大学生自杀潜在风险特征的潜在剖面分析  

Potential profile analysis of potential risk characteristics for suicide among college students

作　　者：刘爱楼 鲍学峰 Liu Ailou;Bao Xuefeng(Department of Student Affairs,Hubei Normal University,Huangshi 435002,China;Graduate School,South China Normal University,Guangzhou 510631,China)

机构地区：[1]湖北师范大学教育科学学院,黄石435002 [2]华南师范大学研究生院,广州510631

出　　处：《中华行为医学与脑科学杂志》2023年第8期738-745,共8页Chinese Journal of Behavioral Medicine and Brain Science

基　　金：教育部人文社会科学研究项目(19YJCZH101);湖北省教育科学规划项目(2021GA046);湖北省高等学校哲学社会科学研究项目(21Y136);教育部高校思想政治工作队伍培训研修中心(华南师范大学)2021年度开放课题(SCNUKFYB011)。

摘　　要：目的分析我国大学生自杀潜在风险的类别特征与变化趋势,及各自杀潜在风险亚群的预测因子,为自杀潜在风险的早期识别与精准干预提供科学依据。方法2017—2021年整群抽取湖北地区6所高校共14653名在校本科生,每年新生入学后10月份收集1次数据(共5次)。采用Mplus 8.3和SPSS 26.0软件进行统计分析。采用潜在剖面分析(latent profile analysis,LPA)分析大学生自杀潜在风险的亚组及其特征,构建无序多分类Logistic回归模型分析性别、入学年份、心理健康因子对自杀潜在风险类别的预测效应,以及自杀潜在风险差异是否会随着时间的推移而扩大、缩小或保持的年份效应。结果(1)大学生自杀潜在风险可分为3个亚群,分别是C1低风险组(75.6%)、C2无助高风险组(3.7%)和C3人格高风险组(20.7%);(2)"C2无助高风险组"和"C3人格高风险组"2个高自杀风险组具有明显的性别和年级效应。在性别上,女生进入"C3人格高风险组"的概率显著小于男生。随着入学年级的增加,进入2个高自杀潜在风险组的概率呈现出增加的趋势;(3)人际关系敏感对进入"C3人格高风险组"具有显著正向预测作用(B=0.150,OR=1.162,95%CI=1.140~1.185),偏执对进入"C2无助高风险组"具有显著正向预测(B=0.103,OR=1.109,95%CI=1.051~1.170),而恐怖则对进入"C2无助高风险组"(B=-0.041,OR=0.976,95%CI=0.937~1.016)和"C3人格高风险组"(B=-0.031,OR=0.987,95%CI=0.965~1.010)均有显著负向预测;(4)随着时间推移,"C2无助高风险组"和"C3人格高风险组"2个高自杀风险组的占比均呈上升趋势,且"C3人格高风险组"占比上升趋势更为明显,5年间占比上升10%。结论大学生自杀潜在风险可分为3个亚群,性别、入学年级和心理健康症状中的人际关系敏感、恐怖和偏执对自杀潜在风险不同亚群的影响差异显著。对于高自杀风险组大学生应根据其特征采取有针对性的干预措施,重点关注高自杀风Objective To explore the characteristics and trends of potential suicide risk among college students in China,as well as the predictive factors for the potential suicide risk subgroups,in order to provide scientific basis for early identification and precise intervention of potential suicide risks.Methods From 2017 to 2021,a total of 14653 undergraduate students were selected from 6 universities in Hubei Province through cluster sampling.Data were collected once a year in October after freshmen enrolled(a total of 5 times).Mplus 8.3 and SPSS 26.0 softwares were used for statistics.Latent profile analysis(LPA)was used to analyze the category characteristic of suicide risk in 5 dimensions,and multinomial logistic regression was constructed to analyze the gender and school year effects of suicide risk categories,as well as whether the differences in suicide risk would expand,decrease or maintain over time.Results(1)According to the fitting indexes of the LPA,the potential suicide risk of college students could be divided into three heterogeneous categories,namely C1 low-risk group(75.6%),C2 helpless high-risk group(3.7%)and C3 personality high-risk group(20.7%).(2)The 2 high suicide risk groups(C2 and C3)had significant gender and grade effects.In terms of gender,female students were significantly less likely to be in the"C3 personality high-risk group"than male students.As the enrollment grade increased,the probability of entering the two high suicide risk groups showed an increasing trend.(3)Interpersonal sensitivity was a significant positive predictor of entry into the"C3 personality high-risk group"(B=0.150,OR=1.162,95%CI=1.140-1.185).Paranoid Ideation was a significant positive predictor of entry into the"C2 helpless high-risk group"(B=0.103,OR=1.109,95%CI=1.051-1.170).Phobic anxiety was a significant negative predictor of entry into both the"C2 helpless high-risk group"(B=-0.041,OR=0.976,95%CI=0.937-1.016)and the"C3 personality high-risk group"(B=-0.031,OR=0.987,95%CI=0.965-1.010).(4)Over time,the probability

关 键 词：大学生 自杀潜在风险 变迁趋势 潜在剖面分析 

分 类 号：R749[医药卫生—神经病学与精神病学]