特种作战军人伞降训练睡眠障碍风险预测模型的构建与验证  

Development and validation of a risk prediction model for sleep disorder in special operations forces during parachute training

作　　者：余淼 黄美胜 丁再雄 陈郁[3] 蒋涛 YU Miao;HUANG Meisheng;DING Zaixiong;CHEN Yu;JIANG Tao(Department of Orthopedics,Second Affiliated Hospital,Army Medical University(Third Military Medical University),Chongqing,400037;Department of Combat Services,Troop 31636 of PLA Army,Southern Theater Command,Kunming,Yunnan Province,650000;Department of Military Medical Geography,Army Medical Service Training Base,Army Medical University(Third Military Medical University),Chongqing,400038,China)

机构地区：[1]陆军军医大学(第三军医大学)第二附属医院骨科,重庆400037 [2]南部战区陆军第31636部队战勤科,昆明650000 [3]陆军军医大学(第三军医大学)陆军卫勤训练基地军事医学地理学教研室,重庆400038

出　　处：《陆军军医大学学报》2024年第10期1068-1074,共7页Journal of Army Medical University

摘　　要：目的调查分析陆军某特种作战部队军人伞降训练期间发生睡眠障碍的影响因素,构建并验证预测模型。方法抽取2022年10-11月有伞降作业的特种作战军人349名作为研究对象,采用自制问卷、焦虑自评量表(Self-Rating Anxiety Scale,SAS)和匹兹堡睡眠质量指数(Pittsburgh Sleep Quality Index,PSQI)量表对其睡眠质量测评后记录基线资料,根据PSQI分值分为睡眠障碍组和非睡眠障碍组。按7∶3比例分为训练集和验证集。基于训练集,经单因素分析和多因素Logistic回归分析筛选出伞降训练军人睡眠障碍风险的影响因素,R语言绘制列线图,受试者工作特征曲线(recevier operating characteristic curve,ROC)评价模型区分度,校准曲线和Hosmer-Lemeshow拟合优度检验评价模型准确度,临床决策曲线DCA评价模型的有效性。利用验证集进行内部验证。结果共有325名特种作战伞降训练军人的数据有效,均为男性。PSQI总分为(4.26±3.30)分,结果提示62名(19.1%)存在睡眠障碍,睡眠障碍组中轻度睡眠障碍46名(14.1%),中度睡眠障碍14名(4.3%),重度睡眠障碍2名(0.6%)。训练集228名,单因素分析和多因素Logistic回归分析结果显示,跳伞次数(OR:0.390,95%CI:0.185~0.811)、吸烟(OR:2.980,95%CI:1.352~7.028)、焦虑状态(OR:3.280,95%CI:1.434~7.570)和慢性疼痛(OR:4.090,95%CI:1.952~8.690)是伞降训练军人睡眠障碍的独立影响因素。构建伞降训练军人睡眠障碍风险预测模型,绘制ROC曲线并计算曲线下面积(area under curve,AUC)为0.778,最佳阈值为0.163,特异度为63.6%,灵敏度为84.1%,区分度良好。校准曲线和Hosmer-Lemeshow拟合优度检验(χ^(2)=8.789,P=0.456)显示模型准确度良好,临床决策曲线提示该预测模型具有较好的临床适用性。验证集97名内部验证该模型有效。结论特种作战伞降训练军人存在不同程度的睡眠问题。跳伞次数、吸烟、焦虑状态和慢性疼痛会加重睡眠障碍。Objective To investigate the influencing factors for sleep disorder during parachute training in special operations forces from a unit of the army,and to develop and validate a prediction model.Methods A total of 349 special operations officers and soldiers undergoing parachute training from October to November 2022 were recruited as research objects.Self-made questionnaires,Self-Rating Anxiety Scale(SAS)and Pittsburgh Sleep Quality Index(PSQI)Scale were used to collect their baseline data and assess sleep quality.According to their PSQI score,the participants were divided into sleep disorder group and non-sleep disorder group,and then assigned into a training set(n=228)and a validation set(n=97)in a ratio of 7∶3.Based on the training set,univariate analysis and multivariate logistic regression analysis were used to screen out the factors influencing the risk of sleep disorders.Then R language was employed to draw a nomogram model,and receiver operating characteristic(ROC)curve was plotted to evaluate the discrimination of the model.Calibration curve analysis and Hosmer-Lemeshow goodness-of-fit test were applied to evaluate the accuracy of the model,and decision curve analysis(DCA)was conducted to evaluate the validity of the model.The validation set was used for internal validation.Results There were totally 325 participants with valid data,and all of them were male.They had a total PSQI score of 4.26±3.30,and 62 of them(19.1%)had sleep disorders,including 46(14.1%)of mild,14(4.3%)of moderate,and 2(0.6%)of severe sleep disorders.Based on the 228 participants in the training set,the results of univariate and multivariate logistic regression analyses showed the number of parachute jumping(OR=0.390,95%CI:0.185~0.811),smoking(OR=2.980,95%CI:1.352~7.028),anxiety state(OR=3.280,95%CI:1.434~7.570),and chronic pain(OR=4.090,95%CI:1.952~8.690)were independent influencing factors of sleep disorder in these parachute officers and soldiers.A risk prediction model for sleep disorder was constructed,and the area under the

关 键 词：伞降训练 睡眠障碍 预测模型 列线图 

分 类 号：E274.2[军事—军事理论] R338.63[医药卫生—人体生理学] R821.3[医药卫生—基础医学]