Growth curves of clients’ emotional experience and their association with emotion regulation and symptoms

Abstract

Objective: Emotional experience during psychotherapy is considered a core mechanism of change. Yet the sheer experience itself may not necessarily be beneficial; instead, the trajectories of emotional experience need to be explored as possible predictors of treatment outcomes. This study investigated whether clients’ pre-treatment levels of emotion regulation and symptoms predicted patterns of session-to-session change in emotional experience. We also explored which patterns better predict clients’ improvement in emotion regulation and symptoms from pre- to post treatment. Method: One-hundred and seven clients undergoing psychodynamic psychotherapy completed questionnaires on their symptoms and emotion regulation at pre- and post- treatment. They also reported their level of emotional experience at the end of each session. Results: Pre-treatment symptoms and difficulties in emotion regulation predicted greater instability in emotional experience. Higher mean levels of emotional experience during treatment were associated with an improvement in emotion regulation, and greater stability during treatment was associated with improvement in emotion regulation and symptoms. Conclusions: These findings lend weight to the idea that experiencing emotion in the therapeutic environment has significant implications for clients’ ability to manage their emotions outside the session. However, emotions experienced in an unstable manner within therapy are associated with poorer outcomes.

Clinical and methodological significance of this article: Therapists can benefit from observing the patterns and not only the level of their clients’ emotional experiences. The identification of clients’ difficulties early in treatment may help therapists guide clients through the delicate process of carefully attending to their emotions.

Abstract

Obiettivo: L’esperienza emotiva durante la psicoterapia è considerata un meccanismo centrale del cambiamento. Tuttavia, la pura e semplice esperienza stessa può di per sé non essere necessariamente benefica; invece, le traiettorie dell’esperienza emotiva devono essere esplorate come possibili predittori degli esiti del trattamento. Il presente studio ha indagato se i livelli di regolazione emotiva e i sintomi dei pazienti al pre-trattamento predicessero i pattern di cambiamento sessione dopo sessione nell’esperienza emotiva. Abbiamo inoltre esplorato quali pattern predicessero meglio il miglioramento dei pazienti a livello di regolazione emotiva e di sintomi dal pre- al post- trattamento. Metodo: 107 pazienti che hanno svolto una psicoterapia psicodinamica hanno completato i questionari sui loro sintomi e la loro regolazione emotiva al pre- e post-trattamento. Essi hanno inoltre riportato i loro livelli di esperienza emotiva alla fine di ogni seduta.

Risultati: sintomi e difficoltà nella regolazione emotiva al pre-trattamento predicevano una maggiore instabilità nell’esperienza emotiva. Livelli di esperienza emotiva con medie più alte durante il trattamento erano associati con un miglioramento nella regolazione emotiva, e una maggiore stabilità era associata con un miglioramento nella regolazione emotiva e nei sintomi.

Conclusioni: questi risultati fornisce sostegno all’idea che fare esperienza emotiva in un contesto terapeutico abbia significative implicazioni per l’abilità dei pazienti di gestire le proprie emozioni al di fuori della seduta. Tuttavia, le emozioni sperimentate in un modo instabile all’interno della terapia sono associate con esiti peggiori.

Significato clinico e metodologico di questo articolo: I terapeuti possono beneficiare dall’osservazione dei pattern e non solo dal livello delle esperienze emotive dei loro pazienti. Identificare le difficoltà dei pazienti prima nel trattamento può aiutare i terapeuti a guidare i pazienti attraverso il delicato processo di prendersi cura con attenzione delle proprie emozioni.

摘要

目標：心理治療期間的情緒經驗被認為是改變的核心機制。然而，單純的經驗本身可能不一定是有 幫助的；相反的，應該進一步探索情緒經驗的變化軌跡是否能作為治療結果的可能預測因子。本研 究調查個案在治療前的情緒調節和症狀程度是否能預測每次療程間的情緒經驗變化模式，我們也探 索哪種模式比較能預測個案在治療前後對情緒調節和症狀的改善。方法：107名接受心理動力取向心 理治療的個案在治療前與治療後完成症狀及情緒調節的問卷調查，他們也回報了每次療程結束時的 情緒經驗程度。結果：治療前症狀和情緒調節困難程度可以預測情緒經驗之不穩定性。治療期間較 高的情緒經驗平均程度與情緒調節的改善有關，且治療期間較高的情緒經驗穩定性與情緒調節和症 狀的改善有關。結論：這些發現凸顯了在治療環境中經驗情緒對於個案在治療外管理情緒的能力具 有重要意涵；然而，在治療中以不穩定的方式經驗情緒則會與較差的治療效果有關。

Resumo

Objetivo: A experiência emocional durante a psicoterapia é considerada um mecanismo central de mudança. No entanto, a pura experiência pode não ser necessariamente benéfica; em vez disso, as trajetórias da experiência emocional precisam ser exploradas como possíveis preditores dos resultados do tratamento. Este estudo investigou se os níveis pré-tratamento de regulação emocional e dos sintomas dos clientes previram padrões de mudança na experiência emocional de sessão para sessão. Nós também exploramos quais padrões melhor predizem a melhora dos clientes na regulação emocional e nos sintomas do pré ao pós-tratamento. Método: cento e sete clientes submetidos à psicoterapia psicodinâmica completaram questionários sobre seus sintomas e regulação de emoções no pré e pós-tratamento. Eles também relataram seu nível de experiência emocional no final de cada sessão. Resultados: Sintomas pré-tratamento e dificuldades na regulação emocional predizem maior instabilidade na experiência emocional. Níveis médios mais elevados de experiência emocional durante o tratamento foram associados a uma melhora da regulação emocional, e maior estabilidade durante o tratamento foi associada à melhora na regulação emocional e sintomas. Conclusões: Esses achados reforçam a ideia de que vivenciar emoção no ambiente terapêutico tem implicações significativas para a capacidade dos clientes de gerenciar suas emoções fora da sessão. Contudo, as emoções vivenciadas de maneira instável na terapia estão associadas a desfechos mais desfavoráveis.

Significância clínica e metodológica deste artigo: Os terapeutas podem se beneficiar da observação dos padrões e não apenas do nível das experiências emocionais de seus clientes. A identificação das dificuldades dos clientes no início do tratamento pode ajudar os terapeutas orientar os clientes através do delicado processo de cuidadosamente atentar para as suas emoções.

Keywords:

• psychodynamic psychotherapy
• emotional experience
• patterns
• emotion regulation

Parole chiave:

• psicoterapia psicodinamica
• esperienza emotive
• pattern
• regolazione emotiva

關鍵詞:

• 心理動力取向心理治療
• 情緒經驗
• 模式
• 情緒調節

Palavras-chave:

• psicoterapia psicodinâmica
• experiência emocional
• padrões
• regulação emocional

Notes

1 Emotional work is usually needed when negative and painful emotions, such as shame or hurt, are experienced (McCullough et al., 2003). Some research efforts have been made to explore the distinct effect of different categories of emotions. For example, Kramer et al. (2014) compared sessions from good vs. poor outcome cases and found that clients with good treatment outcomes experienced at least one moment of grief or hurt during the session. By contrast, it should be noted that the current study does not distinguish between different emotions; instead, it explores the level of the general engagement in the experience of emotions.

2 Affective disorders cluster included the following DSM-IV diagnoses: 296.31, 296.32, 296.63, 300.4, and 296.05.

3 Anxiety disorders cluster included the following DSM-IV diagnoses: 300.01, 300.02, 300.21, 300.22, and 300.23.

4 To differentiate between actual between-session variability and simple scale error, multi-item scales are recommended (see Bolger & Laurenceau, 2013). Unfortunately, this was not possible with the EE-SR. Still, the 50/50 split in variance is encouraging in this regard indication that the EE-SR, which was able to detect reliable between-subject differences, was likely to be sensitive enough to also detect between-session differences.

5 In fact, the correlation between clients’ mean experience level and their individualized intercept parameter (the u_0c) was an almost perfect r = .99.

6 We addressed missing data at post-treatment with a multivariate imputation by chained equations (Azur, Stuart, Frangakis, & Leaf, 2011), as implemented in SAS PROC MI and PROC Analyze. For the multiple imputation step, we identified five measures that were most highly correlated with the predictors, and included them in the imputation model. Ten datasets were generated, each was analyzed using the specified model. The results of this individual analysis were then combined to generate the Type III fixed effect values (Rubin, 1987). The pattern of results from the imputed data was similar to the results obtained using complete-case deletion. Therefore, the results presented in Table III are based on the analysis of all available data.

7 This approach enabled us to remain consistent with the statistical approach employed to examine the pre-treatment characteristics as predictors of emotional experience patterns. We opted not to aggregate the Level-1 variable (e.g., experience rating) into Level-2 variables (e.g., by averaging the experience ratings across treatment) or Level-2 latent estimates (e.g., using Bayesian estimates) to represent emotional experience patterns since both of these approaches would have overlooked the construct reliability by ignoring the variability around each client’s aggregated estimate or the fact that each client provided different amounts of data for creating his/her estimates (see Hoffman, 2007).

8 The current study sought to predict clients’ random error variance (i.e., Level-1 residuals) in emotional experience ratings. We argue that this index can be interpreted as the client’s fluctuation above and beyond the within-person therapy effect (i.e., the client’s linear trend line over the entire course of therapy). The inclusion of predictors of this index (namely, pre-treatment symptoms as well as difficulties in emotion regulation) adds a new perspective to the research on instability. This method has been used extensively outside of psychotherapy studies to evaluate “moderated variation” hypotheses within a variety of intensive measurement designs (e.g., Ferrer & Rast, 2017; Hedeker, Mermelstein, & Demirtas, 2008).

9 To differentiate between actual between-session variability and simple scale error, multi-item scales are recommended (see Bolger & Laurenceau, 2013; Cranford et al., 2006). Unfortunately, this was not possible with the EE-SR. Still, the 50/50 split in variance is very encouraging in this regard. We see it as indication that the EE-SR, which was able to detect reliable between-subject differences, was likely to be sensitive enough to also detect between-session differences.