Experiences of personal recovery in the context of unmoderated online communities centered around self-harming and suicidal behavior: a secondary qualitative analysis

Abstract

Background

Participation in online communities centered around self-harming and suicidal behavior is a complex phenomenon characterized by potentially harmful social dynamics, though it may also involve support. While previous studies have explored peer support within these communities, less is known about how engagement may influence personal recovery and the functions these communities serve. Therefore, this study explored experiences of personal recovery in the context of participating in unmoderated online communities centered around self-harming and suicidal behavior.

Methods

This qualitative deductive–inductive secondary analysis re-examined interview data originally collected to explore peer support exchanges within online communities centered around self-harm and suicidal behavior, using reflexive thematic analysis informed by the CHIME-D (Connectedness, Hope, Identity, Meaning, Empowerment, and Difficulties) framework to examine personal recovery in this context. The dataset consisted of semi-structured interviews with twelve women and one non-binary participant (aged 21–49 years) from all five Danish regions, all with lived experience of psychiatric disorders and previous psychiatric admissions.

Results

Participants described how engagement supported personal recovery through mutual recognition, reciprocity, and shared experience. They also described tensions between their own experiences of these communities as meaningful recovery spaces and professionals’ predominantly risk-oriented interpretations, contributing to stigma and concealment. Recovery was experienced as complex and ambivalent, involving not only support but also ongoing attachment, emotional burden, and difficulty disengaging. Participants also described challenges related to moral responsibility towards others, exposure to distressing content, and balancing recovery with ongoing participation. Examining recovery in digital self-harm networks highlights important nuances in understanding the roles and functions these networks may serve.

Conclusions

Applying the CHIME-D framework provided a nuanced understanding of how online communities centered around self-harming and suicidal behavior may contribute to personal recovery, advancing our understanding of the functions they serve and what motivates participation. The findings highlight recovery as a non-linear process shaped by both supportive and challenging dimensions of online engagement and underscore the importance of recognizing these communities as meaningful, though double-edged, contexts for recovery.

Introduction

Historically, the recovery movement emerged as a rebellion and opposition to a biomedical system and its approach to mental health [1, 2]. However, a clear consensus has yet to be reached as to what recovery as a concept really is and contains [3] and recovery is variously described as a model, an experimental field of knowledge, a philosophy, a vision, an ideology, a slogan, and as a movement [4, 5].

However, recovery can be understood through three distinct concepts [6]: (1) clinical recovery, which refers to a reduction of symptoms below a certain threshold; (2) functional recovery, which pertains to functioning at a level above a defined threshold and; (3) personal recovery, described by Anthony [7] as “a deeply personal, unique process of changing one’s attitudes, values, feelings, goals, skills, or roles.”

While Anthony conceptualized recovery as a deeply personal process [7], later frameworks such as CHIME emphasize that recovery also unfolds through relational and contextual processes through the components Connectedness, Hope, Identity, Meaning, and Empowerment as developed by Leamy et al. [8]. With the later addition of the component Difficulties [9], the framework explicitly acknowledges the ambivalence, setbacks, and complexities inherent in many recovery journeys [3, 10]. Given that participation in online communities centered around self-harming and suicidal behavior may simultaneously foster both positive and harmful outcomes [11], CHIME-D provides an analytical lens for understanding how recovery-oriented processes unfold within these online communities.

Despite the growing conceptual clarity offered by CHIME-D [9], less is known about how these recovery processes unfold in informal, peer-driven contexts beyond formal mental health services such as online peer-supportive communities known to take place on social media platforms through shared experiences in many areas [12, 13]. Unmoderated digital environments centered around self-harming and suicidal behavior constitute a particular type of online communities.

Exchanging peer support in online communities centered around self-harming and suicidal behavior is a double-edged phenomenon. On the positive side, there is evidence that participation in these networks supports empathy, help-seeking and reduces loneliness [14, 15], but on the negative side, there is also evidence that participation is associated with increased rates of self-harm, suicidal thoughts and suicide attempts [16–18], thus it seems that what is beneficial for some is harmful for others [11, 13, 19, 20].

While research has examined both the positive and negative effects of participating in online communities centered around self-harming and suicidal behavior [11, 19], there is limited understanding of how such engagement may affect the process of personal recovery, including how recovery-related processes are negotiated and enacted and which functions participation in these communities serve for the individual.

This study emerged from insights gained during our earlier analysis of peer support in online communities centered around self-harming and suicidal behavior [21]. In that work, we observed how participants not only sought and provided emotional support but also engaged in rich discussions about coping, change, and personal growth. These observations pointed to the need for a focused exploration of how recovery is understood, negotiated, and experienced within these peer-supportive online spaces, and an opportunity to contribute further nuance to the understanding of individual motivations for participation.

Building on this, our aim is to explore how engagement in unmoderated online communities centered around self-harming and suicidal behavior shapes individuals’ personal recovery processes. By examining recovery as it unfolds in such digital peer-support contexts, this study contributes new empirical knowledge about how recovery-oriented processes are lived, shared, and co-constructed outside formal, professionalized mental health services.

Method

The current study constitutes a secondary qualitative analysis of data collected for our previous study [21], using personal recovery as its central theoretical framework to address a distinct research question concerning recovery processes in unmoderated online communities centered around self-harming and suicidal behavior.

Participants

Participants were eligible for inclusion if they: (1) had experience engaging in online communities centered around self-harming and/or suicidal behavior, (2) were fluent in English or a Nordic language, (3) were willing to provide informed written consent, and (4) were 15 years of age or older. Recruitment occurred from April 2024 to July 2024, in both in- and outpatient psychiatric facilities across four of the five Danish regions, in collaboration with residential institutions nationwide, and with support from three Danish NGOs that shared recruitment advertisements on their social media platforms.

The number of participants was assessed continuously during recruitment in relation to information power [22] by MEH, JM and KJ. Rather than aiming for saturation, sufficient information was indicated by the specificity of the sample and the depth of participants’ recovery-related accounts. The estimation of sample adequacy was guided by Malterud et al.’s concept of information power [22], whereby the relatively narrow study aim and the use of the CHIME-D framework to inform both data collection and analysis led us to assess that the sample size of 13 participants we included in the study was appropriate; 12 females and one non-binary, mean age 28 and predominantly diagnosed with attention-deficit disorder, personality disorders, autism disorders and schizophrenia, living in 5 Danish Regions. Characteristics are presented in detail in Table 1.

Table 1.

Participant characteristics (n = 13) – previously published [21]

Characteristics	N=
Age	21–49 years (average 28 years)
Sex
female nonbinary	12 1
Region of Residence
Capital Region of Denmark Region Zealand Region of Southern Denmark Central Denmark Region The North Denmark Region	6 3 2 1 1
Living condition
Living alone or with a spouse Residential Home	9 4
Prior admissions in psychiatry
yes no	13 n/a
Current physical self-harm
Yes No	7 6
Prior suicide-attempts
Yes No	12 1
Diagnosis*
Attention-deficit disorders Personality disorders Autism disorders Schizophrenia Anxiety Bipolar disorder Eating disorders PTSD Adjustment Disorder OCD Relapsing psychosis Schizotypal	5 4 4 4 3 2 2 1 1 1 1 1
Education/work status
Currently studying or having completed a bachelor’s or master’s degree and was working Studied or had completed a bachelor’s or master’s degree but was on sick leave or unemployed Early retirement Unemployed, living from social services Flexible work	5 2 3 2 1
currently active in online communities, centered around self-harming and suicidal behavior.
Yes No	10 3

*Some participants have multiple diagnoses

*n/a = non-applicable

Ethics

The study was conducted in accordance with the Declaration of Helsinki and was approved by the Capital Regions protection agency PRIVACY (Approval number P-2022-41). Further approval was not required, following Danish legislation.

Before conducting the interviews, all participants were provided with both oral and written information detailing the study’s purpose and methods, as well as their rights. Informed consent was obtained in writing, either physically or electronically. Given the need for special consideration when involving vulnerable groups, the participant information and consent forms were developed in accordance with current Danish legislation. Further, given the sensitive nature of the topic, procedures were in place to manage potential distress during interviews, i.e. Follow up with the participant, supportive conversation, possibly give the number of the psychiatric emergency room.

Interviews

The semi-structured interview guide was developed for a prior study on peer support in these online communities [21], informed by existing literature and Shery Mead’s conceptual framework [23], and refined through feedback from patient representatives. The full interview guide is available in the original publication [21]. Participants were informed about the first author’s background, including age, sex, educational qualifications, and motivations for conducting the study. All participant contact was managed exclusively by the first author, who had no prior relationship with the participants. The interviews were conducted by the first author, a Postgraduate Mental Health Nurse Specialist with a Master’s Degree in Nursing Science and more than ten years of clinical and research experience.

Participants were given the option to select their preferred interview format, which included face-to-face meetings (n = 2), online platforms (n = 6), telephone calls (n = 2), or written correspondence (n = 3). The audio recordings from the online and face-to-face interviews were transcribed verbatim by the first author, while written interviews were directly saved. Audio recordings and transcripts were stored in a secure, access-restricted folder, accessible only to the authors MEH and JM. To ensure anonymity, all names and other identifying information were removed before analysis, and all materials were permanently deleted upon completion of the analysis.

The interviews lasted between 44 and 108 min, resulting in 249 pages. Potential differences in depth across interview formats were addressed by applying the same coding and analytic procedures across all data, with analytic emphasis placed on the content of participants’ accounts rather than interview length or format.

Analytical process

This analysis applied the CHIME-D framework [3] as a reflexive analytical lens to the dataset, hence shifting the analytical focus from peer-support exchange to personal recovery processes. The CHIME-D framework conceptualizes personal recovery through six interrelated domains: Connectedness (developing supportive relationships and social inclusion), Hope (maintaining optimism about the future), Identity (rebuilding a positive sense of self), Meaning (finding purpose and meaning in life), Empowerment (developing agency and self-determination), and Difficulties, acknowledging the ongoing challenges and setbacks that may accompany recovery [8, 9].

The analysis followed Braun and Clarke’s six-phase reflexive thematic analysis employing a deductive–inductive hybrid approach. In the first phase MEH engaged with the transcripts, imported to NVivo, through repeated reading and listening to the interviews, with a focus on experiences of personal recovery. During the second phase, coding was conducted using the CHIME-D framework as a theory-informed analytical structure. Data extracts were sorted in relation to the CHIME-D domains (Connectedness, Hope, Identity, Meaning, Empowerment, and Difficulties), while remaining attentive to nuances, overlaps, and tensions within participants’ accounts.

Data extracts were coded in relation to the CHIME-D framework, with each extract assigned to one primary domain and, where relevant, a secondary domain. The analysis was conducted using a reflexive thematic approach, where coding was both guided by the CHIME-D framework and shaped through an ongoing interpretative engagement with the data. This allowed for attention to how recovery-related processes were expressed in diverse and sometimes partial ways across participants’ accounts.

In phases three and four, subthemes were developed inductively within and across the CHIME-D domains by identifying patterned meanings in the coded material. The themes were iteratively reviewed and refined in dialogue with the empirical data and the theoretical framework.

In phases five and six, themes were defined, named, and situated in relation to recovery theory, resulting in an interpretive account of the experience of personal recovery in the context of unmoderated online communities centered around self-harming and suicidal behavior.

Analytic interpretations were continuously discussed within the research team to critically examine assumptions related to these online communities and personal recovery. An example of the analytical development from coded extracts to themes is presented in Table 2.

Table 2.

Example of the analysis process

Coded text (examples)	Chime D primary domain	Chime-d secondary domain	Subtheme	Theme
There is the issue of confidentiality. Those who have been there for many years have reached the point where they consider, if we can see that we are concerned about her, then we must call.’ And then we must disregard confidentiality. There are many who agree that if people's lives are in danger, we cannot just sit and watch. We have to act. We cannot do that. We have to do something. We cannot sit with the guilt afterwards, that nothing was done. Participant 7 I was there for her and would be happy to chat with her throughout the evening to prevent her from harming herself. And then it actually ended up that she didn't harm herself and got through the evening because i was there and made a plan with her and stuff like that, you know?Participant 1people may share something they have done, not necessarily because they want to inspire others to do the same, but because they don't think about how they might influence others. They just need to express themselves to someone who understands. but it can influence others unintentionally. I don't believe that anyone is doing it with bad intentions or consciously wanting to hurt others, but i think it's the unconscious aspect of it, that people aren't aware of it.participant 12	Meaning Connectedness Meaning	Connectedness Hope Difficulties	Issues on confidentiality Disregard confidentiality Feeling of guilt Reaching out to others Providing support Unintentional impact of sharing Influencing others	Finding purpose and identity through mutual care

Results

The analysis led to the development of three themes presented below.

Theme 1: Seeking peer communities in response to perceived caution from professionals

Participants described how professionals in psychiatric settings and supported housing often construed their engagement in online communities centered on self-harm and suicidal behavior as indicative of deterioration, rather than as a potential component of personal recovery. Consequently, participants explained how they were typically advised by health professionals to delete profiles or restrict their participation due to concerns that such engagement might exacerbate self-harming behaviors or suicidal thoughts. This framing was experienced as mistrust, which, in turn, appeared to strengthen participants’ motivation to seek out and engage more actively in these online peer communities.

So, I had that profile for a while, and then one of the staff found out and said, ‘You need to delete that profile. It is not good for you. Participant 1

As such, participants described how health professionals warned against participation, discouraging the exchange of contact details and forming friendships both online and offline. Participants reflected on how these restrictions shaped their sense of agency and their ability to explore coping strategies within social networks. They explained that they felt that their involvement was seen as something to be hidden and to be ashamed of and that there seemed to be no understanding of their need to be surrounded by peers in order to feel less alone while trying to manage difficult periods in their lives.

When you are hospitalized. … I experience this thing where the staff try to… They say, ‘Now you should not share your Facebook, now you should not share your Instagram. You should just be here together. You should not come up with anything to do outside of the psychiatric ward, because now you both have this and that, and you will trigger each other. Participant 8

Participants recounted not experiencing curiosity or recognition from professionals regarding the supportive aspects of their online engagement, such as feeling understood, gaining insight, or discovering strategies to manage distress. The absence of acknowledgment sometimes led participants to conceal their involvement.

I have opened up a little about having been in those kinds of environments. But then people don’t know what it is. And if they do know what it is, they are so negative about it… Really negative in general… It’s really just bad for everyone, they say. Participant 2

At the same time, having to justify or conceal their participation prompted some participants to more clearly articulate for themselves how these communities contributed to their personal recovery and what they gained from them. Participants further described how they felt that societal narratives and media portrayals of risks associated with participating in these online communities influenced professional attitudes, which participants felt reinforced stigma and as a result limited open discussion about the positive potential of exchanging peer support.

It is always the negative aspects that are communicated. No one ever communicates that we are actually very committed to supporting and motivating each other. And saying, ‘Hey, try talking to the staff instead of sitting there thinking about self-harm or something else.’ It is never mentioned that we actually support and motivate each other in there. It is always the dark side that gets mentioned in the media and so on. I think that is a real shame. Because the staff thinks that it is all so bad, because it is all they ever hear. Participant 7

Ultimately, participants described how professional scrutiny of their online engagement contributed to feelings of shame and concealment. The perception that their online participation was something to be hidden, monitored, corrected, and not understood reinforced a sense of being “less than” and of having their personal recovery subject to external evaluation.

Now that I have an education and a completely normal, well-functioning everyday life. So, should I just delete my profile? Should I just leave it at that? I do not think so. Recovery is just a huge thing. It really is… I mean, I would never call it that myself, but that is really what I have been through with this network, even though people do not understand it. Participant 10

In this way, perceived external caution became part of the process through which participants strengthened their sense of agency and ownership of recovery.

Participants further described what they perceived to be a real-life stigma existing around self-harm and suicidal behavior, and how this extended into perceptions of the online communities they engaged in and led to their participation to be misunderstood or devalued. This perception influenced not only whether and how they chose to engage but also shaped the ways in which these spaces could support their personal growth and coping strategies. Some participants described that the health professionals encouraged them to engage with the peer-worker within the mental health system instead of participating in these online communities. However, the participants explained that the peer-workers were perceived to predominantly be a part of the staff, rarely available and they rarely felt that they could mirror themselves in the peer-worker, the same way they could with peers in the online communities.

I felt the peer workers were more part of the staff than they were part of us who were admitted. And that distance meant that I never really bonded with anyone. There is not a peer worker that I look back on and think that she made a difference. I think I have been in contact with three, and it has been the same experience three times. Participant 10

These experiences led participants to reflect critically on what kinds of relationships actually supported them sharpening, their sense of what was helpful, meaningful, or safe in their recovery process.

Theme 2: Learning to navigate connection, dynamics and boundaries

Participants recounted how engagement in online communities involved continuous navigation of relational dynamics, emotional exposure, and personal boundaries. Participation was described as an active and ongoing process in which participants adjusted their level of engagement depending on their emotional state, relational context, and perceived risks. Rather than being a passive form of participation, engagement required ongoing personal decisions about when to be present, how to interact, and when to withdraw.

If I am having a tough time or an especially hard day or something like that, then I definitely do not go on there. And I have also seen some of the others write, “I’m feeling really down; I’m still keeping up, but I’m not really posting much or something like that.” And I actually do that myself—even if I follow 200 people or something, I only check in on the ones I know a little better than the others. And then… I will write if I have the energy. Participant 1

This illustrates how participants actively regulated both the intensity and the form of their engagement. Withdrawal was not necessarily a sign of disengagement, but rather a deliberate strategy to maintain involvement without becoming overwhelmed. Similarly, selective interactions, such as focusing on specific individuals, was described as a way of managing emotional exposure while preserving meaningful connections.

At the same time, participants emphasized that engagement could be both supportive and potentially harmful, which further reinforced the need for ongoing self-monitoring and self-protection.

I can see that there are unhealthy aspects to it as well. You might, maybe a bit unconsciously, end up reinforcing the negative. You really have to be careful. You have to remember to take care of yourself, but that applies more generally to social media as well….Participant 12

This awareness of potential harm appeared to inform a range of concrete strategies aimed at regulating participation. Rather than withdrawing entirely, participants described how they shaped their engagement in ways that aligned with their own needs and limits.

Those who share pictures with blood or where they are in restraints, I have actively unfollowed them, because I think that is too much. And I do not really think that belongs on Instagram. Participant 6

Here, participants described actively curating their social environment by selecting which profiles to engage with and which to avoid. Such practices were not only about preference, but also about managing exposure to potentially triggering content and maintaining a sense of psychological safety within the community.

In addition to regulating what they engaged with, participants also described the importance of stepping back entirely at times.

To begin with, I announced in there that I was taking a break and that I was deactivating my profile. And then it has just been like… I could feel that it was good for me not to have to worry about others all the time and not having to relate to everyone else who is also struggling. Participant 7

Taking breaks or temporarily withdrawing from the community was described as a necessary strategy for managing the emotional demands associated with ongoing exposure to others’ distress. In this way, disengagement was described to function as a form of self-care rather than avoidance.

Participation also involved ongoing regulation of self-disclosure. Participants described being attentive to how their posts might affect others and how they themselves might be perceived, which led them to modify language, limit what was shared, or withhold certain experiences altogether.

I feel that if I share too much, it could be used against me. So, I think very carefully about what I write and how I write it, especially if I know that someone from my residential facility or my therapists might see it. Participant 6

This form of self-regulation reflects an awareness of multiple audiences and potential consequences, both within and outside the community. While such considerations enabled participants to remain engaged without exposing themselves to risk, they could simultaneously constrain opportunities for openness, emotional expression, and validation.

Overall, participation in online communities involved an ongoing process of boundary work and self-regulation. Through continuously negotiating when to engage, what to share, and how to relate to others, participants developed a more differentiated understanding of their own needs, limits, and relational preferences, which formed an integral part of their recovery processes.

Theme 3: Finding purpose and identity through mutual care

Participants described how their engagement involved a continuous negotiation of responsibility, both toward themselves and others. Taking part in these spaces often brought a heightened sense of awareness about how one’s words and actions could affect others, as well as reflections on personal limits, values, and coping abilities, and participants began to see themselves not only as recipients of support, but as active contributors within these communities.

I find that others inspire me to keep fighting on bad days, but I also give others hope. If there is a young girl out there who thinks she has messed everything up because she has scars and her children will ask questions, there is a way forward! And I am happy to show what it was like for me, with all the obstacles along the way. Participant 6

Participants expressed feeling accountable for the well-being of other users, particularly when someone posted about acute distress, self-harming behavior, or suicidal thoughts or actions. In such moments, participants described uncertainty about how to act and how much responsibility to take. Some reached out privately to offer emotional support, while others contacted emergency services or family members when they feared for someone’s safety, despite the acknowledgement of an informal confidentiality agreement within these online communities.

I’m doing this to look after you and to look after myself and the hospital department, but you can’t see that from where you are right now. … Even though there is a duty of confidentiality, I feel: Yes, yes, that is fine. I also have a duty of confidentiality at my job, but I must break it if something serious happens. I have probably brought that with me into my private life. I feel that there have to be exceptions, and I can’t just ignore the fact that someone could potentially die from this stuff. Participant 5

Participants explained that breaking confidentiality could lead to mixed reactions. Some were met with gratitude or understanding, while others experienced criticism or loss of trust. Despite these challenges, participants reflected that acting from care and moral responsibility provided them with a sense of meaning and integrity. Supporting others, they explained, often reinforced their own coping abilities, and strengthened their belief in their capacity to make a difference. As such, these dilemmas were not only experienced as burdens but also as opportunities for participants to act in accordance with their own values. Participants described how such moral engagement strengthened their sense of integrity and reinforced a developing identity as someone capable of making a positive difference.

As part of this process, participants also described learning to navigate emotional boundaries and regulate their engagement to protect their own mental health. Participants described taking breaks, limiting time online, or avoiding distressing content by unfollowing certain profiles as active strategies to maintain balance and sustain their participation on their own terms.

I have also had periods where I have taken breaks from the network because I felt it was too much. Then I just log out for a while, and then I can return, when I have more energy. Participant 1

Such adjustments were described as part of a personal learning process, thus developing awareness of what helped and what burdened them. Participants explained that experiences of mutual support and reciprocity fostered a growing sense of empowerment. Helping others and receiving care in return were described as meaningful experiences that contributed to emotional stability, strengthened participants’ sense of identity and competence, and enhanced their ability to care for themselves.

I have to look after myself as well… I care about the others, but I also know when I need to step back, so I do not get overwhelmed. Participant 6

Overall, participants described that being part of these online communities supported their development of empathy, self-understanding, and a sense of purpose through reciprocal engagement with others. For some, however, recovery involved gradually stepping back from these spaces. After establishing supportive offline relationships and accessing professional care, continued participation was experienced as maintaining a connection to an illness-oriented identity. Others in similar situations chose to remain, describing their engagement as an ongoing and meaningful part of how they sustained their well-being.

When I posted that I had gone a year without self-harming, I got a message from a girl saying, ‘Wow, you’re so cool.’ She also wrote to me, ‘You’re a huge inspiration to me, and the fact that I can follow your journey gives me real hope.’ And then I just feel confirmed that I am doing the right thing in sharing both my victories and my struggles. Participant 7

Being able to inspire others or serve as a role model, even in modest ways, was described as deeply rewarding. Participants explained that when their stories resonated with others, it confirmed that their struggle had not been in vain. These moments, when they inspired others, also strengthened the sense of solidarity, and participants described how this mutual recognition helped them to get through times of crisis.

Discussion

This study explored personal recovery processes in the context of engagement in online communities centered around self-harming and suicidal behavior. The findings suggest that personal recovery, in the context of online peer communities, unfolds as a dynamic and relational process that cannot be meaningfully separated into distinct domains. Rather than aligning with discrete components of the CHIME-D framework, participants’ experiences reflected overlapping and interdependent processes, where connectedness, identity, meaning, empowerment, and difficulties were continuously negotiated. This highlights the need to understand recovery not as a set of isolated elements, but as an integrated and context-dependent process shaped through ongoing social interaction. While the CHIME-D framework provided a useful analytical lens, the findings also suggest that recovery processes in this context are not easily contained within its predefined domains. In particular, participants’ experiences point to forms of connection and identity that were not solely supportive or recovery-oriented, but also involved ongoing attachment, emotional burden, and at times difficulty disengaging from these communities.

This perspective aligns with conceptualizations of recovery as an ongoing, non-linear process that does not require symptom remission, but instead involves rebuilding an individual sense of self, meaning, and control despite the continued presence of mental illness [24].

Participants described how their involvement was often met with professional scrutiny and restrictions, how stigma and self-censorship shaped the ways they engaged and communicated, and how responsibility and reciprocity within these communities had the potential to contribute to personal growth. Together, the findings highlight both the challenges and the transformative potential of such online peer engagement as part of the broader recovery process. Rather than representing separate dimensions, these processes often co-occurred within the same interactions, suggesting that recovery in this context is inherently ambivalent. This challenges the implicit assumption within frameworks such as CHIME-D [9] that domains like connectedness, identity, and empowerment can be understood as predominantly beneficial or directional.

While participants described these communities as meaningful and supportive, it is important to consider that such experiences may not fully capture all dimensions of engagement. As in other areas of mental health, individuals may not always be aware of how certain practices contribute to ongoing attachment, reinforcement of distress, or difficulty disengaging over time. This does not invalidate participants’ accounts but highlights the importance of approaching both participants’ and researchers’ understandings as situated and partial. In this sense, engagement may simultaneously support recovery while also, at times, contributing to processes that are not fully recognized by those involved.

Applying CHIME-D made it visible how participants’ recovery processes were embedded in relational and social dynamics. At the same time, the analysis also highlights that the CHIME-D domains may oversimplify the ways in which supportive and harmful processes are intertwined, as participants’ accounts did not distinguish clearly between what might be considered “recovery” and what might be understood as risk or maintenance of distress. This understanding of recovery as ongoing, multifaceted, and personally defined echoes findings reported by Lewis et al., who similarly describe recovery as extending beyond the mere discontinuation of self-injury [25].

In this study, participants described how professionals’ concerns were experienced as reinforcing a sense of shame and an urge to conceal their involvement. Such approaches might reflect a biomedical framing in which online interaction among people with lived experience of self-harming behavior or suicidality is understood primarily through the lens of risk rather than as a potential resource for empowerment or connection. This risk-oriented perspective may be experienced as unintentionally undermining the autonomy, agency, and relational dimensions that personal recovery has been described to depend upon [26]. This tension highlights a broader challenge within recovery-oriented practice, where risk is often approached as something to be eliminated rather than engaged with [6]. The concept of dignity of risk, originally defined by Robert Perske in 1972 [27] captures how taking risks may be essential for autonomy and personal development [26, 28], while harm reduction approaches emphasize supporting individuals in navigating, rather than avoiding, such risks [29]. Together, these perspectives align with understandings of recovery as a process involving ongoing negotiation of vulnerability, responsibility, and personal limits [30]. Although the dignity of risk framework originates in disability studies [27, 28] and has since been applied mainly in contexts such as aged care, rehabilitation and community support [28]^, the underlying ethical argument being that risk-taking can be integral to autonomy, agency and identity [29] is relevant for understanding how individuals negotiate risk within these online self-harm communities.

Participants’ accounts in this study suggest that engagement in these online communities, despite potential harms, provided opportunities to navigate boundaries, take responsibility for oneself and others, and develop relational competencies. However, the predominantly risk-averse responses described among professionals may inadvertently limit these developmental processes. This highlights the importance of approaching such engagement with curiosity and openness, acknowledging both the potential risks and the experiential value embedded in these practices, in line with harm reduction principles.

While personal recovery is often described as an individual and subjective process, it is increasingly understood as fundamentally shaped by social contexts and relationships, arguing that recovery cannot be understood solely as an individual process but must be seen as fundamentally embedded in social relationships and everyday interactions [31]. Their work highlights how relationships with others, including peers with similar lived experiences, can provide opportunities for mutual support, reciprocity, and the reconstruction of identity beyond the role of patient [31–33]. In this study, such processes were particularly evident in participants’ descriptions of online communities as spaces where they could both receive and provide support, thereby actively contributing to their own and others’ recovery.

However, the growing institutionalization of recovery within psychiatry has shifted it from a grassroots movement towards a professional framework [31], and recovery has increasingly been shaped by formal requirements and professional frameworks [26, 31], leading to a predominant focus on clinical recovery [26]. This institutionalization has been widely criticized for narrowing the concept of recovery, as the increasing emphasis on standardization, measurability and professionally defined outcomes risks marginalizing the original user-led values of autonomy, hope and self-determination [26]. In this context, frameworks such as CHIME-D may also contribute to stabilizing particular understandings of recovery, potentially overlooking forms of recovery that do not align with linear progression, symptom reduction, or clearly positive change.

Participants’ narratives draw attention to the known tension between the rhetoric of recovery and the reality of its implementation in practice [26, 34]. While recovery-oriented principles emphasize self-determination and hope [7], the professionals in this study were described as interpreting recovery through a clinical or institutional lens that prioritized safety and control. For participants, this was associated with feelings of being misunderstood and morally judged. Such experiences are consistent with previous findings, suggesting that when recovery becomes codified within professional frameworks, it may lose its emancipatory, autonomous, and person-led qualities [26, 35–37]. From participants’ perspectives, a more recovery supportive approach might involve professionals approaching online engagement with curiosity, seeking to understand what individuals gain from these communities within their personal and social recovery processes rather than assuming such engagement to be inherently risky, respecting individuals’ autonomy sometimes involves honoring decisions that clinicians may perceive as potentially detrimental [26].

The findings further highlight how the stigma, known to be associated with self-harming and suicidal behavior [38–40] was experienced as extending into perceptions of online communities. Participants described an extensive social and professional discourse that was perceived as portraying these communities as harmful, thereby contributing to the silencing of nuanced and positive experiences participants described their engagement also contained, resonating with individuals concealing or modifying stigmatized behaviors to avoid discreditation [41]. Participants in this study engaged in various forms of self-censorship e.g. deleting posts, avoiding certain language, or hiding their participation altogether to protect themselves from criticism or loss of support. However, participants described that such concealment limited their ability to express vulnerability and to receive validation and empathy, which are commonly described as central components of recovery [8, 32]. The need to hide one’s sources of support can be understood as revealing a paradox: in trying to protect people from risk, professionals, media portrayals and societal narratives may inadvertently be experienced as reinforcing isolation and shame, highlighting that persistent impulses and fluctuations do not necessarily invalidate recovery, but can be understood as part of a broader, ongoing process [25].

This tendency toward concealment and self-censorship also points to the moral hierarchy that might exist between professionally approved forms of support and those that emerge naturally between peers [26, 32, 34]. Several participants contrasted their experiences in online communities with their encounters with professional peer workers in the mental health system, who were often described as being perceived as part of the staff rather than equal peers. Although peer workers are formally designated as supporters, their roles are bound by institutional expectations, documentation requirements, and hence some professional distance [42, 43]. By contrast, relationships in online communities centered around self-harming and suicidal behavior were experienced as more genuine, reciprocal, and emotionally resonant. Rather than understanding this primarily as a distinction between professional and peer support, our findings suggest that recovery processes unfolded in contrast to, and sometimes in opposition to, established mental health structures.

The theme of shared responsibility adds further nuance to the understanding of how recovery is lived and practiced in these online contexts. Participants described how their engagement prompted reflections on moral and emotional boundaries, i.e. how to care for others without losing oneself, and when to act or when to step back. These negotiations were described as reflecting the relational and social nature of recovery [31, 32]. For participants, supporting others was described as not merely an unselfish act but an integrated part of their own healing. Helping peers manage crises or navigating despair was experienced by participants to reinforce their sense of competence, purpose, and self-worth, which are elements commonly described as central to personal recovery [7]. The feeling of being needed and the feeling of making a difference for a peer, was described as transforming participants’ self-perception from patient to capable helper.

At the same time, these interactions could be emotionally demanding and ethically challenging. Participants sometimes struggled with the dilemma of breaking confidentiality to prevent harm or with the emotional weight of witnessing distress they could not ease. These descriptions provide insight into the complexity of responsibility within these unmoderated online communities. Importantly, several participants described developing strategies to balance involvement and self-care, such as taking breaks or unfollowing distressing content. These actions were described as demonstrating that recovery is not only about connection but also about personal limits and the development of the ability to protect one’s own mental health. Participants’ experiences suggest that recovery, within this sample, was co-constructed through peer interactions outside formal mental health systems, where individuals draw on shared experience, emphasizing the social and relational dimensions that are often described as being less visible within clinical settings [26, 31]. This resonates with Topor’s critique of the psychiatrization of recovery, where a fundamentally social process becomes relocated within professional systems and reframed in clinical terms [31].

Rather than viewing these online communities as opposed to professional treatment, they may be understood as complementary arenas that extend recovery beyond the boundaries of the mental health system. From this perspective, participants’ engagement can be understood as reclaiming recovery as an everyday, relational practice rather than a clinically regulated intervention. From this perspective, participants’ experiences also raise questions about what is recognized as legitimate recovery, and whether existing frameworks sufficiently capture recovery processes that unfold in informal, peer-driven, and at times contradictory social contexts. From participants’ accounts, this points to the relevance of openness and curiosity rather than prohibition. Recognizing such online peer engagement as part of the individuals’ personal recovery journey does not mean ignoring risks, but rather engaging with them constructively, thus acknowledging both the potential harm and the potential benefits, which has been described as a double-edged sword [11].

This study of recovery in the context of unmoderated online communities centered around self-harming and suicidal behavior suggests that personal recovery is not restricted to professional settings but experienced as emerging in the everyday contexts where people connect, reflect, and support one another online.

Implications

The implications for practice and further research are significant. If mental health professionals dismiss these online communities as purely harmful, there is a risk of alienating and stigmatizing individuals who experience them as meaningful and supportive. On the other hand, by approaching these spaces with curiosity and openness, professionals may be better able to build trust and gain insight into how individuals themselves understand recovery.

Rather than assuming participation is either beneficial or harmful, professionals may benefit from exploring with individuals what functions these communities serve, what they gain from participation, and whether aspects of engagement also create challenges or distress. Such conversations may support collaborative reflection on how to maximize supportive aspects of participation while identifying and managing potential risks.

At the same time, this raises an important tension. When we begin to frame these communities as “recovery contexts,” there is a risk that we impose professional and theoretical understandings onto practices that are, at their core, self-organized and participant driven. In that sense, we may unintentionally contribute to formalizing or regulating something that derives much of its meaning from being informal, flexible, and outside traditional systems. From a recovery perspective, this points to the importance of maintaining an awareness of whose knowledge counts, and of approaching these communities with a degree of humility, recognizing that meaning and support are created within the communities themselves.

For research, the findings point to a need for more participatory and longitudinal studies that can capture how engagement in these communities unfolds over time. In particular, there is a need to better understand what it is within participation that, for some, seems to support recovery, while for others it may contribute to ongoing distress or maintenance of self-harming practices. This includes exploring what shapes different trajectories over time, such as why some remain in these communities while others disengage, and how experiences of support and harm may shift. Rather than identifying fixed or universal effects, future research should aim to capture the complexity and variability of these processes as they are lived and negotiated over time.

Methodological considerations

Methodological rigor was sought by acknowledging the overlap in data sources while clearly differentiating the research questions, analytical approach, and theoretical contributions of the present study. Further, to ensure transparency and avoid duplication, we explicitly reference our previous publication, disclose the roles of each author, and clarify the distinction in both aims and findings, aligning with recent recommendations [44].

As the analysis is based on data collected for another purpose, some areas might have been explored more comprehensively through targeted interview questions. However, the dataset is rich and rigorous, offering nuanced descriptions of participants’ experiences that are essential for addressing the objectives of the current study. The application of the CHIME-D framework [3] strengthened the analysis by providing a structured lens through which the complexity and interrelated nature of recovery processes could be identified and interpreted as both personal and socially embedded.

The participants predominantly consist of female participants under the age of 25, with 12 of 13 participants identifying as female. While this reflects patterns observed in existing research on online self-harming and suicidal behavior [14, 45], it may limit the transferability of the findings, particularly regarding how the experience of recovery identified in this study applies across different gender and age contexts.

Individuals identifying with genders other than female may have distinct experiences, needs, or barriers regarding online peer support, which future research should aim to explore. Further, only participants aged 18 years and older were included. Legal requirements for parental consent may have constrained the inclusion of younger participants in anonymous online communities, which may limit the transferability of the findings, as also noted previously [21]. Finally, as with all research on digital communities, findings are shaped by the specific platforms and of the online environments in which participants engaged. Experiences may differ in other digital contexts or across cultural settings. Despite these limitations, the current study provides valuable insights into how online peer-support networks contribute to personal and social recovery, highlighting the relational, ethical, and reflective dimensions of engagement.

Conclusion

This study demonstrates that engagement in unmoderated online communities centered around self-harm and suicidal behavior serves a meaningful yet ambivalent function in personal recovery. Using the CHIME-D framework, recovery emerged as a relational and socially embedded process shaped by both supportive peer connections and significant challenges. Participation appeared to fulfil important needs for connectedness, recognition, and being seen as an equal person, while simultaneously involving stigma, secrecy, moral responsibility, and exposure to distressing content. This underscores recovery as a non-linear and negotiated process in which difficulties form an integral part. The findings further highlight a tension between participants lived experiences of recovery and professional interpretations shaped by risk management. While professional responses were often experienced as restrictive, online peer communities were described as offering reciprocal and emotionally resonant support. Recognizing digitally mediated peer spaces as legitimate - though complex - recovery contexts may contribute to a more nuanced understanding of individual trajectories towards personal recovery.

Author contributions

MEH and JM initiated and developed the study idea. MEH was responsible for the data generation, transcriptions, and the initial analysis. JM and KJ contributed to the analysis process. MEH wrote the first draft of the article, while JM, KJ, KA, MV and MN critically revised the draft and provided valuable feedback and amendments. All authors have approved the submitted manuscript and consent to publication.

Funding

Open access funding provided by Copenhagen University. This research is part of a Ph.D. project, funded by the Novo Nordisk Foundation (Grant number NNF20OC0066179). The funder has no role in the development, implementation, or completion of the study.

Data availability

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

Declarations

Ethics approval and consent to participate

The study was approved by the Capital Regions protection agency PRIVACY (Approval number P-2022-41). Further approval was not required, following Danish legislation. Prior to conducting the interviews, all participants were provided both oral and written information detailing the study’s purpose and methods, as well as their rights. Informed consent was obtained in writing, either physically or electronically.

Consent for publication

All authors approved the manuscript and consented to its publication.

Competing interests

The authors declare no competing interests.