Exploring illness and recovery in eating disorders: a qualitative study of patients’ group therapies experiences

Abstract

Background

Recovery in eating disorders (EDs) is often defined in terms of symptom remission, potentially overlooking the broader meanings individuals attribute to illness and recovery. Qualitative research has highlighted the multidimensional nature of recovery, yet less is known about how therapeutic contexts may shape how individuals make sense of illness, change, and recovery. This study aimed to explore how individuals with EDs conceptualize illness and recovery, and how these meanings may be shaped through participation in group psychotherapy.

Methods

A qualitative study was conducted within an interpretivist framework using reflexive thematic analysis. Ten adult outpatients with EDs participated in semi-structured interviews at the end of a group psychotherapy program integrated within a multidisciplinary treatment setting.

Findings

Participants described a shift from understanding the ED mainly through food, weight, body image, and symptom control toward a more emotionally and relationally informed understanding of distress. Recovery was increasingly constructed as a multidimensional and non-linear process involving emotional awareness, self-acceptance, agency, identity renegotiation, and connection with others. The group was experienced as a space that fostered recognition, belonging, and reflection, while also exposing participants to shame, fear of judgment, and emotional vulnerability.

Conclusions

Findings suggest that group psychotherapy may support a reconfiguration of how individuals with EDs understand illness and recovery, moving from a primarily symptom-centred perspective toward a more holistic, relational, and personally meaningful framework. Rather than operating mainly through direct behavioural change, the group appeared to function as a space for meaning-making, emotional reflection, and relational experimentation. This study highlights emotional awareness, identity renegotiation, self-acceptance, and interpersonal connection as potential targets for fostering recovery in EDs.

Trial registration The study protocol (n. 0034565/i) was approved by the local Institutional Review Board.

Plain language summary

Eating disorders are often understood mainly in terms of eating behaviours, weight, and body image. However, people with eating disorders may experience illness and recovery in broader and more personal ways, involving emotions, relationships, identity, shame, and self-acceptance. This study explored how adults with eating disorders understood their illness and recovery after taking part in group therapy within a multidisciplinary treatment setting. Ten adults were interviewed after completing a group therapy program. Before the group experience, participants often described their eating disorder mainly through symptoms such as food restriction, bingeing, purging, body dissatisfaction, and weight control. After the group, many described their difficulties in a broader way, linking the eating disorder to emotions, relationships, and personal experiences. Recovery was seen not only as reducing symptoms, but also as becoming more aware of emotions, accepting oneself, developing personal agency, and connecting with others. The group helped some participants feel understood and less alone, but it could also involve shame, fear of judgment, and emotional vulnerability. Overall, group therapy appeared to support broader and more personally meaningful understandings of recovery.

Background

Eating disorders (EDs) are complex and multifactorial psychiatric conditions, often characterized by chronic and relapsing courses and significant psychosocial impairment [1–3]. Despite advances in diagnosis and treatment, EDs remain difficult to conceptualize and treat effectively due to heterogeneous presentations and the intricate interplay of psychological, biological, and social factors [4].

The Diagnostic and Statistical Manual of Mental Disorders (DSM-5; [5]) has played a key role in standardizing diagnostic criteria and facilitating research and clinical communication. However, its categorical approach has shown substantial limitations: frequent diagnostic cross-over, poor longitudinal stability, and a reduction of complex subjective experiences to objective behavioral and somatic criteria [6]. In response, dimensional and transdiagnostic models such as the Research Domain Criteria (RDoC; (7)) and the Hierarchical Taxonomy of Psychopathology (HiTOP [8, 9]) have been proposed. These models integrate neurobiological, cognitive, and behavioral dimensions; however, they have not produced a significant advance for clinicians given they were not able to date to provide useful prognostic and therapeutic indications [10]. However, these models primarily rely on observable and measurable criteria, which may not fully capture the subjective and lived experience of individuals—an aspect that is crucial for understanding both illness and recovery [3, 11].

The issue of recovery definition is tightly connected to these diagnostical limitations. Within current diagnostic systems, remission is largely based on symptom reduction and weight restoration, overlooking broader dimensions such as empowerment, hope, identity, and quality of life. This narrow definition neglects psychological criteria, failing to reflect what recovery means from the patient’s perspective. Recent qualitative syntheses have emphasized the importance of integrating patients’ voices into the conceptualization of recovery, showing that this process extends beyond symptom remission to include personal growth, emotional awareness, and social reconnection [12]. Importantly, individuals with lived experience consistently describe recovery as a subjective and evolving process, rather than a fixed endpoint. From this perspective, recovery involves the construction of personal meanings and is shaped by relational, emotional, and contextual factors. Within this framework, quality of life has emerged as a central dimension of recovery, encompassing broader aspects of functioning, well-being, and life satisfaction beyond the mere absence of symptoms [13, 14].

Given that recovery is shaped by relational and experiential processes, group-based interventions for EDs may represent a particularly relevant context in which meanings of illness and recovery are negotiated and transformed. Systematic reviews and meta-analyses highlight considerable methodological heterogeneity—differences in treatment format, group composition, and outcome measures—which limit the comparability and generalizability of results [15, 16]. Trombetta et al. [17] reviewed psychodynamic group approaches and emphasized how shared reflection and emotional attunement can promote insight and self-acceptance—yet empirical work exploring these phenomena remains limited. Theoretical models such as Yalom’s therapeutic factors [18] continue to provide valuable guidance, but the specific ways in which these factors operate within ED populations have yet to be elucidated. In particular, group cohesion—defined as the sense of belonging, trust, and emotional connection among group members—has been consistently identified as a central therapeutic process in group psychotherapy and has been associated with improved treatment outcomes across different clinical populations [19].

In this context, qualitative analysis offers a unique opportunity to explore patients’ lived experiences, emotional meanings, and perceived processes of change—dimensions often inaccessible through quantitative approaches ([20]. Integrating such insights with existing quantitative evidence can enrich our understanding of recovery trajectories and guide the development of more person-centred and psychologically informed interventions [21]. The present study aims to explore how individuals with EDs conceptualize illness and recovery, and how these conceptualizations may shift through participation in group psychotherapy. It also explores participants’ experiences of the group process and the meanings they attribute to change within this context in order to better understand how recovery is subjectively constructed and transformed within a relational therapeutic setting.

Methods

Study design

This study was conducted within an interpretivist paradigm and adopted a qualitative approach using reflexive thematic analysis as outlined by Braun and Clarke [22, 23]. An iterative and reflexive analytic approach was adopted, allowing both inductive identification of patterns of meaning and theoretically informed interpretation guided by the research aims.

The primary aim was to explore how individuals with EDs conceptualize illness and recovery, and how these conceptualizations may shift within the context of group psychotherapy. The analysis explored session-related emotions and themes, alongside participants’ satisfaction with how behavioral aspects of the illness (i.e., relationships with food and compensatory behaviors) were addressed.

Reflexivity and positionality

The first author (M.C., male) is a psychiatrist–researcher and systemic–family psychotherapist with extensive experience in the outpatient treatment of EDs. M.C.’s clinical background and research expertise provided a deep understanding of the therapeutic processes and systemic dynamics that can shape individuals’ experiences of group therapy. The principal investigator was a university professor–psychiatrist (A.M.M.) with experience in cognitive-constructivist psychotherapy and extensive knowledge of the scientific literature on EDs.

The research team included a clinical psychologist (A.C.) with more than ten years of experience in group therapy and eating disorder treatment, and two sociology students. One sociology (V.L.) student was a lived-experience expert in remission from an ED, with prior training in qualitative research and valuable insight into the lived experience of EDs. The other sociology student (G.L.) had prior academic training in qualitative methodology and contributed expertise in the rigor and transparency of the analytic process. One psychiatry resident (F.M.) had received specific training in qualitative research methods and semi-structured interviewing prior to data collection. The interviewer had no prior therapeutic relationship with the participants, which was intended to reduce the risk of response bias. The other authors were psychiatrists (V.R., G.P., E.B.) or a psychologist (F.R.) who had completed cognitive psychotherapy training, had clinical experience working with individuals with EDs, and conducted individual psychotherapies at our outpatient center.

The diversity of professional backgrounds and experiential perspectives was considered an asset to the credibility and depth of the analysis. However, we acknowledge that our disciplinary training, clinical experience, and, in one case, personal recovery from an ED, may have shaped the ways in which we interpreted participants’ narratives. During study design, coding, and theme development, the research team engaged in reflexive discussions to identify and critically examine these potential biases. These discussions aimed to ensure that the final themes reflected participants’ experiences of group therapy as faithfully as possible, rather than the preconceptions or disciplinary priorities of the research team.

Group intervention

The clinical intervention examined in this study was delivered through a group psychotherapy format integrating principles of Emotion-Focused Therapy with techniques derived from the cognitive-constructivist tradition, specifically adapted for EDs. This integrative approach is supported by evidence highlighting the relevance of EFT-oriented interventions in addressing difficulties in emotional processing and regulation commonly observed in EDs [24, 25]. The therapeutic stance was grounded in an emotion-centered and attachment-informed perspective, with attention to relational dynamics involved in the development and maintenance of ED. The group was conceived as an experiential and relational context in which participants could explore emotionally salient experiences through narrative reflection and shared discussion, inviting them to consider how others’ narratives resonated with their own histories and emotional patterns, thereby fostering emotional awareness, interpersonal exchange, and reduced isolation. Each full cycle consisted of 12 sessions delivered over 6 months. Groups were conducted in a closed format. Sessions were co-facilitated by a psychiatrist and a clinical psychologist, both trained in cognitive-constructivist and emotion-focused approaches.

Participants and recruitment

Participants were recruited from an outpatient ED treatment program for adults, offering structured group therapy twice a month, each lasting 90 min, as part of a multidisciplinary care (including individual psychotherapy, nutritional counseling, and monitoring of medical complications). Participants were consecutively invited to take part in the group therapy as part of this multidisciplinary outpatient treatment programme. All participants were within the first three months of treatment. Among the patients invited to participate into the group therapy, six declined. Three patients declined because they did not wish to take part in a group-based intervention, and three declined due to scheduling incompatibilities with the group sessions. No refusals were related to dissatisfaction with the ongoing treatment programme.

Inclusion criteria to participate into the study were:

1. diagnosis of an ED according to the DSM-5 criteria;

2. participation in at least one full cycle of group therapy.

All individuals participating in the group therapy were invited to take part in the study: two participants declined. All participants were informed about the study and provided written informed consent prior to participation. Participation was entirely voluntary and did not affect access to treatment, and participants were free to withdraw at any time. The study protocol (n. 0034565/i) was approved by the local Institutional Review Board.

Data collection

Data were collected at the end of group therapy through semi-structured, face-to-face interviews, each lasting approximately 30 min, conducted in a quiet, distraction-free setting. All interviews were carried out by a single interviewer (F.M.)—a psychiatry resident with experience in EDs—who was part of the research project but had not participated in the group therapy sessions. This methodological choice was intended to minimize the risk of bias related to any pre-existing therapeutic relationship between interviewer and participant. Interviews were video-recorded and later transcribed verbatim and anonymized, with identifying information removed and replaced by codes. The transcripts were not sent to the participants for member checking. Demographic and clinical data were collected at the first clinical assessment through a clinical interview.

Interviews began with broad, open-ended questions (e.g., “Tell me about your experience of the group”), followed by more focused prompts only if specific topics had not been spontaneously addressed. Interview questions pertaining to the current study are provided in Table 1.

Table 1.

Semi-structured interview guide

Interview questions
Q1. Tell me about your experience of the group
Q2. What does having an eating disorder mean to you? Do you think the group experience changed your understanding of what it means to have an eating disorder?
Q3. What does recovery from an eating disorder mean to you? Do you think the group experience changed your understanding of what recovery means?
Q4. At the end of this experience, do you think the group helped you cope with your eating disorder?
Q5. During the group sessions, were there times when you engaged in specific activities?
Q6. How did you experience taking part in these activities? If no activities were reported: before starting group therapy, did you expect to take part in specific activities?
Q7. Were you satisfied with the topics addressed during the group activities?
Q8. Do you think there were topics that were not sufficiently addressed?
Q9. Do you think addressing those topics would have improved your experience within the group?
Q10. Which topic discussed in the group do you consider to have been particularly effective for your therapeutic process?

The interview guide comprised two main sections. The first, Conceptual exploration, asked participants to describe their understanding of illness and recovery before and after the group experience, and to reflect on any perceived changes. Participants were also invited to evaluate the relevance and completeness of the topics discussed during the sessions.

The second section, Impact of group therapy, focused on participants’ perceptions of the group’s influence, including its impact on eating-related behaviors, emotions, and salient themes experienced during the group. Participants also reflected on the overall relevance of the topics addressed and identified any unmet needs. These areas were used to facilitate discussion but were not used as a predefined coding framework during analysis.

Sample size was guided by the concept of information power, according to which the more relevant information the sample holds in relation to the study aim, the fewer participants are required [26]. Given the specific study aim, the relatively homogeneous sample, and the in-depth analytic approach, the sample was considered to provide sufficient information power.

Data analysis

Transcripts were analyzed using NVivo software to facilitate systematic coding and organization. Coding was conducted iteratively and reflexively, guided by the research questions rather than by predefined domains, while remaining open to patterns of meaning arising from the data. Coding was conducted by the first author and reviewed by a second researcher to ensure credibility through peer debriefing rather than quantitative inter-rater reliability. The research team engaged in ongoing reflexive discussions throughout the analytic process to examine how their perspectives and assumptions may have influenced interpretation. Themes were refined through iterative comparison of codes and participant quotations, and an audit trail of coding decisions was maintained. Representative quotations were selected to illustrate each theme. Theme development involved active interpretation by the researchers, moving beyond descriptive coding to identify patterns of shared meaning relevant to the study aims. These strategies were adopted to enhance trustworthiness, including credibility, dependability, and confirmability of the analysis. The analysis followed Braun and Clarke’s six-phase reflexive Thematic Analysis [22].

Findings

Overview

A total of 12 patients were invited to participate into the study, of whom 10 agreed to participate. (mean age: 23.4 years, SD: ± 5.9 years; mean illness duration: 3.8 SD: ± 1.75; mean age of onset: 19.2 years, SD: ± 5.47 years; gender: 80% female, 20% male, 0% non-binary). Among participants, 50% had a diagnosis of AN, 20% BN, 30% Binge Eating disorder (BED).

Themes

The reflexive thematic analysis generated four main themes: from symptom control to emotional meaning-making; negotiating identity: from fusion with the disorder to ambivalent separation; being understood: recovery as a relational and shared process; and the group as a space of relational exposure: safety, shame, and emotional risk.

From symptom control to emotional meaning-making

Participants initially described the ED primarily through behavioural and bodily terms, framing distress in relation to food restriction, binge–purge cycles, body dissatisfaction, and weight control. Early narratives were often dominated by concrete symptom-focused descriptions, with limited emotional elaboration and reduced awareness of the relational and psychological meanings associated with eating behaviours. The ED appeared to function as an organizing framework through which distress was experienced and communicated primarily at the level of the body.

“I’m on a diet… counting calories under 1,000 a day, but my weight still isn’t right.” (Participant 9).

“I couldn’t go out because I was terrified people would see me… or even just looking at myself in the mirror disgusted me.” (Participant 3).

Within these accounts, emotional suffering was frequently displaced onto bodily control and appearance concerns. Although some participants referred to painful life experiences or emotional deprivation, these experiences were often described retrospectively and remained only partially integrated into their understanding of the disorder. For several participants, the ED initially appeared as something externally imposed or detached from subjective agency, almost as if the illness had “arrived” from outside rather than emerging within a personal emotional context.

“When the ED arrived, I saw it as… a hand falling down from the sky.” (Participant 2).

This externalized understanding of the illness coexisted with experiences of emotional disconnection and difficulty identifying internal states. In this sense, eating-related behaviours seemed to operate not only as symptoms but also as embodied strategies for managing distress that could not yet be symbolized emotionally.

Across the group process, participants increasingly articulated more emotionally nuanced understandings of both the disorder and recovery. Rather than describing eating difficulties solely in terms of symptom reduction or behavioural control, many began to recognize emotional instability, relational experiences, and psychological vulnerability as central components of their suffering. The relationship with food was no longer framed exclusively as a behavioural problem but became increasingly understood as intertwined with emotional regulation and subjective experience.

“The ED is a way to let my emotions out… but they come out through food, not in a healthy way.” (Participant 10).

Similarly, participants began to describe recovery less as the mere elimination of symptoms and more as the development of emotional awareness and reflective understanding. Emotional recognition, which was largely absent from pre-group conceptualizations of recovery, progressively emerged as a meaningful dimension of change.

“I realized it’s not always about food or my body… sometimes it’s about the situation I’m in.” (Participant 2).

“I caught myself thinking: ‘Do I really feel this?’ I’m so detached from myself that I don’t even realize it; I don’t feel my emotions, I feel almost apathetic most of the time.” (Participant 9).

Importantly, this shift did not represent a complete abandonment of symptom-focused concerns. Eating-related behaviours and fears remained highly salient after the group experience, often described as unstable, fluctuating, and difficult to control. However, participants’ narratives increasingly reflected the capacity to contextualize these experiences within broader emotional and relational processes.

“Some days I’m eating better, but I don’t expect anything, because the day after tomorrow it could all change.” (Participant 8).

Rather than suggesting a linear transition from pathology to recovery, these accounts indicate a movement from concrete, body-centred understandings of distress toward more reflective and emotionally mediated interpretations of the ED experience. The group appeared to facilitate a process through which previously unsymbolized emotional experiences became increasingly thinkable, speakable, and relationally shareable.

Negotiating identity: from fusion with the disorder to ambivalent separation

Participants’ accounts revealed that the ED was often experienced not simply as an illness, but as a deeply embedded component of identity. Before participating in the group, several participants described the disorder as something intimately tied to their sense of self, functioning as a source of safety, familiarity, and psychological continuity. In these narratives, the ED was not only harmful but also emotionally meaningful, making the prospect of recovery potentially threatening because it implied the loss of a valued or stabilizing part of the self.

“I couldn’t imagine myself without the problem with food… eating little has always pleased me, it made me feel safer. And it also made me feel like I had something… something that was mine.” (Participant 4).

Within the group process, participants increasingly demonstrated the capacity to reflect on their relationship with the disorder in more differentiated and ambivalent ways. Rather than describing the ED as entirely fused with the self, some participants began to position it as something simultaneously familiar, protective, and burdensome. This shift did not necessarily involve rejecting the disorder outright; instead, participants often articulated contradictory feelings of attachment and frustration, closeness and exhaustion.

“I love my problem with food, but not how I act because of it… I wouldn’t want to give it up, even though it’s so annoying.” (Participant 6).

The persistence of this ambivalence is analytically important because it complicates simplistic recovery narratives based on symptom rejection or full psychological separation from the illness. Participants’ accounts instead suggest that recovery involved an ongoing negotiation between maintaining aspects of a familiar identity and imagining alternative ways of relating to themselves beyond the disorder.

This process of renegotiation was also reflected in emerging narratives of self-acceptance and agency. After participation in the group, some participants increasingly described recovery not solely in terms of symptom reduction, but as the possibility of relating differently to themselves, including acknowledging imperfection, cultivating compassion, and taking an active role in change.

“I’ve learned that accepting myself, even with flaws, is part of getting better.” (Participant 4).

“It’s about making peace with what happened and taking control of my life instead of waiting for someone else to make it happen.” (Participant 10).

These narratives suggest that the group may have facilitated the development of a more reflective and agentic position toward the self and the disorder. Importantly, this did not imply complete autonomy from the illness or the disappearance of ambivalence. Rather, participants appeared to move from a relatively fixed identification with the ED toward a more dynamic and negotiable relationship with it. The group context may have contributed to this shift by creating opportunities for participants to witness multiple ways of experiencing and narrating EDs. Through exposure to others’ stories, participants were potentially able to recognize aspects of themselves while simultaneously imagining alternative identities not entirely organized around the illness. In this sense, the group functioned not only as a supportive environment, but also as a relational space in which identity positions could be questioned, mirrored, and gradually reconfigured. Overall, this theme highlights how recovery was experienced not as a linear movement away from illness, but as a complex process of identity renegotiation.

Being understood: recovery as a relational and shared process

Participants’ accounts suggested that the group experience fundamentally reshaped how recovery itself was understood. Before participating in the group, the ED was often described as an isolating experience, characterized by withdrawal, interpersonal mistrust, and fear of judgment. Relationships were frequently experienced as sources of threat, misunderstanding, or loss, reinforcing feelings of separateness and emotional disconnection.

“It took away the relationship I had with my parents, with my friends; I lost friendships and everything else.” (Participant 3).

In these early narratives, recovery was primarily conceptualized as an individual struggle focused on symptom management, bodily stabilization, or personal control. Relational dimensions of healing were largely absent from participants’ definitions of recovery prior to the group experience.

Across the group process, however, participants increasingly described recovery as something that occurred through contact with others. Listening to peers with similar experiences appeared to generate a powerful sense of recognition, reducing feelings of alienation and fostering the perception of being emotionally understood in ways participants had rarely experienced elsewhere.

“A doctor knows you, a friend understands you, but someone going through the same struggle… shares common factors that make you reflect and help you a lot.” (Participant 5).

The therapeutic value of the group was frequently linked not only to receiving support, but to encountering others whose experiences resonated with participants’ own internal struggles. This shared recognition seemed to create a form of interpersonal legitimacy, allowing emotional experiences that had previously felt isolating or shameful to become mutually recognizable and therefore more tolerable.

“When we talk here, a word of comfort from them has a very different effect than if it came from someone else, because I know it’s coming from someone who has lived or is living the same thing I’m describing.” (Participant 8).

In this sense, recovery increasingly emerged as a relationally mediated process grounded in shared experience, emotional reciprocity, and interpersonal validation. Participants described the group as a space in which they could recognize aspects of themselves in others, while simultaneously feeling recognized by others in return. These processes appeared to weaken rigid assumptions of being uniquely flawed, incomprehensible, or alone in their suffering.

The experience of sharing also seemed to facilitate greater openness toward interpersonal relationships more broadly. Several participants described shifts in how they perceived and approached others outside the group, reporting reduced expectations of criticism and greater willingness to engage socially.

“It helped me realize that not everyone is there to judge you…” (Participant 3).

“Now I say: ‘They listen to me, so I have to make myself heard too.” (Participant 8).

These accounts suggest that the group functioned as a relational micro context in which participants could experiment with alternative interpersonal experiences. Through repeated exposure to mutual listening, validation, and emotional disclosure, participants appeared to develop less defensive and more trusting relational positions.

Importantly, the sense of connection described by participants was not limited to explicit support or encouragement. For some, simply recognizing common emotional difficulties in others reduced feelings of abnormality and enabled forms of self-reflection that had previously been inaccessible.

“It reassured me to find this sharing, to see that other people also often don’t recognize or can’t express certain feelings.” (Participant 4).

The group therefore appeared to operate not merely as a setting for symptom discussion, but as a social environment in which recovery could be reimagined as fundamentally relational rather than purely individual. Participants increasingly constructed change as something emerging through emotional contact, mutual recognition, and the experience of belonging to a shared human struggle. At the same time, these narratives suggest that the therapeutic effects of the group were linked less to direct behavioural intervention than to the creation of interpersonal conditions that made new meanings possible. Feeling understood by others with similar experiences appeared to reduce shame, increase openness, and support the development of more compassionate and socially connected understandings of both illness and recovery.

The group as a space of relational exposure: safety, shame, and emotional risk

Although many participants described the group as supportive and meaningful, their accounts also revealed that participation involved considerable emotional vulnerability and interpersonal risk. Entering the group often meant confronting fears of exposure, judgment, and inadequacy, particularly in relation to speaking openly in front of others who shared similar struggles. Rather than being experienced as immediately safe, the group frequently emerged as a psychologically demanding space in which participants were required to negotiate visibility, self-disclosure, and emotional closeness.

Fear was especially prominent during the initial phases of participation and was commonly linked to anticipated negative evaluation from others.

“I was afraid to speak openly… worried they’d think I was crazy.” (Participant 9).

Feelings of embarrassment and self-consciousness further contributed to participants’ sense of interpersonal vulnerability. For some, the prospect of discussing intimate emotional experiences in a group context activated concerns about being exposed, misunderstood, or perceived as inadequate.

“Maybe I’m not the only one who feels a bit embarrassed at certain moments.” (Participant 10).

These accounts suggest that the group did not simply provide connection and validation; it also activated interpersonal anxieties that mirrored broader relational difficulties commonly associated with EDs, including shame, fear of judgment, and difficulties tolerating emotional visibility. In this sense, the therapeutic group appeared to function as a context of relational exposure, in which participants encountered forms of interpersonal closeness that were both desired and threatening.

Importantly, some participants described a gradual transformation of these initial fears through ongoing participation in the group. Experiences of surprise were particularly significant in this regard, as participants contrasted their anticipatory anxiety with the discovery of a more accepting and emotionally containing environment than expected.

“At first, I was sceptical and afraid I would be negatively influenced. Instead, I found a very welcoming environment…” (Participant 3).

This movement from anticipated threat toward partial experiences of safety and belonging appeared to emerge through repeated encounters with shared vulnerability and mutual recognition. Listening to others’ experiences often reduced participants’ sense of isolation and facilitated feelings of emotional connection.

“I like being in the group, even if I don’t speak sometimes—it still helps me to listen to others. I recognize myself in many of the things they say, and that gives me positive support. Coming to the group gives me a feeling of serenity.” (Participant 1).

However, participants’ accounts also indicate that emotional safety remained fragile and incomplete rather than fully achieved. For some, hearing others’ suffering intensified emotional distress rather than alleviating it, highlighting the emotionally contagious and sometimes overwhelming nature of shared pain.

“Hearing others’ pain made me realize I’m not alone… but it was not comforting.” (Participant 4).

Similarly, a few participants described experiencing pressure or constraint within the group setting, particularly regarding expectations to speak or emotionally engage.

“In the group I felt embarrassed, almost forced to talk.” (Participant 2).

These narratives complicate idealized understandings of group cohesion by suggesting that participation itself may evoke forms of tension, discomfort, and emotional overload. The group was therefore not experienced uniformly as safe or reparative; instead, participants appeared to oscillate between connection and withdrawal, recognition and self-protection.

Feelings of guilt further reflected participants’ sensitivity to the emotional impact of the group on others. Some worried that sharing painful experiences might burden or negatively affect fellow members, revealing heightened awareness of interpersonal responsibility within the group dynamic.

“I worried heavy topics might hurt someone already struggling.” (Participant 9).

In terms of behavioural impact, most participants reported moderate or limited changes in their eating-related behaviours. Some felt that the indirect approach to symptom discussion experienced in the group limited its effect on concrete behavioural strategies and voiced dissatisfaction with the limited focus on behavioural coping strategies:

“Often the focus is on the ‘background’ of the problem, on its deeper causes, but for someone who is suffering, the central thought is the one that is most difficult to manage.“(Participant 3).

“We never really talked about eating disorders… so I keep doing what I always did.” (Participant 6).

From an analytic perspective, these accounts suggest that the group may function primarily as a meaning-making and relational intervention, with indirect or delayed behavioural effects, rather than as a skills-based symptom-management format. The therapeutic value of the group may have resided not in the absence of difficult emotions, but in the possibility of encountering and negotiating them within a shared relational context. Safety did not appear to function as a pre-existing condition of the group; rather, it emerged as something gradually co-constructed through repeated experiences of emotional exposure, mutual tolerance, and interpersonal survival. The group thus operated simultaneously as a source of support and a site of emotional challenge. By confronting participants with experiences of visibility, vulnerability, and relational uncertainty, it created opportunities for new interpersonal meanings to emerge.

Discussion

This study employed a qualitative approach to explore how individuals with EDs conceptualized illness and recovery before and after participation in a group therapy, as well as the perceived emotional, relational, and behavioral impact of the group experience. Overall, participants retrospectively described a contrast between their pre-group and post-group understandings of illness and recovery from a predominantly symptom-focused and externally oriented understanding of illness toward a more emotionally and relationally informed perspective. Recovery was retrospectively defined almost exclusively in symptomatic terms whereas, following a group experience integrated within a multidisciplinary treatment, it increasingly encompassed emotional awareness, empowerment, social reconnection, and self-acceptance. Group therapy was perceived as having a stronger impact on emotional and interpersonal domains than on eating-related behaviors [15, 17].

Before entering group therapy, participants predominantly framed their illness in terms of ED symptoms and external determinants, often neglecting emotional and relational dimensions underlying ED psychopathology [27]. This finding aligns with previous quantitative studies describing marked difficulties in emotional awareness and expression [28], as well as reliance on avoidance and inhibition strategies in individuals with EDs [29]. Such narratives are consistent with the limitations of symptom-centered diagnostic systems, which may inadvertently reinforce a narrow understanding of illness [5]. Following the group experience, participants increasingly described the ED as a maladaptive strategy for managing emotional distress and interpersonal difficulties: the ED became progressively re-signified from a problem located in the body to a meaningful expression of emotional and relational suffering. This shift toward emotional and relational awareness echoes findings from research showing that therapeutic progress in EDs is often accompanied by greater recognition of affective processes and attachment-related dynamics [30]. Importantly, participants reported a reduced tendency to attribute causality exclusively to external pressures and an increased sense of personal responsibility for change. This transition from external to internal locus of control has been identified as a key process in recovery-oriented models of EDs [31].

Consistent with prior research and diagnostic manual systems, recovery was initially equated with symptom remission, weight normalization, and freedom from food-related preoccupations [32]. After participation in group therapy, recovery was increasingly described as a broader, multidimensional process involving emotional regulation, empowerment, self-acceptance, and improved social functioning. These findings strongly align with the concept of personal recovery, which emphasizes meaning-making, agency, and relational connectedness beyond symptom reduction [12]. More broadly, these findings support a view of recovery as a dynamic and context-dependent process, shaped by individual meanings as well as relational experiences [13, 33]. In this sense, recovery appears not only as an individual trajectory but also as something that is negotiated within interpersonal contexts. Our findings extend this perspective by suggesting that group therapy may actively contribute to this process, providing a space in which meanings of illness and recovery are shared, challenged, and redefined. This highlights the potential role of group processes in shaping recovery beyond individual-level change.

The increased recognition of emotional and relational dimensions of the disorder observed after group therapy reflects a broader process through which patients begin to reclaim agency and self-reflection [34]. The emergence of empowerment and self-mastery themes in post-group narratives suggests that group therapy may foster a sense of personal agency—an aspect closely tied to control, identity formation, and the internal negotiation of the anorexic voice—yet often underrepresented in traditional outcome measures for EDs [35]. Importantly, the study findings suggested that recovery was experienced not as a linear movement away from illness, but as a complex process of identity renegotiation. The illness remained emotionally significant even as participants began to establish greater psychological distance from it, suggesting that recovery involved learning to coexist with ambivalence while progressively reclaiming forms of self-definition not wholly dependent on the disorder. This aligns with previous qualitative studies emphasizing that a key aspect of recovery is developing a sense of identity beyond the ED and suggesting that successful treatment involves not only symptom improvement but also the experience of personal agency, personal growth, and the reconstruction of a meaningful self-concept [36].

Participants consistently highlighted the relational nature of group therapy as a central component of their experience, particularly the value of shared experience and peer support. Participants reported that being understood by peers with similar experiences reduced feelings of isolation and promoted emotional expression, supporting previous qualitative evidence that group settings offer a particularly validating environment for individuals with EDs [37], although they require careful facilitation to manage intense emotions and member dynamics [38]. These relational processes may reflect the development of group cohesion, a core mechanism in group psychotherapy that has been consistently associated with improved therapeutic outcomes [19]. This experiential validation appears to support emotional insight and self-reflection rather than direct behavioral change, suggesting that group therapy may operate primarily through indirect psychological pathways. From a broader perspective, these findings suggest that group therapy may function less as a direct behavioral intervention and more as a context for meaning-making and relational reorganization. The changes described by participants point to a shift in how they interpret their experiences, rather than to immediate symptom resolution, highlighting the importance of targeting underlying psychological and interpersonal processes in treatment. Activation of these psychological processes may enhance the therapeutic work undertaken in individual psychotherapy, suggesting that group therapy may serve as an adjunct to standard treatment in facilitating recovery. The prominence of peer support and shared meaning-making supports the validity of therapeutic factors indicated by Yalom and Leszcz [18] such as universality, group cohesion, and interpersonal learning. Similar mechanisms have been observed in group-based interventions for other severe mental disorders, where relational safety and mutual recognition were essential for therapeutic change [39]. The observed impact of group therapy on emotional and interpersonal domains is also consistent with prior qualitative and narrative reviews of group interventions in EDs [15, 17].

Despite the overall positive appraisal of group therapy, fear of judgment and embarrassment remained salient emotional experiences for some participants, particularly in relation to express themselves within the group. This finding reflects previous literature highlighting shame, social comparison, and fear of evaluation as central maintaining factors in EDs [40, 41]. Group settings may therefore activate threatening interpersonal dynamics, especially in patients with high levels of interpersonal sensitivity; however, they may simultaneously be therapeutic by constituting a safe space in which individuals can be exposed to and work through these dynamics. An ambivalent nature of the group process emerged from participants’ accounts: healing appeared to be closely intertwined with emotional risk, and the group became meaningful precisely because it allowed difficult relational experiences to be enacted, shared, and, at times, transformed in the presence of others. In this sense, therapeutic change may have occurred not through the elimination of shame or fear, but through participants’ increasing capacity to remain emotionally present in relationships despite these experiences.

Clinically, these findings underscore the importance of therapists actively monitoring group processes and explicitly addressing fear of judgment and perceived pressure to participate arising from peers or the therapist. Clinicians should not consider safety as an a-priori condition of the group, but something that should be co-constructed through repeated experiences. Establishing psychological safety and validating ambivalence toward disclosure may be crucial to prevent disengagement and maximize the therapeutic potential of group interventions [18]. Furthermore, group-based interventions may be especially useful in facilitating processes that are less accessible in individual therapy, such as shared meaning-making, interpersonal recognition, and the development of a sense of belonging. Integrating these elements more explicitly into treatment planning may help support broader recovery processes beyond symptom reduction. However, the need for additional behavioral strategies to manage eating behaviors has been highlighted, and clinicians are advised to work with patients to enhance their coping skills.

The present findings support a growing body of evidence suggesting that meaningful change in EDs extends beyond symptom reduction and involves emotional, relational, and identity-related processes. By foregrounding patients’ perspectives, this study complements quantitative research outlining putative mechanisms of change and reinforces calls for more person-centered outcome frameworks in ED treatment [30, 42].

Limitations of this study need to be acknowledged. First, researchers’ positionality inevitably shapes the research process [43]. Consistent with this, participants’ narratives may partly reflect internalization of the therapeutic language and conceptual framework. As such, the themes identified may represent, at least in part, the assimilation of the group’s explanatory model rather than entirely spontaneous reformulations. Second, the sample size is modest, although this aligns with most studies conducted in EDs through a qualitative method, and participants were recruited from a single clinical site, which may limit the transferability of the findings. Third, the retrospective nature of the interviews may have influenced how participants reconstructed their experiences, particularly in relation to perceived changes over time. Fourth, the absence of a comparison group receiving treatment as usual without adjunctive group therapy limits the ability to differentiate the effects of the various treatment approaches. Fifth, although this was not the aim of the study, the lack of characterization of participants’ symptom levels before and after the group, limits the possibility of examining the impact of group therapy on symptomatology. Sixth, the interview guide was developed to address the research aims and was not formally pre-tested. Finally, the sample included some variability in clinical and demographic characteristics, which may influence how illness and recovery are conceptualized. Future research should explore these differences more explicitly. Mixed-methods and longitudinal qualitative designs may be particularly valuable in clarifying the mechanisms through which group therapy contributes to recovery. Future research should further investigate how the observed relational and emotional processes evolve over time and how they interact with symptom change.

Conclusions

Participants’ narratives indicate that group therapy integrated into multidisciplinary treatment promotes a transition from a primarily symptom-focused view of illness to one emphasizing emotional and relational aspects, with notable group benefits for emotional awareness, interpersonal functioning, and self-acceptance. These effects may accelerate psychotherapeutic processes potentially serving as mechanisms of change facilitated by group therapy, although their relevance for clinical outcomes remains to be explored.

Author contributions

M.C.: Led the study conceptualisation and design, conducted data collection and analysis, and drafted the manuscript.F.M., G.P., V.R.: Contributed to data collection.V.L., G.L., F.R., E.B.: Contributed to data analysis and interpretation of the findings.A.C.: Contributed to study conceptualisation and manuscript drafting.A.M.M.: Contributed to study conceptualisation, supervised the study, and contributed to manuscript drafting and revision.All authors contributed to manuscript revision and approved the final version.

Funding

Not applicable.

Data availability

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

Declarations

Ethics approval and consent to participate

The study protocol (n. 0034565/i) was approved by the local Institutional Review Board.

Consent for publication

Consent to publication was gained for the data collected from all participants.

Competing interests

The authors declare no competing interests.