Demand for adult mental health services in England continues to rise. According to the latest Mental Health Services Monthly Statistics, almost 1.5 million people were in contact with adult mental health services at the end of December 2025. While this reflects improved access to care, it also highlights the significant pressure facing NHS services.
Community Mental Health Teams (CMHTs) provide ongoing support for people with complex and enduring mental health difficulties. However, workforce shortages and high demand mean that many service users face long waiting times for psychological therapies. As a result, services are increasingly exploring ways to deliver psychological support more efficiently while maintaining clinical benefit.
One potential approach is group-based psychological interventions. Delivering therapy in groups can allow services to reach more people with limited clinical resources compared with individual therapy alone. Group formats may be especially suited to interventions targeting transdiagnostic processes, such as shame and self-criticism.
Compassion-focused therapy (CFT) aims to address these processes by helping individuals develop compassion towards themselves and others. Drawing on an evolutionary model of emotion regulation, CFT seeks to reduce shame and self-criticism by strengthening the brain’s soothing system and rebalancing affect-regulation processes (Gilbert, 2009). Through psychoeducation and experiential exercises, the approach aims to normalise emotional difficulties and reduce self-blame. Although CFT is increasingly used in clinical settings, its empirical evidence base is still developing (Craig et al., 2020). Group delivery may be particularly relevant for CFT, as the relational environment can support shared understanding, reduce feelings of isolation, and foster compassion towards both self and others.
Against this backdrop, Vivolo et al. (2025) conducted a service evaluation of compassion-focused therapy groups delivered in two CMHTs in the East of England.
Group-based interventions with a focus on compassion may be a tool for targeting transdiagnostic symptoms and improving accessibility of support in overstretched community mental health services.
Methods
Vivolo and colleagues conducted a mixed-methods service evaluation, combining routine outcome measures with qualitative service user feedback.
Before the groups were delivered, the psychologists leading the intervention met twice with a local Lived Experience Advisory Group (LEAG) to gather feedback on the group’s structure, content, and implementation. This input helped inform the design of the programme.
Twelve participants attended two compassion-focused therapy groups (six participants per group), delivered within CMHTs. All participants identified as White British, with a mean age of 53 (range 28-65), and most were female (N=9). The most common diagnosis was depressive disorder; other diagnoses included bipolar disorder, anxiety disorders, obsessive-compulsive disorder, and eating disorders.
Intervention
The intervention took a transdiagnostic approach, whereby participants were referred based on common difficulties, e.g., high levels of shame or self-criticism, rather than specific diagnoses. It was an eight-week closed group intervention, with sessions lasting between 90 and 120 minutes.
Groups were facilitated by clinical psychologists and supported by assistant or trainee psychologists and other mental health professionals. The following topics combined teaching, experiential exercises, and between-session practice tasks, as in CFT:
- psychoeducation about emotional regulation and the “three systems” model;
- compassion-focused imagery exercises;
- developing a “compassionate self”;
- working with self-criticism; and
- cultivating self-soothing and compassionate behaviour.
Measures
Pre- and post-intervention self-report measures included:
- PHQ-9 (depression)
- GAD-7 (anxiety)
- Rosenberg Self-Esteem Scale
- Compassion-related measures assessing self-criticism, self-reassurance, and self-compassion.
Participants also completed feedback questionnaires, including open-ended questions about their experience of the group.
Results
Given the small sample size, the researchers examined reliable change and clinically significant change at the individual level, rather than conducting group-level statistical tests. Reliable change indicates whether a change in scores is greater than would be expected due to measurement variability, calculated using the Reliable Change Index (Jacobson & Truax, 1991). Whereas clinically significant change refers to whether a participant’s score moves from the range typical of a clinical population to that of a non-clinical population following treatment (Jacobson & Truax, 1991).
Quantitative outcomes
Overall, the results suggested improvements for some participants, particularly in compassion-related outcomes.
Several participants demonstrated reliable or clinically significant improvements in self-compassion, self-kindness, self-reassurance, and reduced self-criticism. Improvements were also observed in general mental health outcomes. Approximately half of the participants showed reliable improvement or recovery on depression and anxiety measures, while some improvements were also reported in self-esteem.
However, results were not consistent across all measures. Some participants showed deterioration on certain subscales, particularly those relating to compassion directed towards others or received from others. A small number also showed declines in self-esteem. These mixed findings highlight that responses to the intervention varied considerably between individuals.
Qualitative feedback
Participants also completed feedback questionnaires about their experience of the group. Responses to closed-ended questions were summarised using frequency analysis, while open-ended responses were analysed using content analysis, a method suited to identifying common themes in brief qualitative responses.
Despite the variability in quantitative outcomes, participants generally reported positive experiences of the group. Most participants indicated that they enjoyed the sessions and found the content useful. Many described the group as supportive and emphasised the importance of a safe and non-judgemental environment, especially when engaging with emotionally challenging topics such as shame and self-criticism.
Two broad themes emerged from the qualitative responses:
- Perceived benefits of the group: participants reported that the facilitators created a compassionate and supportive atmosphere that helped them feel comfortable sharing their experiences. Several participants highlighted specific techniques, such as imagery exercises and strategies for managing anxiety and self-criticism.
- Suggestions for improving the intervention: participants recommended increasing the number of sessions, incorporating more visual materials, and maintaining consistent session length.
The mixed-methods service evaluation provided evidence on symptom changes and experiences of CFT groups in routine CMHT care.
Conclusions
Vivolo and colleagues suggest that compassion-focused therapy groups may help some CMHT service users develop greater self-compassion and reduce self-criticism. Some participants also experienced improvements in depression, anxiety, and self-esteem.
However, the authors emphasise that the findings should be interpreted cautiously. The study was a small service evaluation rather than a controlled research study, and therefore cannot establish whether the observed improvements were caused by the intervention.
Group-based CFT may help CMHTs address service users’ experiences of shame and self-criticism while waiting for individual therapy.
Strengths and limitations
This evaluation offers useful insights into how compassion-focused therapy groups may function within routine NHS services. The study was conducted in a real-world CMHT setting, increasing its relevance for clinical practice in the UK. Importantly, the intervention targeted transdiagnostic processes, such as shame and self-criticism, which are common across many mental health conditions and may be particularly relevant for individuals in secondary care services, especially if we consider the shift of NHS services to offer group interventions to combat long waiting lists cost-effectively.
However, several limitations constrain the interpretation of the findings.
- First, the sample size was small, with twelve participants across two groups. This limits statistical power and makes it difficult to draw reliable conclusions.
- Second, the study was conducted as a service evaluation without a comparison group, meaning improvements cannot be confidently attributed to the intervention itself. Participants may have improved over time due to other factors, such as ongoing treatment or natural recovery.
- Third, the two sites used different compassion outcome measures, which reduced comparability across participants and limited the analyses that could be conducted.
- Finally, the sample lacked diversity, with all participants identifying as White British, limiting the generalisability of the findings.
Taken together, these limitations mean that the results should be interpreted as preliminary and exploratory for the use of CFT group interventions in secondary mental health settings, rather than evidence of effectiveness.
Small, real-world service evaluations can provide useful early insights into how interventions work in CMHT routine practice, but findings should be interpreted cautiously.
Implications for practice
Despite these limitations, the study raises important questions about how psychological support can be delivered within resource-constrained services. Many CMHTs face substantial waiting lists for individual therapy. In this context, group-based interventions may offer a pragmatic way to increase access to psychological support while people wait for individual treatment. In practice, structured group interventions may offer a way for CMHTs to provide earlier psychological input while individuals wait for individual therapy, potentially preventing symptoms from escalating during long waiting periods.
Compassion-focused therapy groups may be particularly relevant for individuals experiencing shame and self-criticism, which are common across a wide range of mental health difficulties. Whereas the group format may help foster a sense of shared experience and reduce feelings of isolation.
However, it is important to avoid drawing firm conclusions about effectiveness based on this study alone. Future research should aim to conduct larger and more rigorous studies, ideally including randomised controlled trials and longer-term follow-up. Standardising outcome measures and recruiting more diverse samples would also strengthen the evidence base. Given the pressures currently facing CMHTs, further research exploring whether group-based psychological interventions can reduce waiting times or improve access to care could be particularly valuable.
Amid growing pressures on CMHTs, group-based CFT may offer a potential route to expanding cost-effective psychological support to people on long waiting lists.
Statement of interests
Sofiia Kornatska has no conflicts of interest to declare.
Edited by
Dr Dafni Katsampa.
Links
Primary paper
Marco Vivolo, Gabriel Ardeman, Catherine Ford (2025). Compassion-Focused Therapy Groups in Secondary Care Adult Mental Health Services: A Service Evaluation. International Journal of Cognitive Behavioral Therapy, 18(1), 94-113. https://doi.org/10.1007/s41811-025-00230-x
Other references
Craig, C., Hiskey, S., & Spector, A. (2020). Compassion focused therapy: a systematic review of its effectiveness and acceptability in clinical populations. Expert Review of Neurotherapeutics, 20(4), 385-400. https://doi.org/10.1080/14737175.2020.1746184
Gilbert, P. (2009). Introducing compassion-focused therapy. Advances in Psychiatric Treatment, 15(3), 199–208. https://doi.org/10.1192/apt.bp.107.005264
Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12–19. https://doi.org/10.1037/0022-006X.59.1.12
