Peiper Nicholas, Clayton Richard, Wilson Richard, Illback Robert, O'Brien Elizabeth, Kerber Richard, Baumgartner Richard, Hornung Carlton
Department of Epidemiology and Population Health, University of Louisville, Louisville, KY, USA,
Soc Psychiatry Psychiatr Epidemiol. 2015 Jun;50(6):983-94. doi: 10.1007/s00127-015-1017-2. Epub 2015 Feb 5.
The heterogeneity of serious emotional disturbance has been thoroughly documented among adolescents with nationally representative data derived from structured interviews, although use of these interviews may not be feasible within the context of brief and self-administered school surveys. This study seeks to identify distinct subtypes of serious emotional disturbance in a large school-based sample.
A total of 108,736 students fully completed the K6 scale that was included on the 2012 Kentucky Incentives for Prevention Survey. Latent class analysis was used to derive subtypes of serious emotional disturbance among students receiving a positive screen (n = 15,147). To determine significant predictors of class membership, adjusted rate ratios and 95 % confidence intervals were calculated using multinomial logistic regression.
A four-class model was the most parsimonious, with four distinct subtypes emerging that varied by both symptom type and severity: comorbid moderate severity, comorbid high severity, anxious moderate severity, and depressed high severity. Age, gender, race/ethnicity, family structure, substance use, antisocial behavior, role impairments, and peer victimization were significant predictors of class membership, although the magnitude of these effects was stronger for the two high severity groups.
Our results suggest heterogeneity of serious emotional disturbance by both symptom type and severity. Prevention programs may benefit by shifting focus from specific disorders to the core features of serious emotional disturbance, including psychological distress, high comorbidity, and role impairments.
尽管在简短的自我管理学校调查中使用结构化访谈可能不可行，但在具有全国代表性数据的青少年中，严重情绪障碍的异质性已通过结构化访谈得到充分记录。本研究旨在在一个大型的基于学校的样本中识别严重情绪障碍的不同亚型。
共有108,736名学生完整填写了2012年肯塔基州预防激励调查中包含的K6量表。潜在类别分析用于在筛查呈阳性的学生（n = 15,147）中得出严重情绪障碍的亚型。为了确定类别归属的显著预测因素，使用多项逻辑回归计算调整后的率比和95%置信区间。
四类模型最为简约，出现了四种不同的亚型，这些亚型在症状类型和严重程度上各不相同：共病中度严重、共病高度严重、焦虑中度严重和抑郁高度严重。年龄、性别、种族/族裔、家庭结构、物质使用、反社会行为、角色损害和同伴受害是类别归属的显著预测因素，尽管这些影响的程度在两个高度严重组中更强。
我们的结果表明，严重情绪障碍在症状类型和严重程度上存在异质性。预防项目可能会从将重点从特定障碍转移到严重情绪障碍的核心特征中受益，这些核心特征包括心理困扰、高共病率和角色损害。
