节点文献

精神科急诊患者自杀风险的预测研究

A study on the prediction of suicide risk in psychiatric emergency patients

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 潘轶竹王刚朱辉尹利李京渊及晓李小强

【Author】 PAN Yi-zhu;WANG Gang;ZHU Hui;YIN Li;LI Jing-yuan;JI Xiao;LI Xiao-qiang;Beijing Anding Hospital,Capital Medical University;

【通讯作者】 王刚;

【机构】 首都医科大学附属北京安定医院

【摘要】 目的:对精神科急诊患者的自杀行为风险进行预测。方法:收集精神科急诊求医患者的一般人口学及临床资料,并进行自杀危险因素评估表评定,以10分为分界值将患者分为安全组(≤10分)及危险组(>10分),比较两组患者的人口学及临床资料,分析影响自杀行为发生的相关因素,构建自杀行为预测模型。结果:12 345例急诊求医的患者中,安全组8 909例,危险组3 436例;单因素Logistic回归分析显示,自杀的危险因素为户籍地、居住地、年龄、男性、民族、无伴侣、独居、无职业或目前未在职/在读、家庭月收入水平低、父母非原生健在、与父母关系不好、有>1个兄弟姐妹、与兄弟姐妹关系不好、于22∶00至次日7∶00就诊、3至5月及9至11月就诊、发病有诱因、病前性格内向、精神疾病家族史阳性、首次发病年龄小、总病程与本次病程短、未曾在精神科医院住院或精神科医院住院1次、有明确精神科诊断、伴躯体疾病、攻击风险评估表≤Ⅰ级、简明精神病评定量表(BPRS)、汉密尔顿焦虑量表(HAMA)及汉密尔顿抑郁量表(HAMD)总分高(P均<0.01);保护因素为受教育年限、发病诱因为家庭事件、Young躁狂评定量表(YMRS)总分高(P均<0.01)。多因素Logistic回归建立自杀风险预测模型,其区分度良好,通过内部验证,且模型校正能力良好(χ~2=14.644,P=0.066)。结论:精神科急诊患者自杀风险预测模型对精神科急诊患者自杀风险有较好的预测能力。

【Abstract】 Objective: To predict the risk of suicidal behavior in psychiatric emergency patients. Method:The general demographic and clinical data of patients in psychiatric emergency department were collected, and the suicide risk factors were evaluated.The patients were divided into safety group(≤10 points) and risk group(>10 points) with 10 points as the cut-off value.The demographic and clinical data of the two groups were compared, the related factors influencing the occurrence of suicide were analyzed, and the prediction model of suicide behavior was constructed. Results:Among 12 345 patients receiving emergency medical treatment, 8 909 were in the safety group and 3 436 in the risk group.Univariate Logistic regression analysis showed that the risk factors of suicide were registered permanent residence, current residence, age, male, nationality, no partner, living alone, no occupation, currently not working and studying, low monthly family income, parents not native living, bad relationship with parents, having more than one sibling, bad relationship with siblings, visiting doctor from 22∶00 to 7∶00 the next day, visiting in March to May and September to November, predisposing factors, introversion before illness, positive family history of mental illness, younger age of first onset, shorter total and current course of disease, never hospitalized in a psychiatric hospital or hospitalized once in a psychiatric hospital, with a clear psychiatric diagnosis, accompanied by physical diseases, aggressive risk assessment table ≤ grade I,high total scores of Brief Psychiatric Rating Scale(BPRS),Hamilton Anxiety Scale(HAMA),Hamilton Depression Scale(HAMD)(all P<0.01).Protective factors were years of education, family events, high total score of Young Mania Rating Scale(YMRS)(all P<0.01).Multivariate Logistic regression was used to establish the suicide risk prediction model.The prediction model had good discrimination and passed the internal validation with good calibration ability(χ~2=14.644,P=0.066). Conclusion:The suicide risk prediction model of psychiatric emergency patients has good prediction ability for the suicide risk of psychiatric emergency patients.

【基金】 首都卫生发展科研专项-青年项目(首发2014-4-2125)
  • 【文献出处】 临床精神医学杂志 ,Journal of Clinical Psychiatry , 编辑部邮箱 ,2021年03期
  • 【分类号】R749
  • 【下载频次】143
节点文献中: 

本文链接的文献网络图示:

本文的引文网络