青少年心境障碍患者非自杀性自伤风险预测模型的建立及验证  

Establishment and validation of a risk prediction model for non-suicidal self-injury in adolescents with mood disorders

作　　者：吴莹玉 王文娟 肖媛 汪雅静 陈玉平[3] Wu Yingyu;Wang Wenjuan;Xiao Yuan;Wang Yajing;Chen Yuping(Second Department of Psychology,Qingdao Mental Health Center,Qingdao 266000,China;Department of Gastroenterology,Qingdao Municipal Hospital,Qingdao 266000,China;Department of Nursing,Qingdao Mental Health Center,Qingdao 266000,China)

机构地区：[1]青岛市精神卫生中心心理二科,青岛266000 [2]青岛市市立医院消化内二科,青岛266000 [3]青岛市精神卫生中心护理部,青岛266000

出　　处：《中国实用护理杂志》2025年第10期741-749,共9页Chinese Journal of Practical Nursing

基　　金：青岛市医药卫生科研指导项目(2022-WJZD164)。

摘　　要：目的构建青少年心境障碍患者非自杀性自伤(NSSI)风险预测模型,为高风险患者的识别及干预提供依据。方法采用便利抽样法,选取2019年1月至2023年7月在青岛市精神卫生中心开放式病房住院治疗的724例青少年心境障碍患者作为研究对象进行回顾性分析,将2019年1月至2022年12月住院的641例患者按照7∶3比例随机分为建模组和内部验证组,将2023年1—7月住院的83例患者作为外部验证组。根据二元多因素Logistic回归分析的结果构建青少年心境障碍患者NSSI风险的列线图预测模型,采用曲线下面积(AUC)、灵敏度、特异度等综合评价模型的预测性能。结果724例患者中男179例,女545例,年龄15(14,17)岁;建模组449例,内部验证组192例,外部验证组83例。建模组患者NSSI发生率为32.96%(148/449)。饮食减少(OR=10.980,95%CI 4.462~27.017)、接触被动(OR=4.681,95%CI 1.986~11.031)、17项汉密尔顿抑郁量表评分>17分(OR=12.235,95%CI 4.657~32.141)、汉密尔顿焦虑量表评分>15分(OR=27.888,95%CI 8.700~124.630)、失眠严重指数量表评分>15分(OR=6.357,95%CI 2.257~17.899)和更高的过往自伤水平(OR=1.663,95%CI 1.428~1.935)是青少年心境障碍患者发生NSSI的独立危险因素(均P<0.05),基于此6项因素建立的建模列线AUC为0.973,灵敏度为0.94,特异度为0.89。结论青少年心境障碍患者NSSI风险预测模型具有良好的区分度、准确度和实用性,有助于识别青少年心境障碍患者中的高风险NSSI人群。Objective To construct a risk prediction model for non-suicidal self-injury(NSSI)in adolescents with mood disorders,providing a basis for identifying and intervening in high-risk patients.MethodsA convenience sampling method was used to retrospectively analyze 724 adolescent patients with mood disorders admitted to the open ward of Qingdao Mental Health Center from January 2019 to July 2023.Patients admitted from January 2019 to December 2022(641 cases)were randomly divided into a modeling group and an internal validation group in a 7∶3 ratio,while patients admitted from January to July 2023(83 cases)were used as an external validation group.A binary multivariate logistic regression analysis was used to construct a nomogram prediction model for NSSI risk in adolescents with mood disorders.The predictive performance of the model was evaluated using the area under the curve(AUC),sensitivity,and specificity.ResultsThere were 179 males and 545 females,aged 15(14,17)years old.Among the 724 patients,449 were in the modeling group,192 in the internal validation group,and 83 in the external validation group.The incidence of NSSI in the modeling group was 32.96%(148/449).Reduced food intake(OR=10.980,95%CI 4.462-27.017),passive contact(OR=4.681,95%CI 1.986-11.031),Hamilton Depression Rating Scale-17 score>17(OR=12.235,95%CI 4.657-32.141),Hamilton Anxiety Rating Scale score>15(OR=27.888,95%CI 8.700-124.630),Insomnia Severity Index score>15(OR=6.357,95%CI 2.257-17.899),and higher levels of past self-injury(OR=1.663,95%CI 1.428-1.935)were independent risk factors for NSSI in adolescents with mood disorders(all P<0.05).The AUC of the nomogram model based on these six factors was 0.973,with a sensitivity of 0.94 and a specificity of 0.89.ConclusionsThe risk prediction model for NSSI in adolescents with mood disorders has good discrimination,accuracy,and practicality,and can help identify high-risk NSSI populations among adolescents with mood disorders.

关 键 词：青少年 心境障碍 非自杀性自伤 列线图 预测模型 

分 类 号：R749.4[医药卫生—神经病学与精神病学]