Predictors and Moderators of Post-Treatment Depression Severity After Hypnotherapy and Cognitive Behavioral Therapy in an RCT for Mild-to-Moderate Depression · Brockmann AL, Pfeiffer L, Batra A et al.

1. Introduction

In the year 2019, around 280 million people worldwide were diagnosed with a depressive disorder (WHO, 2021). Of the years lived with disability, depressive disorders were second among all diseases and highest for mental health conditions in 2021 (The Lancet Psychiatry, 2024; GBD 2021 Diseases and Injuries Collaborators, 2024). There are various treatments for depressive disorders, including antidepressant medication (ADM) and psychotherapy. Cognitive behavioral therapy (CBT) and interpersonal psychotherapy (IPT) are considered to be the most evidence-based forms of psychotherapy (e.g., Hautzinger, 2008; Luty et al., 2007). Essential elements of CBT are behavioral activation (e.g., Ekers et al., 2014), cognitive therapy and modifying automatic thoughts (Jacobson et al., 1996), in addition to elements like psychoeducation, social skill training, and relapse prevention (J. S. Beck, 2020). At least ten sessions are recommended for CBT for depression to increase its efficacy (Honyashiki et al., 2014). For those with relatively severe depression, especially, a treatment length of 16 sessions improved treatment outcomes compared to those with milder symptoms (Shapiro et al., 1994). When similar effective treatments are available, for example, IPT and CBT (Lemmens et al., 2015), the question arises as to which works better for whom (Cohen & DeRubeis, 2018). Consequently, predictors and moderators for differential treatment success should be identified (Simon & Perlis, 2010; Kazdin, 2007, 2009) to reduce individual suffering, time, and direct and indirect costs. Sociodemographic variables such as age (Fournier et al., 2009), marital status (Barber & Muenz, 1996), gender, and socioeconomic status (Fournier et al., 2009) were found to predict outcome. Regarding disorder-related characteristics such as the duration of the current episode (Blom et al., 2007) and the severity of the episode before the start of treatment (Hamilton & Dobson, 2002), there is evidence for a negative influence on the outcome. The quality of life or expectations regarding the efficacy of a treatment are also related to the outcome (Fournier et al., 2009, Beard et al., 2016); higher expectations were related to a better treatment outcome (Beard et al., 2016; Webb et al., 2013). A high self-efficacy and the belief of a person in their own capacities can have positive effects on the course of the depression (Schindler & Körkel, 1994; Brenninkmeijer et al., 2019). Taking ADM against depression when starting with psychotherapy may also influence outcomes. Several studies and meta-analyses found that a combined treatment of psychotherapy with ADM is superior to stand-alone treatments for moderate-to-severe depression (Cuijpers et al., 2014; Cuijpers et al., 2020; Nemeroff et al., 2003).

1.1. Treatment-Specific Outcome Predictors for CBT

Within the theoretical framework of CBT, cognitive and behavioral variables, for example, a lack of reinforcement (Lewinsohn, 1974), were hypothesized to moderate or mediate the outcome of a treatment for depression. As one of the cognitive variables, rumination, which refers to thinking about negative events and experiencing negative thoughts in loops, can predict higher symptoms after the end of a treatment (Jones et al., 2008; Schmaling et al., 2002; Teismann et al., 2008) and lowers the probability for a remission from depression (Jones et al., 2008). Similarly, dysfunctional attitudes assessed before treatment can predict a worse outcome after treatment (Hamilton & Dobson, 2002).

1.2. Hypnotherapy in the Treatment of Depression

While there is notable evidence for the successful treatment of depression with CBT, hypnotherapy (HT) for depression has not yet been significantly researched within the gold standard of randomized controlled trials (RCTs). Hypnotherapy (HT) covers a broad range of techniques such as directive suggestion, ego-strengthening, Ericksonian approaches, imagery-based work, age progression or regression, resource activation, cognitive hypnotherapy, and others. Previous RCTs compared HT as an add-on to other treatments, for example, CBT. Meta-analyses showed additive effects of HT compared to CBT alone and also for depressive symptoms (Kirsch et al., 1995; Ramondo et al., 2021). However, only a few RCTs were published with samples of patients with a clinically confirmed diagnosis of depression (Alladin & Alibhai, 2007; Butler et al., 2006, Chiu et al., 2018; Fuhr et al., 2021). Only some of the previous RCTs studied HT as a stand-alone treatment (Chiu et al., 2018; Fuhr et al., 2021). The generalizability of RCT results and the usage of those designs in studies with HT can be questioned, for example, because of the challenges involved in manualizing the treatment (McCann & Landes, 2010). Nevertheless, decisions in the health care system as well as the development of treatment guidelines within Europe rely on RCTs and meta-analyses, as stated, for example, by the European Medicines Agency. Treatments lacking empirical evidence within the gold standard approach therefore remain “unseen” and excluded from standardized treatment protocols and recommendations. Even though the number of RCTs published during recent years has been increasing (see, for example, Ramondo et al., 2024; Çınaroğlu et al., 2025), the interpretation of results is insufficient, mainly due to small sample sizes or methodological shortcomings.

1.3. Treatment-Specific Outcome Predictors for Hypnotherapy

Hammond (2005)’s model suggests that different aspects such as cognitive and motivational variables, as well as a dissociative capacity, suggestibility, absorption and aspects of the therapeutic relationship, can influence outcomes of HT. In the literature, hypnotic susceptibility or suggestibility was considered as a main predictor for treatment outcome. Alladin and Alibhai (2007) found that higher hypnotic suggestibility was associated with better treatment outcomes in both CBT and HT patients. However, one very recent study found no moderation effect of hypnotic susceptibility for depression when comparing hypnosis with muscle relaxation (Siewert et al., 2026). Additionally, as suggested by the social cognitive theory for hypnosis put forward by Lynn et al. (2008), efficacy might also be influenced by expectancies, attributions, and motivations. Outcome expectation predicted better treatment outcomes in the study by Ramondo et al. (2024). HT outcome was also influenced by treatment expectancies and treatment credibility in patients with pain (Milling et al., 2007).

1.4. Aims of the Study

The aim of this study was to identify treatment-specific predictors and moderators of post-treatment depression severity after CBT and HT in mild-to-moderate depression in a secondary analysis based on the RCT of Fuhr et al. (2021). Regarding CBT, potential treatment-specific predictors were variables such as behavioral activation level and rumination. Regarding HT, hypnotic susceptibility was selected as a potential predictor or moderator. Since the outcome of both CBT and HT may also be influenced by self-efficacy and treatment expectations, those variables were considered to be global predictors. Other global predictors were sociodemographic, depression-related, and psychological variables such as quality of life. Predictors were identified in three steps. First, based on correlation analyses, potential predictors were identified for the overall sample and for a treatment with CBT and HT separately. Second, linear regression analyses were conducted to select the most prominent predictors. Third, a moderator analysis was conducted to predict treatment-specific moderators.