Wolf Erika J, Miller Mark W
National Center for PTSD at VA Boston Healthcare System, Boston, MA, USA.
Boston University Chobanian & Avedisian School of Medicine, Department of Psychiatry, Boston, MA, USA.
J Mood Anxiety Disord. 2025 May 23;11:100130. doi: 10.1016/j.xjmad.2025.100130. eCollection 2025 Sep.
Latent class analysis (LCA) is a "person-centered" analytic method designed to identify subgroups of individuals defined by a common characteristic that distinguishes them from other groups within a larger population. Many recent studies have applied LCA to data from self-report trauma exposure measures in an effort to identify clinically useful and/or nosologically informative trauma history "types." In this article, we provide a non-technical overview of this analytic approach and its application to trauma exposure data. We raise concerns about the use of LCA for identifying trauma exposure types relating to: (a) the application of a person-centered approach to variables that reflect environmental exposures; (b) lack of evidence that use of LCA is more informative than other more straightforward and generalizable methods for quantifying trauma exposure; (c) failure to show meaningful differences in the correlates (e.g., risk factors, outcomes, treatment response) of latent classes; (d) forcing severity-based categories on variables that are dimensional, promotion of small classes, and misinterpretation of fit statistics; and (e) interpretation of changing class definitions over time as individual-level changes. Collectively, these concerns lead us to ask, "what is latent about trauma exposure?" and suggest the need for alternative approaches to quantifying and summarizing trauma exposure.
潜在类别分析（LCA）是一种“以人为主”的分析方法，旨在识别由共同特征定义的个体亚组，这些特征将他们与更大人群中的其他群体区分开来。最近的许多研究已将LCA应用于自我报告创伤暴露测量的数据，以努力识别临床上有用和/或在疾病分类学上有信息价值的创伤史“类型”。在本文中，我们对这种分析方法及其在创伤暴露数据中的应用提供了一个非技术性概述。我们对使用LCA识别创伤暴露类型提出了以下担忧：（a）将以人为主的方法应用于反映环境暴露的变量；（b）缺乏证据表明使用LCA比其他更直接和可推广的量化创伤暴露的方法更具信息价值；（c）未能显示潜在类别的相关因素（如风险因素、结果、治疗反应）存在有意义的差异；（d）将基于严重程度的类别强加于具有维度的变量、推广小类别以及对拟合统计的错误解释；以及（e）将随时间变化的类别定义解释为个体水平的变化。总体而言，这些担忧促使我们提出“创伤暴露的潜在性是什么？”这一问题，并表明需要采用替代方法来量化和总结创伤暴露。
