信度及影响因素的研究[J]. 中国健康心理学杂志,2007,15(1):6-9.
[12] 罗澍,刘平,毕亚杰,等. 中国心身疾病患者中DSM-5与ICD-11关于创伤后应激障碍的比较[J]. 中国临床心理学杂志,2023,31(6):1438-1441.
[13] 中国高血压防治指南修订委员会,高血压联盟(中国),中国医疗保健国际交流促进会高血压分会,等. 中国高血压防治指南(2024年修订版)[J]. 中华高血压杂志,2024,32(7):603-700.
[14] 中华医学会糖尿病学分会. 中国糖尿病防治指南(2024版)[J]. 中华糖尿病杂志,2025,17(1):16-139.
[15] 赵志茹,刘焕,高娜,等. 甲状腺切除术患者创伤后应激障碍发生风险及影响因素研究[J]. 临床精神医学杂志,2024,34(6):465-468.
[16] CAPALDI J M,SHABANIAN J,FINSTER L B,et al. Post-traumatic stress symptoms,post-traumatic stress disorder,and post-traumatic growth among cancer survivors:a systematic scoping review of interventions[J]. Health Psychol Rev,2024,18(1):41-74.
[17] 张辰,张明洁,王重汉,等. 肿瘤患者共病抑郁障碍与创伤后应激障碍的研究进展[J]. 精神医学杂志,2024,37(4):439-443.
[18] 王玉琛,苏宇玉,杨军. 眼外伤患者发生创伤后应激障碍的危险因素[J]. 国际精神病学杂志,2025,52(1):283-286.
[19] 王莉莎,钱国安,方群英. 乳腺癌患者术后创伤后应激障碍危险因素分析[J]. 中国妇幼保健,2021,36(12):2841-2843.
[20] 郑韦,杨荣,赵海艳. 宫颈癌患者创伤后应激障碍现状调查及影响因素[J]. 国际精神病学杂志,2025,52(3):890-893.
[21] 唐艳,张锦笑. 卵巢癌患者术后创伤后应激障碍有关因素调查[J]. 中国妇产科临床杂志,2024,25(5):460-462.
[22] 周芷晴,罗淑华,郑玲玲,等. 甲状腺癌患者创伤后应激障碍发生现状及其影响因素分析[J]. 中华现代护理杂志,2025,31(7):914-919.
[23] 刘佳晨,涂发妹,熊小艳,等. 乳腺癌患者创伤后应激障碍危险因素的Meta分析[J]. 护理管理杂志,2023,23(4):315-320.
[24] 汪依佳,潘爱红,刘芮,等. ICU患者家属创伤后应激障碍影响因素的Meta分析[J]. 中华急危重症护理杂志,2025,6(8):938-944.
[25] 郭迎坤,袁苗,张嘉慧,等. 基于前瞻性队列研究的产妇产后创伤后应激障碍风险预测模型的构建及验证[J]. 徐州医科大学学报,2025,45(6):404-411.
[26] 孙燕华,赵静,胡艳梅,等. 多次体外受精-胚胎移植失败女性创伤后应激障碍轨迹及影响因素分析[J]. 中国实用护理杂志,2024,40(2):129-136.
[27] 杜小梅,康凤英,宋秋香,等. 自我管理效能在乳腺癌术后病人疾病进展恐惧与创伤后应激障碍间的中介效应[J]. 护理研究,2022,36(7):1287-1292.
[28] SABA T, REHMAN A, SHAHZAD M N, et al. Machine learning for post-traumatic stress disorder identification utilizing resting-state functional magnetic resonance imaging[J]. Microsc Res Tech,2022,85(6):2083-2094.
[29] WHYTE-NESFIELD M M, TRUJILLO RIVERA E A, KAPLAN D, et al. Predicting parental post-traumatic stress symptoms following their child's stay in a pediatric intensive care unit, prior to discharge[J]. J Intensive Care Med,2025,40(4):396-403.
[30] SIMIE D, AZALE T, GASHAW F, et al. Post-traumatic stress disorder symptoms among youths in war-affected areas of Northeast Ethiopia[J]. BMC Psychiatry,2025,25(1):871.
PDF(2385 KB)
