Evaluating the efficacy of rumination-focused cognitive-behavioral therapy in alleviating depression, negative affect, and rumination among patients with recurrent major depressive disorder: a randomized, multicenter clinical trial

Abstract

Background

Rumination plays a pivotal role in the onset, recurrence, and persistence of residual symptoms in major depressive disorder (MDD), making it a critical therapeutic target in the development of novel interventions. Rumination-Focused Cognitive-Behavioral Therapy (RFCBT), an adapted form of cognitive-behavioral therapy specifically designed to address ruminative thinking patterns, has garnered increasing attention in clinical research. In light of these developments, the present study was conducted to evaluate the efficacy of RFCBT in reducing rumination severity, depressive symptoms, and negative affect among individuals with recurrent MDD.

Methods

This randomized clinical trial enrolled 44 university students diagnosed with MDD. Participants were randomly assigned in equal proportions to either the group-based Rumination-Focused Cognitive-Behavioral Therapy (g-RFCBT) condition or a waitlist (WL) control group. The g-RFCBT received 10 weekly group therapy sessions supplemented by two individual sessions (the initial session and the 12th session), while participants in the WL group received no intervention until the conclusion of the study. Assessments were conducted at four time points; pre-treatment, post-treatment, 2-month follow-up, and 6-month follow-up. Outcome measures included the Beck Depression Inventory-II (BDI-II), the Repetitive Thinking Questionnaire-31 (RTQ-31), and the Negative Affect subscale of the Positive and Negative Affect Schedule (PANAS-NA).

Results

Repeated measures ANOVA revealed statistically significant improvements in the g-RFCBT group compared to the WL group. Within-group analyses indicated that RFCBT led to a 65% reduction in depressive symptoms and a 30% reduction in both rumination and negative affect from baseline to post-treatment. Furthermore, the g-RFCBT exhibited significantly greater reduction in depression, rumination, and negative affect compared to the WL group, with treatment effects sustained through the 6-month follow-up period.

Conclusions

Group-based Rumination-Focused Cognitive-Behavioral Therapy (g-RFCBT) demonstrated significant efficacy in reducing the severity of depressive symptoms, rumination, and negative affect among individuals with major depressive disorder (MDD). Moreover, these therapeutic benefits were maintained over a 6-month follow-up period, suggesting durable treatment effects.

Trial registration

This randomized clinical trial, consisting of an intervention and a waitlist control condition, that was registered in the Iranian Registry of Clinical Trials (IRCT Id: IRCT20181117041676N1) (Registration Date: June 3, 2019).

Graphical Abstract

Background

Major depressive disorder (MDD) is a prevalent and debilitating disorder causing significant clinical distress and impairs the quality of life. Once a second episode occurs, depression often becomes chronic, with the risk of recurrence increasing after each subsequent episode. Following a third episode, the likelihood of relapse rises to approximately 90% [1]. Cognitive-behavioral therapy (CBT) and antidepressant medications are currently the most widely recommended and effective treatments for recurrent major depressive disorder (MDD). However, emerging evidence suggests that the efficacy of antidepressants may be more limited than previously assumed. One study reported no significant differences in mental and physical health outcomes between users and non-users [2]. Furthermore, another study also reported a higher mortality rate in users of these drugs (11.4% vs. 7.0%), suggesting that the effects of these medications may be more complex or potentially associated with adverse outcomes [3]. As a result, CBT may offer distinct advantages that position it as the preferred treatment for MDD. Indeed, evidence demonstrates its superior long-term effectiveness compared to antidepressants [4]. According to cognitive-behavioral models, depression is primarily driven by distorted and repetitive negative cognitive processes. These maladaptive thought patterns result in negatively biased interpretations of internal and external events, perpetuating a dysfunctional cycle of depressive behaviors [5, 6] and sustained negative affect [7].

The rumination-focused model of depression elucidates how targeting rumination can enhance both the understanding and treatment of depression [8]. Rumination is defined as a maladaptive cognitive process characterized by negative, repetitive, and recurrent thoughts focused on oneself, feelings, personal concerns, and distressing experiences [9]. Although numerous studies have established rumination as a key factor in the onset and maintenance of depression [9–11], other findings indicate that elevated levels of rumination are associated with reduced responsiveness to both cognitive-behavioral therapy (CBT) and antidepressant medications [7, 12, 13]. Furthermore, evidence has supported that rumination functions as a transdiagnostic process, contributing to the psychopathology of various disorders, such as generalized anxiety disorder, substance abuse [14], and eating disorder [15]. Rumination in depression persists post-treatment, with meta-analytic evidence identifying it as the most robust residual symptom predicting relapse following cognitive-behavioral therapy [16]. Therefore, targeting rumination—both as a persistent residual symptom linked to relapse and as a transdiagnostic mechanism contributing to the psychopathology of major depressive disorder (MDD)—may improve treatment efficacy and foster sustained long-term remission [17, 18].

Various therapies have been developed to target rumination, including Metacognitive Therapy MCT; [19], Mindfulness-Based Cognitive Therapy MBCT [20], Cognitive Bias Correction [21], and Rumination-Focused Cognitive-Behavioral Therapy (RFCBT) [22]. RFCBT conceptualizes rumination as an habitual avoidance strategy wherein individuals engage in repetitive thinking to suppress or evade negative emotions—a process reinforced through negative reinforcement—which ultimately solidifies rumination as a maladaptive response style [23]. RFCBT differentiates between two modes of thought processing—abstract and concrete —which have distinct consequences [24]. RFCBT links abstract cognitive processing (general, decontextualized, maladaptive) to depressed mood and impaired problem-solving abilities, while concrete processing (specific, contextualized, adaptive) is associated with enhanced problem-solving and reduced impulsivity [25]. RFCBT targets the shift from abstract to concrete processing, alongside functional analysis, behavioral experiments, patients values and relapse prevention as core mechanisms of change [24]. An uncontrolled group study of RFCBT demonstrated significant reductions in depressive symptom severity, rumination, and worry among 10 patients with chronic depression, with sustained improvements at 3-month follow-up [26]. A randomized controlled trial found RFCBT to be more effective than CBT in reducing symptoms of MDD, suggesting it may serve as a viable alternative to group-based CBT [27]. However, it also has been noted that the evidence base for RFCBT remains insufficient, necessitating further research to establish its efficacy across diverse patient populations [27].

Although prior studies have highlighted the potential benefits of targeting rumination, no validated randomized controlled trials have been conducted to examine the efficacy of RFCBT among Persian-speaking populations. To address this gap, the present randomized clinical trial aimed to evaluate both the clinical efficacy of group-based RFCBT in reducing depressive symptoms, rumination, and negative affect, as well as its feasibility and cultural adaptability within a Persian-speaking context. By providing empirical evidence from an understudied population, this study addresses a critical gap in the existing literature on RFCBT.

Method and material

Participants and procedure

During the initial screening phase, 207 university students seeking treatment for depression were contacted via telephone and invited to participate in a group therapy intervention. Of these, 92 individuals who provided informed consent were invited for further eligibility screening, including structured clinical interviews. The inclusion criteria were, a) A clinical diagnosis of recurrent MDD, defined as at least two episodes separated by an interval of two months or more, confirmed by a research psychiatrist; b) A Beck Depression Inventory-II (BDI-II) total score meeting the clinical cut-off (≥ 14), c) ability to attend therapy sessions in person, d) Continued enrollment as a university student during the study period. Exclusion criteria were as follows: a) Current depression secondary to medical conditions, bipolar disorder, history of at least one of psychosis, addictive disorders or substance abuse, personality disorders, b) Prior receipt of CBT or any psychological interventions within the past year (more than five sessions), c) history of using psychotropic medications, hospitalization, and electroconvulsive therapy (ECT) for MDD. Participants who missed more than two consecutive therapy sessions or failed to complete all four assessment phases (pre-treatment, post-treatment, 2-month follow-up, 6-month follow-up) were excluded from the study. A total of 8 (18.2%) participants were excluded: four from the intervention group (2 due to excessive absences, 2 due to incomplete post-treatment data) and four from the WL group (all due to incomplete post- treatment assessments). Ultimately, 36 participants (18 in the intervention group; 18 in the WL) were included in the final analysis. This study, registered in the Iranian Registry of Clinical Trials (IRCT Id: IRCT20181117041676N1), comprised an intervention and a wait-list (WL) group. Participants were recruited from university students seeking treatment for depression at three counseling centers across the city of Zanjan: the counseling center of Zanjan University of Medical Sciences, Zanjan University, and Islamic Azad University-Zanjan Branch. These centers were selected to ensure a diverse sample of students diagnosed with recurrent MDD. A standardized protocol was implemented across all sites, and assessments were administered by trained clinical psychologists to ensure consistency and reliability. Prospective participants were referred to the study assessor at Zanjan University of Medical Sciences. Following the screening phase, individuals who failed to meet the inclusion criteria were excluded, and the remaining participants were randomly assigned to one of two conditions: the intervention group or the Waitlist Control (WL) group. The primary outcome variables in this study were depressive symptoms, rumination, and negative affect, assessed at four standardized timepoints (pre-treatment and post-treatment, 2 and 6-month follow-up) using self-report questionnaires.

Randomization and blinding

Participants were randomly assigned to groups using a 1:1:1:1 allocation ratio, generated via Random Number Generator V 1.3 software. A researcher, independent of the therapeutic team, prepared and executed the randomization sequence to minimize selection bias. Finally, Statistical analyses were performed by a data analyst blinded to group allocation, using anonymized group codes.

Intervention

g-RFCBT is grounded in the principles of CBT, in which rumination is conceptualized as a learned habitual behavior, maintained through negative reinforcement [28]. Central to this approach is the shift from challenging the content of ruminative thoughts to targeting the process of rumination itself, achieved through functional analysis. In functional analysis, patients are taught to identify the antecedents and consequences of the maladaptive style (abstract processing style), then practice repeatedly alternative behaviors to replace with a more adaptive style (concrete processing style). These alternative strategies include contingency plans (e.g.,"if–then"scenarios), visualization exercises, assertiveness training, stress management and adaptive coping styles, relaxation practice, retraining experiences of absorption or flow, and problem-solving (Table 1). The intervention comprised two individual sessions (the initial session and the 12th session) and ten weekly 2-h group sessions. At the end of each session, the worksheets were provided to participants as homework, and they were reviewed at the beginning of the following sessions.

Table 1.

Overview of RFCBT Sessions

Session No	Content of intervention sessions	Assignments
Session 1 (Individual)	Introduction and recognition of repetitive thoughts (especially rumination), the introduction of RFCBT
Session 2	Introducing the group members and an Icebreaker, formulation of rumination as a mental habit, training the self-monitoring, consequences, and functions of rumination	Handout 1. Key Facts about Rumination Handout 2. Avoidance
Session 3	Noticing warning signs, introducing If–Then plans and then practicing if–then plans based on the examples presented in the group	Handout 3. Goal Setting Handout 4. Self-Monitoring Handout 5. Tracking Rumination and Avoidance
Session 4	The idea of changing circumstances to prevent triggering of habit, introducing assertiveness behavior as the opposite of avoidance behavior as part of an if–then plan	Handout 4; Handout 5
Session 5	Introduce different styles of thinking, abstract thinking (passive, general, non-experimental) versus concrete thinking (active, specific, experimental)	Handout 5; Handout 6 Handout 7. Antecedent–Behavior–Consequence (ABC) Form
Session 6	Investigating the different outcomes of"why"thoughts versus"how"thoughts; Training experiential alternatives to rumination by mental imagery practice (e.g., concrete thinking, Why–How experiential exercise), and linking into If–Then plan	Exercise: writing concrete thinking based on the"how"question and completing Handout 7
Session 7	Training of functional analysis; Exclusive conditions: Whom? What? How? When? Where? identifying and planning alternatives to rumination that serve the same function; useful rules of thumb to discriminate helpful versus unhelpful rumination	Handout 8. Being More Effective (Practicing Functional Analysis)
Session 8	Introduce the idea of absorption, increasing absorbing activities. Process-focused approach: integration of action and awareness, the balance between challenge and skill, focus on the task, focus on the present moment	Handout 9. Being More Effective Form Handout 10. Engaging with Life through Concreteness
Session 9	Training stress management and stress reduction; Relaxation training and practice; Problem-solving training	Exercise: Practicing Relaxation and Problem-solving; Handout 10
Session 10	Introducing the style of self-compassion; Investigating obstacles to self-compassion, acting in a more caring way toward self, and increasing compassionate activities
Session 11	Discussion of therapeutic effects; comparing effective versus ineffective strategies, resilience; review skills, plan for ongoing activity, review experience of the group
Session 12 (Individual)	Functional analysis review for each client; Re-evaluation of warning signs and stresses, and Relapse prevention plans and flexibility

All the handout are based on Watkins'book Rumination-Focused Cognitive-Behavioral Therapy for Depression

Therapeutic team

The group therapy sessions were delivered by a therapeutic team comprising a licensed clinical psychologist with a PhD and extensive experience in RFCBT, and a master’s-level clinician trained in CBT under direct supervision for two years. Additionally, to adhere to the treatment protocols, all participant records were systematically documented and monitored throughout the intervention by the study supervisor.

Instruments

Structured Clinical Interview for DSM-5—Clinician Version (SCID-5)

SCID-5-CV is a structured clinical interview for diagnosing DSM-5 psychiatric disorders. The Persian adaptation demonstrates acceptable psychometric properties, including good inter-rater agreement (kappa = 0.78–0.80), high sensitivity (0.82–0.94), and specificity (0.78–0.98), except for anxiety disorders (Kappa = 0.03) [29].

Beck Depression Inventory—Second Edition (BDI-II)

Developed by Beck et al. (1996), the BDI-II is a 21-item self-report measure evaluating depressive symptom severity [30]. The total score ranges from 0 to 63, categorized into four levels: no symptoms, mild, moderate, and severe [31]. The original version has robust internal consistency (Cronbach's alpha 0.91) and one-week test–retest reliability (0.93). The Persian adaptation has also shown high reliability (r = 0.93). [32].

Repetitive Thinking Questionnaire (RTQ-31)

RTQ-31 is a self-report measure designed to assess repetitive thinking. It consists of two subscales: Repetitive Negative Thinking (RNT; 27 items) and Absence of Repetitive Thinking (ART; 4 items). Responses are rated on a 5-point Likert scale ranging from"Not at all true"(1) to"Completely true"(5). The RTQ-31 has demonstrated good internal consistency, with a reported Cronbach’s alpha of 0.89. [33].

Positive and Negative Affect Schedule: (PANAS-NA)

The PANAS is a self-report questionnaire comprising 10 questions for negative and 10 for positive affect, rated on a 5-point Likert scale. The original version demonstrated satisfactory internal consistency for negative and positive subscales (respectively α = 0.87 and 0.88) and acceptable test–retest reliability (r = 0.71 for negative affect and r = 0.67 for positive affect) [34]. Also, the Iranian sample indicated a good internal consistency (α = 0.85), and construct validity for both subscales [35].

Data analysis method

Statistical analysis

First, a descriptive analysis was conducted to explain participants'characteristics. T-tests and chi-square tests were employed to examine group differences. The Kolmogorov–Smirnov test was used to assess the normality of dependent variables, while Levene’s test evaluated the homogeneity of variances. Baseline equivalence between groups was confirmed for depressive symptoms (F = 0.81, p > 0.375) and rumination (F = 0.08, p > 0.777). Considering the nonconformity to the assumption of the homogeneity of variances for negative affect (F = 7.49; p > 0.010), an independent t-test was used in the pre-test of negative affect. To examine the effectiveness of RFCBT, a mixed analysis of variance (ANOVA) was employed. This approach was chosen due to the small sample size and limited number of measurement points, enabling assessment of both within-subject and between-group effects over time. The Huynh–Feldt correction was applied when the sphericity assumption was violated, and Tukey’s post-hoc test was used to further examine significant intergroup differences. These methodological steps were taken to ensure statistical rigor and clarity throughout the analysis.

The Leeds Reliable Change Index (RCI) method was employed to evaluate clinical significance [36]. The Reliable Change Index (RCI) and Clinically Significant Change (CSC), introduced by Jacobson and Truex [37], were developed by Morley and Dowser for clinical and group-based applications [38]. The method answers two basic questions: 1) Is the observed change greater than measurement error (significant change)? 2) Does the change reach a threshold of clinical relevance (clinically significant change)? RCI is calculated as the ratio of the difference between the post-test and pre-test scores to the “standard error of the difference.” This index determines whether the observed difference is greater than what would occur solely due to measurement error. CSC is also a measure of whether a subject’s post-test scores have moved into the “nonclinical” range or reached a level that distinguishes them from the clinical group [37].

Ethical considerations

This study was initiated after approval by the Ethics Committee of Zanjan University of Medical Sciences (No.: IR.ZUMS.REC.1397.202) and adhered to international ethical guidelines. Researchers and therapists strictly followed core ethical principles such as informed consent, confidentiality, privacy, and safeguarding participants’ mental health throughout all study phases. Prior to enrollment, participants were provided with detailed information about the study’s objectives, procedures, and their rights, and written informed consent was obtained from all individuals. Ethical considerations were regularly reviewed in executive team meetings, with decisions guided by the professional ethics code for psychologists. To maintain confidentiality, participant information was recorded with a numerical code and entered into statistical software.

Results

Baseline demographic and clinical characteristics

The chi-square test for gender and the independent t-test for age revealed no statistically significant differences in demographic variables between the two groups (p > 0.05). These results confirm the baseline comparability within the g-RFCBT and WL groups regarding demographic characteristics. (Table 2). The t-test results indicated no significant differences between the groups on BDI-II scores (t = 0.69; p = 0.496) or PANAS-NA scores (t = 0.44; p = 0.660) during the pre-test. However, the mean RTQ-31 scores in the g-RFCBT group were significantly higher compared to the WL group (t = 3.36; p = 0.002) (Table 3).

Table 2.

Participants demographic characteristics

Variables		g-RFCBT	WL	Total	x2	Sig
		N (%)	N (%)	N (%)
Gender	Male	6(33.3)	5(27.8)	11(30.6)	0.13	0.717
	Female	12(66.7)	13(72.2)	25(69.4)
	Total	18(100)	18(100)	36(100)
Age		Mean ± SD	Mean ± SD	Mean ± SD	t	Sig
		23.44 ± 3.71	22.67 ± 2.14	23.06 ± 3.01	0.77	0.448

RFCBT group-based Rumination-Focused Cognitive-Behavioral Therapy for Depression, WL wait list, M Mean, SD Std. Deviation

Table 3.

Pre-treatment comparisons and general linear model (GLM) repeated measure

	g-RFCBT		WL		t (df)	F (df)	F (df)	F (df)	Eta
	M	SD	M	SD	Pre-treatment comparison	Within-group comparison	Between-group comparison	Interaction timegroup
BDI-II
Pre-treatment	27.16	9.85	25.11	6.64	t(34) =.69, p =.496	F(2,44) = 14.68, p <.001	F(1) = 19.65, p <.001	F(2,44) = 17.76, p <.001	0.34
Post-treatment	9.77	7.24	25.83	7.02
2-Month Follow-up	10.61	11.91	27.22	9.25
6-Month Follow-up	12.22	11.46	24.00	6.05
RTQ-31
Pre-treatment	108.77	16.13	90.44	17.11	t(34) = 3.36, p =.002	F(2,60) = 21.88, p <.001	F(1) = 0.48, p >.05	F(2,60) = 13.35, p <.001	0.28
Post-treatment	77.55	15.48	86.38	17.01
2-Month Follow-up	71.72	24.83	85.38	17.66
6-Month Follow-up	77.00	26.45	87.72	15.32
PANAS-NA
Pre-treatment	30.61	7.22	29.77	4.54	t(34) =.44, p =.660	F(2.27) = 10.92, p <.001	F(1) = 9.33, p <.01	F(2.27) = 5.82, p <.01	0.14
Post-treatment	22.00	4.97	27.88	5.94
2-Month Follow-up	20.22	8.63	28.38	4.84
6-Month Follow-up	22.44	8.48	28.61	4.52

g-RFCBT group-based Rumination-Focused Cognitive-Behavioral Therapy, WL Wait-List

Efficacy of g-RFCBT

The general linear model repeated measures showed that based on the Huynh–Feldt correction, the null hypothesis is rejected for BDI-II (time*group: F = 17.76, df = 2.44, p < 0.001, η² = 0.34), RTQ-31 (time*group: F = 13.35, df = 2.60, p < 0.001, η² = 0.28), and PANAS-NA (time*group: F = 5.82, df = 2.27, p < 0.01, η² = 0.14). This means that the g-RFCBT is effective for reducing depression, rumination, and negative affect (Table 3).

Further analysis revealed that the main effect of time on depression was significant (BDI-II; p < 0.001), indicating changes from pretreatment to 6-month follow-up in the g-RFCBT are significant. In addition, the difference between g-RFCBT and WL on depression was also significant, as revealed by GLM repeated measures (F (2.44) = 17.76, p < 0.001, partial η² = 0.34), indicating a significant difference between the g-RFCBT and WL groups on depression. (Fig. 1.1).

Fig. 1.

Between-group comparisons of outcomes from pre-treatment to 6-month follow-up

The main effect of time for RTQ-31 was also significant (p < 0.001). There was a significant difference between the means of RTQ-31 from pre-treatment to 6-month follow-up (Fig. 1.2). However, the main effect of group on rumination (RTQ-31) was not significant, indicating no significant difference between g-RFCBT and WL. These results are because RTQ-31 scores in g-RFCBT were higher than WL in pre-treatment and this difference was significant, meaning that as RTQ-31 scores in g-RFCBT decreased, WL scores remained constant over time. This was confirmed by GLM repeated measures, which revealed a nonsignificant main effect of group (F (2.60) = 13.35, p < 0.001, partial η² = 0.28).

The main effect of time for PANAS-NA was also significant (p < 0.001). Additionally, a significant difference was observed in the mean RTQ-31 scores from pre-treatment to the 6-month follow-up, indicating changes over time. (Fig. 1.3). The main effect of the group was also significant, as revealed by GLM repeated measures (F (2.28) = 5.82, p < 0.01, partial η² = 0.14), which indicates the significant difference between g-RFCBT and WL in the PANAS-NA.

Sustaining treatment outcomes

Tukey’s post-hoc test was used to compare the results of 2-month follow-up and 6-month follow-up with post-treatment. Tukey's results for BDI-II, RTQ-31, and PANAS-NA showed that in the g-RFCBT, the results showed that there was no significant change in participants'scores from the end of treatment to the 2-month and 6-month follow-ups (p > 0.01). These findings indicate that participants were able to maintain their treatment gains during both follow-ups.

Furthermore, 72.2% of participants in the intervention group showed a 50% or greater reduction in BDI-II scores at post-treatment. This increased to 77.8% at 2-month follow-up and 66.7% at 6-month follow-up. Also, a 70% or greater score reduction was observed in 44.4% at posttest, 55.6% at 2-month follow-up, and 50% at 6-month follow-up. These findings indicate a significant and sustained improvement in participants'depressive symptoms. In this study, a 50% reduction is considered a"response,"and a 70% reduction is regarded as an"improvement."

Clinical significance

In the Leeds reliable change analysis for BDI-II, Cronbach's alpha was considered to be 0.91, and the RCI was estimated at 6.69, which means that changes greater than 6.69 are reliable. Additionally, the BDI-II clinical cut-off of 13 (representing general population norms) was used to determine clinically meaningful improvement. As a result, 13 participants achieved clinically significant changes from pre-treatment to post-treatment (Fig. 2.1). Additionally, in the pre-treatment to the 2-month follow-up phase, 12 participants obtained reliable clinical changes. As indicated by an increase in their BDI-II, one person deteriorated between pre-treatment and 2-month follow-up (Fig. 2.2). Moreover, between post-treatment and 6-month follow-up, it was found that 6 participants had no change and 11 participants showed a reliable change (Fig. 2.3).

Fig. 2.

Leeds reliable change analysis

In the Leeds formula for RTQ-31, Cronbach's alpha value was considered 0.94 in the Iranian population (11). Based on this, the RCI value equal to 11.05 was obtained, which means that changes greater than 11.05 are valid. To establish criteria for clinical significance, we predefined a 30% reduction in RTQ-31 scores (corresponding to a post-treatment score < 73) as the threshold for treatment response, given the absence of an established cut-off point for RTQ-31. That is, we consider participants whose score is less than 73 as a response to treatment and of clinical significance. As a result, 14 participants had valid changes in RTQ-31 from pre-treatment to post-treatment, and the changes of 8 of them were clinically significant (Fig. 2.4). Furthermore, between pre-treatment and 2-month follow-up, 12 participants had reliable changes (Fig. 2.5), and from pre-treatment to 6-month follow-up, the results of 8 participants were in clinical significance level (Fig. 2.6).

Moreover, a Cronbach's alpha of 0.82 was considered for the clinical significance analysis of PANAS-NA (8). Accordingly, the Reliable Change Index (RCI) value was calculated to be 7.75, indicating that changes greater than 7.75 are reliable. To establish criteria for clinical significance, a 30% reduction in PANAS-NA scores (corresponding to a post-treatment score < 23) was predefined as the threshold for treatment response, given the absence of an established clinical cut-off for this measure. Participants achieving scores below this threshold were classified as demonstrating clinically meaningful improvement. The analysis showed that 9 participants had valid changes between pre-treatment and post-treatment, and 5 participants'changes were clinically significant (Fig. 2.7). Furthermore, from pre-treatment to the 2-month follow-up, 9 participants had clinical significance status (Fig. 2.8), and from pre-test to the 6-month follow-up, 9 participants showed clinical significance (Fig. 2.9).

Discussion

We hypothesized that g-RFCBT would reduce the depressive symptoms, negative affect, and rumination, with treatment effects maintained for at least six months. Findings supported these hypotheses: statistically significant reductions in depressive symptoms (BDI-II scores) were observed in the g-RFCBT group compared to the waitlist (WL) control group from pre-treatment to post-treatment, and these improvements were sustained at the 6-month follow-up (Fig. 1.1). The findings of this study showed that a significant proportion of participants in the intervention group experienced not only statistically significant but also clinically significant reductions in depressive symptoms. Specifically, 13 individuals achieved a 50% reduction in BDI-II score (72.2%) and 8 individuals achieved a 70% reduction (44.4%). The maintenance of these improvements at the six-month follow-up may indicate the sustained long-term efficacy of the intervention.

In the post-treatment phase, 14 participants (77.8%) had reliable improvement in depressive symptoms, of which 9 participants (50.0%) obtained clinically significant changes (Fig. 2.1). Similarly, in the 2-month follow-up, 13 participants (72.2%) showed reliable improvement in depressive symptoms and 12 participants (66.7%) showed clinically significant improvement (Fig. 2.2). By the 6-month follow-up, 12 participants (66.7%) continued to show reliable changes, and 11 participants (61.1%) experienced clinically significant changes (Fig. 2.3). These results show that g-RFCBT produces clinically meaningful improvements for in 50% of participants in post-treatment and after 6 months this amount reached 61.1%.

The findings of present study align with prior evidence from Watkins et al. [22], Hvenegaard et al. [39], Theisman et al. [40], and Jacobs et al. [41] regarding effectiveness of RFCBT. Watkins et al. [22] showed that the effect size of RFCBT in reducing depression was between 2 and 2.5. While their study focused on drug-resistant patients with residual symptoms, the comparable outcomes observed in our student population suggest that g-RFCBT may achieve similar efficacy as individual RFCBT. In another study, Hvenegaard et al. [39] compared g-RFCBT with g-CBT and showed that g-RFCBT has relative superiority over g-CBT. Despite a notable attrition rate (approximately 50% loss to follow-up at 6 months), the authors concluded that g-RFCBT could serve as a beneficial alternative to g-CBT for MDD. Teismann et al. [40] replicated these findings in a separate group therapy trial.

The findings demonstrated that g-RFCBT significantly reduced rumination. In the intervention group, rumination levels decreased significantly from baseline to post-treatment, with these effects maintained at the 6-month follow-up (Fig. 1.2). According to the Leeds Clinical Significance Index, 15 participants (83.3%) achieved reliable improvement post-treatment, and 8 participants (44.5%) showed clinically significant changes (Fig. 2.4). At the 2-month follow-up, 16 participants (88.9%) maintained reliable improvement, with 12 participants (66.7%) achieving clinical significance (Fig. 2.5). By the 6-month follow-up, 15 participants (83.3%) continued to show reliable improvement, and 8 participants (44.5%) sustained clinically significant changes (Fig. 2.6). These findings indicate that g-RFCBT produced significant clinical reductions in rumination for 44.5% of participants immediately post-treatment, with equivalent maintenance at the 6-month follow-up. This result aligns with prior evidence from Watkins et al. [22] and Teismann et al. [40], which reported significant reductions in rumination following RFCBT interventions, with sustained effects observed during follow-up assessments.

Hvenegaard et al. [39] reported that RFCBT reduced rumination; however, no significant difference was observed between RFCBT and CBT in this regard. In contrast, Spinhoven et al. demonstrated in a meta-analysis that interventions directly targeting rumination yielded greater reductions in rumination compared to conventional treatments [6].

Negative affect, a construct closely linked to rumination, was also examined in this trial. The results revealed a statistically significant decrease in negative affect in the intervention group from baseline to post-treatment, with these effects sustained at both the 2-month and 6-month follow-ups, indicating the enduring efficacy of g-RFCBT in mitigating negative affect over time (Fig. 1.3). According to the Leeds Clinical Significance Index, post-treatment assessments showed that 9 participants (50.0%) achieved reliable improvement, while 5 participants (27.8%) demonstrated clinically significant changes (Fig. 2.7). At the 2-month follow-up, 10 participants (55.6%) maintained reliable improvement, and 9 participants (50.0%) sustained clinical significance (Fig. 2.8). By the 6-month follow-up, these outcomes remained consistent: 10 participants (55.6%) showed reliable improvement, and 9 participants (50.0%) retained clinically significant changes (Fig. 2.9). These findings suggest that g-RFCBT effectively reduces negative affect, with 50% of participants maintaining clinically meaningful improvements at the 6-month follow-up.

Following randomization, assessments showed that RTQ-31 scores at pre-test significantly differed between the intervention group and the WL. Given that randomization was based on BDI-II scores and the sample size was relatively small, there is a possibility that such a difference could have occurred in some variables (e.g., RTQ-31). Nevertheless, results from mixed-effects analysis of variance (ANOVA) and the Leeds Clinical Significance Index consistently demonstrated the efficacy of g-RFCBT in reducing rumination over time.

While the efficacy of RFCBT on negative affect has not been directly investigated yet, the interventions grounded in classical CBT frameworks targeting rumination or repetitive thoughts have demonstrated efficacy in reducing negative affect. For instance, the findings of our study align with studies that used special diagnosis CBT to reduce negative affect. Including the study of Saxon et al. showed the effectiveness of CBT on negative emotion with a significant and large effect size (hedges'g = 0.98) [42]. Similarly, McEvoy et al. demonstrated that group-based metacognitive therapy for negative repetitive thoughts in generalized anxiety disorder led to significant reductions in negative affect [43].

Although limited research has directly examined the efficacy of RFCBT on negative affect, its therapeutic mechanisms align with established theoretical models of rumination. For example, Watkins and Nolen-Hoeksema conceptualize depressive rumination as a habitual cognitive process triggered by depressed mood [28]. According to their model, this maladaptive habit is activated by diverse contextual cues—such as negative emotional states, unmet goals, or empathic responses to others’ distress—and involves passive focus on the causes and implications of depressive symptoms rather than adaptive problem-solving. This cyclical pattern perpetuates and exacerbates depressive symptoms [28].

During the treatment, patients engage in RFCBT techniques—including functional analysis, if–then planning, behavioral/imagination exposure, absorption training, self-compassion exercises, and problem-solving training—to shift from abstract to concrete thinking styles, ultimately fostering adaptive problem-solving. By strengthening concrete cognitive processing, RFCBT effectively disrupts the cycle of depressive rumination. Furthermore, the Repetitive Thinking Questionnaire-31 (RTQ-31), a transdiagnostic measure used in this study to assess rumination [44], demonstrated significant reductions following RFCBT. This outcome suggests that RFCBT may not only target rumination but also broader repetitive negative thought patterns (RNTs). However, these findings require further investigation to confirm the generalizability of RFCBT’s effects across transdiagnostic cognitive processes.

RFCBT reduces negative affect by shifting cognitive processing away from self-focused attention (e.g., bodily sensations and negative emotions) toward more adaptive mental strategies. A core feature of rumination involves abstract-evaluative processing—characterized by"why"questions, comparative thinking, and self-referential evaluation—which perpetuates self-blame, dissatisfaction, and negative emotions. Watkins posits that transitioning from abstract to concrete processing disrupts this maladaptive cycle, thereby reducing negative affect [45]. hence, by focusing on the concrete processing style, the patient is taught to practice"how"based questions (instead of focusing on"why"based questions) to specifically and actively seek the solution so that the possibility of solving the problem becomes more successful, and as a result, it can affect the reduction of negative emotions.

In addition, in some theories, especially the Response Styles Theory (RST), particularly as articulated by Nolen-Hoeksema, posits that rumination is primarily triggered by negative emotional states. Empirical evidence further indicates that, in the absence of negative affect, rumination lacks a demonstrable causal role in the onset or maintenance of depressive symptoms. This underscores the centrality of negative emotion in the ruminative process within the RST framework, suggesting that rumination does not independently drive depression but rather amplifies distress in the context of preexisting negative mood states [46, 47].

Finally, this study represents one of the first investigations into the efficacy of g-RFCBT in reducing negative affect. As with many early-phase clinical trials, methodological limitations inherent to pilot studies may influence the interpretation of findings. Therefore, the results should be interpreted with caution and viewed as preliminary until replicated in larger, more diverse populations.

Conclusion

This study demonstrates robust evidence for the efficacy of g-RFCBT in reducing depressive symptoms, rumination, and negative affect, with sustained effects observed at 6-month follow-up. Statistically significant reductions in depressive symptoms were identified from pre-treatment to post-treatment, accompanied by reliable and clinically significant improvements. These effects were maintained at both the 2- and 6-month follow-up assessments. Notably, 61.1% of participants achieved clinically significant changes at the 6-month follow-up, underscoring the clinical utility of g-RFCBT in fostering durable recovery. These findings highlight g-RFCBT as a promising intervention for recurrent major depressive disorder (MDD), particularly given its capacity to sustain therapeutic gains over time.

Moreover, the study highlights g-RFCBT’s impact on rumination, showing substantial reductions in rumination levels that persisted across follow-up periods. While findings indicate g-RFCBT effectively targets rumination and RNT, additional research is needed to evaluate its applicability across diverse clinical populations. Notably, over 80% of participants achieved reliable improvement on the Repetitive Thinking Questionnaire-31 (RTQ-31), reinforcing the intervention’s capacity to mitigate rumination-related cognitive patterns linked to depressed mood. These results underscore RFCBT’s potential to enhance emotional regulation and cognitive flexibility over time.

Furthermore, the findings show that g-RFCBT can significantly reduce negative affect, a construct closely linked to rumination. A notable proportion of participants demonstrated valid and clinically significant improvements in negative affect at post-treatment and follow-ups, indicating the intervention’s sustained emotional benefits. The observed improvements in negative affect following g-RFCBT intervention also suggest that this approach effectively addresses emotional regulation, further supporting its therapeutic value in mood-related disorders.

Limitation

This study aimed to minimize potential confounding factors that could influence its validity. However, the sample of this study was limited, and future research with larger sample is needed to enhance the robustness of findings and strengthen statistical power. Additionally, the present study was conducted exclusively among a student sample, which may limit the generalizability of the findings to the broader population. To address this limitation, future studies should evaluate g-RFCBT can be used on other populations in order to establish its broader applicability and external validity.

This study did not employ an intention-to-treat (ITT) analysis. While primary analyses were conducted on participants who completed all study phases (both the intervention and the assessments), the absence of ITT analysis may limit the generalizability and robustness of findings particularly in interventional studies. This limitation underscores the importance of adopting ITT frameworks in future interventional studies to enhance the validity of conclusions.

A further limitation of the present study is the baseline imbalance in RTQ-31 scores between the intervention and WL groups. Although randomization was performed correctly using BDI-II scores at pretest, since this allocation was not based on rumination index and the sample size was relatively small, this initial difference could have affected the results regarding changes in rumination. Therefore, caution should be exercised when interpreting the effect of the intervention on reducing rumination.

Another limitation of this study is using a waiting list control group rather than an active comparison group. Given the modest sample size and the exploratory nature of this pioneering investigation in Iran, the study prioritized evaluating the feasibility, acceptability, and preliminary efficacy of the RFCBT protocol. However, the absence of therapeutic engagement in the control group may have confounded the interpretation of outcomes, particularly regarding the specificity of treatment effects. Therefore, it is suggested that future investigations should employ active control conditions with equivalent therapeutic attention.

Abbreviations

CBT

Cognitive-Behavioral Therapy

RFCBT

Rumination-Focused Cognitive-Behavioral Therapy

g-RFCBT

Group-Rumination-Focused Cognitive-Behavioral Therapy

WL

Wait List

MDD

Major Depressive Disorder

BDI-II

Beck Depression Inventory—Second Edition

RTQ-31

Repetitive Thinking Questionnaire – 31 items

RNT

Repetitive Negative Thinking

ART

Absence of Repetitive Thinking

PANAC-NA

Positive and Negative Affect Schedule—Negative Affect

PANAC-PA

Positive and Negative Affect Schedule—Positive Affect

IRCT

Iranian Registry of Clinical Trials

MCT

Metacognitive Therapy

MBCT

Mindfulness-Based Cognitive Therapy

ECT

Electroconvulsive Therapy

DSM-5

Diagnostic and Statistical Manual of Mental Disorder, 5th Edition

SCID-5

Structured Clinical Interview for DSM-5

RCI

Reliable Change Index

Authors’ contributions

Mohsen Hasani:  Investigation, Data Curation, Validation, Writing- Original draft preparation; Saeedeh Zenoozian: Investigation, Conceptualization, Supervision, Formal analysis; Reza Ahmadi: Writing- Original draft preparation, Writing- Reviewing and Editing; Sahel Khakpoor: Investigation, Visualization; Reza Pirzeh: Investigation, Resources; Soleiman Saberi: Investigation, Data Curation; Omid Saed: Conceptualization, Methodology, Formal analysis, Funding acquisition, Project administration.

Funding

This study was supported by Zanjan University of Medical Sciences. The funding body played no role in study design, analysis or interpretation of data in this paper.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The APA ethical standards were followed in the conduct of the study, and we received approval from the Zanjan University of Medical Sciences review board (No. IR.ZUMS.REC.1397.202), which it is approved in the Iranian Registry of Clinical Trials (IRCT Id: IRCT20181117041676N1) (Registration Date: 03/06/2019).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.