Centre for Outcomes Research and Effectiveness (CORE), Research Department of Clinical, Educational, and Health Psychology, UCL, London, UK.
Centre for Urban Mental Health, University of Amsterdam, Amsterdam, The Netherlands.
Sci Rep. 2022 Jun 27;12(1):10881. doi: 10.1038/s41598-022-14901-8.
Psychotherapy is an effective treatment for many common mental health problems, but the mechanisms of action and processes of change are unclear, perhaps driven by the focus on a single diagnosis which does not reflect the heterogeneous symptom experiences of many patients. The objective of this study was to better understand therapeutic change, by illustrating how symptoms evolve and interact during psychotherapy. Data from 113,608 patients from psychological therapy services who completed depression and anxiety symptom measures across three to six therapy sessions were analysed. A panel graphical vector-autoregression model was estimated in a model development sample (N = 68,165) and generalizability was tested in a confirmatory model, fitted to a separate (hold-out) sample of patients (N = 45,443). The model displayed an excellent fit and replicated in the confirmatory holdout sample. First, we found that nearly all symptoms were statistically related to each other (i.e. dense connectivity), indicating that no one symptom or association drives change. Second, the structure of symptom interrelations which emerged did not change across sessions. These findings provide a dynamic view of the process of symptom change during psychotherapy and give rise to several causal hypotheses relating to structure, mechanism, and process.
心理治疗是许多常见心理健康问题的有效治疗方法，但作用机制和变化过程尚不清楚，这可能是由于专注于单一诊断，而单一诊断并不能反映许多患者的异质症状体验。本研究的目的是通过说明症状在心理治疗过程中的演变和相互作用，更好地了解治疗变化。从完成抑郁和焦虑症状测量的心理治疗服务的 113,608 名患者中分析了三到六次治疗疗程的数据。在模型开发样本（N=68,165）中估计了面板图形向量自回归模型，并在一个单独的（保留）患者样本（N=45,443）中进行了验证模型的测试。该模型表现出极好的拟合度，并在验证保留样本中得到了复制。首先，我们发现几乎所有的症状都在统计学上相互关联（即密集连接），这表明没有一个症状或关联会驱动变化。其次，出现的症状相互关系的结构在整个疗程中并没有改变。这些发现提供了心理治疗过程中症状变化的动态视图，并提出了几个与结构、机制和过程相关的因果假设。
