‘I’ve seen you at your worst, but I still want to be your friend’: A qualitative study of peer support exchange from the perspective of participants in online communities centred around self-harming and suicidal behaviour

Abstract

Objective

Research on online communities centred around self-harming and suicidal behaviour often focuses on harmful effects. The concept of peer support, which may help explain why individuals engage with such forums, remains under-explored. This study aimed to examine how peer support is experienced in these communities.

Methods

Thirteen in-depth, semi-structured interviews were conducted with current and former participants in online communities focused on self-harming and suicidal behaviour (12 women and one non-binary person, mean age 28). A reflective thematic analysis was used to identify patterns in the data.

Results

Four main themes emerged: (1) You are allowed to feel bad and express it on your own terms, (2) Anonymity and intimacy is not a paradox in the online world, (3) It is necessary to have friends in real life as well, and (4) When support, relations, and motivation change over time, one must ask: Should I stay or should I go? The findings not only illustrate the communities as emotionally supportive spaces that foster belonging and intimacy, but also point to challenges in maintaining boundaries and disengaging when support becomes unhelpful.

Conclusion

Online communities centred around self-harming and suicidal behaviour provide an alternative to traditional mental health services, often perceived as paternalistic and unsupportive of autonomy. Participants describe online peer support as a way to share emotional burdens and form meaningful relationships. However, the informal nature of support may complicate recovery, especially when individuals become entrenched in the community. These dynamics warrant further research and consideration in mental health practice.

Introduction

Peer support involves individuals using their lived experiences to support others facing similar challenges.^1–3 Within mental health care, peer support has a well-documented positive effect on both clinical and personal recovery. ² Shery Mead et al. ¹ defines peer support as:

A system of giving and receiving help founded on key principles of respect, shared responsibility, and mutual agreement of what is helpful. Peer support is not based on psychiatric models and diagnostic criteria. It is about understanding another's situation empathically through the shared experience of emotional and psychological pain. This connection, or affiliation, is a deep, holistic understanding based on mutual experience where people are able to ‘be’ with each other without the constraints of traditional (expert/patient) relationships.

This definition highlights that peer support emphasises relational connection beyond clinical diagnosis and professional roles, centring on empathy and shared experience. Traditionally peer support was delivered through face-to-face interactions or clinical settings. ⁴ Today peer support is increasingly facilitated by social media platforms, which provide unique affordances such as anonymity, immediacy, and multimodal communication, that is, through text, videos, or pictures.^5–7 Unlike offline peer support, online communities allow for rapid, anonymous, and multimedia exchanges that can reduce barriers to disclosure and create novel support opportunities.^3,7,8 Further, the ability to algorithmically promote content, sustain large-scale visibility, and enable persistent engagement through features such as likes, shares, and follows marks a significant shift from earlier, more static online forums.^9,10

Online communities centred on self-harm and suicidal content, especially on image- and video-based social media platforms like Instagram, Tumblr, and TikTok, have grown substantially.^11,12 These spaces allow individuals to share experiences, express distress, and seek understanding from peers facing similar challenges. The anonymity and immediacy afforded by these platforms facilitate open and timely peer interactions that might not be possible offline, creating a sense of safety and belonging.^8,13,14 These communities are often attractive to individuals who may face stigma, social isolation, or lack of offline support, providing a space where they can connect with others who truly understand their experiences.^4,7,15

However, these communities present a double-edged sword. Minimal content moderation on social media and easy access to graphic or triggering materials can contribute to harmful outcomes, such as normalising self-harming behaviours or exacerbating suicidal ideation.^16,17 Participation has been linked both to increased self-harming urges and suicidal ideation as well as reduced feelings of loneliness and improved access to peer support.^18,19 A recent review found that posting or viewing self-harm images can lead to both positive outcomes, like empathy and help-seeking, and negative outcomes, such as learning new methods to self-harm, which may normalise harmful behaviours. ⁶ This suggests that the effects of these online communities vary widely depending on individual experiences, supported by current evidence.^6,8,11,19,20

What distinguishes self-harm and suicidality-focused online communities on social media from other mental health forums is this complex duality, where peer support can both foster recovery and exacerbate risk. This dynamic interplay highlights the importance of understanding not only why individuals engage in these spaces but also how peer support relationships emerge and develop within them. While prior research has investigated motivations for participation and broad outcomes of engagement, there remains a gap in knowledge about the formation and functioning of peer support exchanges in these online community contexts.

Understanding these relational processes is essential to harness the benefits of peer support in reducing stigma and promoting recovery while mitigating potential harms associated with exposure to self-harm and suicidal content. This study aims to explore the subjective experiences of peer support among participants in online communities centred on self-harm and suicidal behaviours.

Specifically, we ask how peer relations emerge and form in social media-hosted online communities centred around self-harm and suicidal

Method

Recruitment

We employed criterion-based purposive sampling to select participants who could provide rich, detailed accounts relevant to our research questions. Participants were included if they (1) had self-reported experience engaging in online communities centred around self-harming and/or suicidal behaviour, regardless of level of involvement, (2) were fluent in English or Nordic languages (Danish, Swedish, or Norwegian), (3) provided informed written consent, and (4) were 15 years of age or older.

Participants were recruited between April 2024 and July 2024 through psychiatric in/outpatient facilities in four out of five Danish regions and through residential institutions across the country. Further, three Danish non-governmental organisations (NGO) shared the recruitment advert through their social media accounts.

Throughout the recruitment period, MEH, JM, and KJ evaluated the number of participants in relation to information power ²¹ with the understanding that saturation does not conform to the values and assumptions of reflexive thematic analysis but is inevitably situated and subjective and can therefore only be determined on an ongoing basis. ²² Based on the inclusion of theory, a focused objective, and expecting comprehensive data through rich dialogue, we estimated that 10–15 participants would be sufficient, depending on the variation achieved within our sample.

Participants

A total of 13 participants aged 21–49 were included in the study, representing all five regions of Denmark. None of the participants was known by the authors prior to study commencement. No participants were excluded. Of these, 12 identified as women and one as non-binary. All participants had previously experienced psychiatric hospitalisations, including emergency room or intensive care stays. Yet, only one participant was recruited from mental health settings; the remaining 12 responded to an advertisement shared on social media by three Danish NGOs. Their diagnoses included attention-deficit disorders, personality disorders, autism spectrum disorders, and schizophrenia. All but one participant reported prior suicide attempts. All had engaged in physical self-harm, although six reported no longer engaging in self-harming behaviours. Four participants resided in group homes, while nine lived alone or with a spouse. Five had completed or were studying for a bachelor's or master's degree and worked full or part-time. Two participants were on sick leave or unemployed. One participant worked flexibly, three were on early retirement, and two were unemployed, relying on social services. Ten participants were currently active in online communities related to self-harm and/or suicidal behaviour, while three were not active anymore. All participants chose to remain anonymous and created profiles using pseudonyms, when participating in social media-hosted online communities centred on self-harming and suicidal content.

Ethics

Prior to the interviews, participants received oral and written information about the study's purpose and methods, and their rights. They then provided written consent either physically or electronically. It is important to pay particular attention to vulnerable groups in research, for example, people suffering from mental health problems ²³ ; participant materials and consent forms were therefore prepared in accordance with Danish legislation and recommendations from two patient representatives, currently employed as mental health peer workers in the Capital Region of Denmark, but not affiliated with the study nor part of the research team. Post-interview, participants were asked about their current state of mind and were offered support if needed, for example, in the form of supportive conversation or contact details for the psychiatric emergency room, and so on (see Appendix A). No participant reported discomfort or thoughts of self-harm. Data were safely stored in accordance with the European Union General Data Protection Regulation. ²⁴ Anonymity and confidentiality were established by coding data and identifying respondents using ID numbers.

Interviews

The guide for our semi-structured interviews was inspired by existing interview guides,^25–28 evidence identified in our recent work, ²⁹ and Shery Mead et al.'s conceptualisation of peer support. ¹ We also used feedback from the patient representatives to refine the guide and ensure that our questions were relevant to people with lived experience. ³⁰ The interview guide is included in Appendix A.

The interviews were conducted by MEH, a female mental health nurse with a master's degree, 10+ years of clinical experience, and prior qualitative research experience. MEH was unknown to the participants, who were given background information about them, including age, sex, education, and study motivations. All contact was through MEH.

The participants chose their preferred format for interview, that is, face-to-face (n = 2), online (n = 6), telephone (n = 2), and mail/in writing (n = 3). On average, interviews lasted 70 min (min 44 min/max 108 min) regardless of format. The resulting transcriptions were 249 pages (ranging from 14 to 25 pages each). This comprehensive textual corpus formed the basis of our analysis.

Analytical process

MEH transcribed the audio recordings of the telephone, online, and face-to-face interviews verbatim. The interviews in writing were uploaded in un-edited form. Transcriptions were uploaded to NVivo for analysis. We followed Braun and Clarke's six-phase approach to reflexive thematic analysis, which allows inductive exploration of data, focusing on patterned themes.^31,32 In phase one, for further familiarisation with the data, MEH re-listened to and re-read the interviews and transcriptions. In phase two, initial codes were generated by identifying patterns and meanings within the data. These codes were grouped into themes in phase three. Phase four involved reviewing and further developing the themes by consolidating content across the initial themes before finalising and naming them in phase five. The analysis process was iterative, ³² revisiting each phase several times to improve and refine results (see Table 1 for examples).

Table 1.

Overview of the iterative analysis process.

Coded text	Code name	Subtheme	Theme
SOMETIMES IT'S JUST REALLY NICE TO BE IN A PLACE WHERE YOU DON'T HAVE TO WORK TO STOP AND CAN SAY WHAT YOU WANT. PARTICIPANT 13	Freedom of speech, when feeling bad	Express it in your own way	You are allowed to feel bad and express it in your own way
IT HAS BEEN LONG MESSAGES ABOUT WHAT HAPPENED TODAY… AND IT IS JUST SO NICE TO HAVE SOMEONE YOU CAN TALK TO ABOUT EVERYTHING. THERE IS NO JUDGEMENTS, THERE’S NOTHING. AND SHE DOES NOT GET, YOU KNOW, SUPER WORRIED ID I WRITE THAT I HAVE SELF-HARMED. PARTICIPANT 10	Being acknowledged and supported when feeling bad	Express it in your own way
YOU DON'T NEED ALL THAT FORMAL STUFF FIRST. THERE'S NO CHEMISTRY CHECK, THERE'S NO WHAT DO YOU REALLY THINK OF ME? BECAUSE YOU ALREADY KNOW EACH OTHER'S INNERMOST THOUGHTS. PARTICIPANT 5	Non-superficial relation	The depth of friendship	Anonymity and intimacy is not a paradox in the online world
IT'S LIKE YOU START YOUR FRIENDSHIP ON A DIFFERENT LEVEL. YOU UNDERSTAND EACH OTHER IN A DIFFERENT WAY AND HAVE A DIFFERENT UNDERSTANDING OF EACH OTHER AND EACH OTHER'S CHALLENGES. PARTICIPANT 4	Deep understanding of one another	Friendship on another level
I ALSO NEED MY FRIENDSHIPS WITH THOSE WHO AREN'T MENTALLY ILL, BUT I'LL JUST TELL YOU THAT THERE AREN'T MANY LEFT. IT'S BECAUSE THERE ARE SO MANY PEOPLE WHO HAVE ENDED THEIR FRIENDSHIP WITH ME, BECAUSE HOW COULD THEY EVER REFLECT THEMSELVES IN ME AND IN WHAT I'VE GONE THROUGH? AND IT'S PROBABLY BEEN INSANELY HARD FOR THEM TO BE WORRIED ABOUT ME. PARTICIPANT 10	Offline friends leave	When the offline friends leave	It is necessary to have friends offline as well
I JUST REALLY NEEDED NORMAL FRIENDS TOO. PARTICIPANT 11	The need for normal relations	The need for normal relations
IT NOT THE ONLY KIND OF PEOPLE YOU WANT TO HANG OUT WITH. PARTICIPANT 3	Can’t just be with social media relations	The need for normal relations
NOW I MOSTLY USE IT AS A SCARE CAMPAIGN. FOR WHAT HAPPENS IF I DON’T STAY AWAKE. LIKE A MIRROR OF WHAT I DON’T WANT TO GO BACK TO. IT MAKES ME FEEL LESS ALONE. IF I GET TRIGGERED, I HAVE THE RESPONSIBILITY MYSELF NOT TO FOLLOW IT OR ASK MYSELF ‘WHY DOES THIS TRIGGER SOMETHING IN ME?’ THEN IT BECOMES A WORK POINT FOR ME. PARTICIPANT 6	Uses it as motivation	Reasons for staying	When support, relations and motivation change over time, one must ask, ‘Should I stay or should I go’
I DON'T THINK I COULD TAKE CARE OF MYSELF IN THAT ENVIRONMENT. IT ALWAYS ENDED UP IN MORE SELF-HARM AND MASSIVE SUICIDAL THOUGHTS AND SOME SUICIDE ATTEMPTS. I COULDN'T BE IN THAT ENVIRONMENT. I SEE THAT TODAY. PARTICIPANT 9	Leave to take care of oneself	Reasons for leaving

Phase six involved writing a structured report following the Consolidated Criteria for Reporting Qualitative Research (COREQ) checklist. ³³ The COREQ checklist is provided in Appendix B.

MEH, JM, and KJ contributed to phases one to four, and MEH, JM, KJ, KA, MV and MN contributed to phases five and six.

Results

The analysis resulted in four themes.

Theme 1: you are allowed to feel bad and express it on your own terms

Participants described the option to participate anonymously in online communities centred around self-harming and suicidal behaviour as crucial in alleviating concerns about the impact of their online activity on their current or future lives. In particular, participants experienced significant stigma related to their mental health diagnoses, self-harming behaviours, and suicidal urges and thus felt compelled to participate anonymously to avoid jeopardising their chances of a normal life. They described experiencing anonymous participation as a pre-condition of authenticity.

It's really because I think I’m going to work as a teacher at some point again, and therefore I would be really worried if students or parents could connect me to some of the things I write in there.

Participant 10.

Participants’ attachment to online communities seemed to be reinforced by the reactions of family and friends when exposed to their self-harming behaviour and suicidal urges. Participants shared how their loved ones struggle to understand their reasons for self-harm, often scolding them and demonstrating a lack of understanding. Encountering such reactions made being a part of online communities even more appealing, as they could share thoughts and feelings without experiencing guilt or judgement and without being asked to take responsibility for using self-harm as a coping strategy.

You don’t have to use your skills. You’re allowed to feel bad. And it's sometimes really nice. If you say you feel like shit in real life, it becomes a cross-examination: In what way? How bad is it? Is it dangerous? If I say I feel like shit, it's not just ‘Do you need a hug?’ It's ‘oh no! We’re going on vacation! The travel insurance won’t work. Do you remember what this did to your sister last time? It doesn’t benefit me. It's nice that there are people who understand in there (online communities ed.)

Participant 13.

Participants reported their experiences with the mental health system, describing it as one that is often paternalistic, under-resourced, patronising, and unsupportive of autonomy.

There are a lot of staff who criticise the network and don’t really listen to what we think. They tell us not to share our Facebook or Instagram. We’re not supposed to connect outside the psychiatric hospital, because of our self-harming behaviour. I think it's because they don’t know much about it.

Participant 8.

They reported experiencing mistrust and stigmatisation due to their self-harming behaviour, with these feelings being further reinforced when staff forbid the formation of relationships with peers through psychiatric hospitalisation or outpatient treatment. They describe the failure of staff to understand or be curious about their online engagements, resulting in participants intensifying their secrecy around and attachment to online communities.

Theme 2: anonymity and intimacy is not a paradox in the online world

Participants highlighted the unique bonds formed with peers through shared experiences, despite the fact that participants engage from anonymous profiles and do not identify themselves with their actual identity. A recurring narrative was the profound understanding and lack of judgement within these relations, emphasising that they are built through shared struggles. This creates a strong foundation of empathy and mutual support, leading most participants to call these relations their friends, rather than just acquaintances or peers.

It's like the worst thing in your life that has united you. We’ve seen each other in the most broken situations, and still choose each other. I mean really, really providing strong support, care and very, very beautiful people-to-people interactions. I've seen you at your worst and I still want to be your friend.

Participant 10

Participants stressed that these relations start at a deeper level than conventional ones, bypassing the superficial layers often present in novel relationships and diving straight into deeper, more meaningful interactions. Anonymity and intimacy are not perceived to be paradoxes; rather, anonymity appears to create a safe space where vulnerability can emerge more readily. By removing the pressures of identity performance and social judgement, participants described feeling freer to share personal experiences, fostering a sense of immediate emotional connection.

As soon as you've met someone from the network and you click personality-wise, it's never superficial. You've already seen their worst side of themselves. It's almost the other way around. Usually, it's often superficial ‘I'm fine, things are going well’ and stuff like that. So it's a deeper relationship from the start compared to real-life meetings.

Participant 2

Having witnessed one another's most vulnerable moments constitutes an acceptance that is inherent to these relationships. Participants contrasted the freedom to be oneself without fear of judgement with the pressure to conform to societal norms when interacting with people in social settings offline.

Because of my ADHD, I'm very fidgety. So, meeting people in real life, I make these rules. ‘In an hour you can move this many times on a chair’, because you don't want to seem completely fucked up or uninterested. But if I meet with my friends from the network, I can stand on my head if I want. I don't care. There's not that barrier.

Participant 5

Participants described the deep emotional connection despite the physical distance or lack of face-to-face interaction as a peculiar but poignant aspect of these relationships. A term that surfaced repeatedly was ‘mirror’, with participants describing the experience of exceptionally solid connections; some characterised meeting these individuals – either online or in real life – as ‘coming home’, representing the epitome of safety and cohesion. The participants recounted finding a mirror so recognisable and powerful that they experienced a true meeting with their own reflection, thus becoming united with and more forgiving and understanding of themselves.

Theme 3: it is necessary to have friends offline as well

Participants described needing offline friendships with people with and without mental illness, stressing that surrounding themselves exclusively with others who are struggling can lead to a cycle of illness and worries.

I often surrounded myself with people who were also struggling. My friend said, ‘you need healthy friendships with someone who isn't mentally ill’. And he's right. Because it can quickly turn into illness and worry and hospitalizations… It's good to have friends where you talk about something else and do something else.

Participant 10

Participants expressed the need to have friends with different life experiences to provide a balanced and stable social environment. They explained that having friends who are not struggling with mental health issues offers a different perspective and a sense of normalcy, which is crucial for their overall well-being. Real-life friends often contrast with online friends – where online relationships are rooted in shared experiences and the ability to mirror one another, offline friends reflect ‘normality’ and having relationships that represent everyday life.

It's very liberating that you didn't have to talk about it all the time. One day my friends introduced me to a program called ‘Naked Attraction’ (A TV-show, eds.). I almost died! I didn't understand the point. But it felt normal and nice.

Participant 4

Despite acknowledging the need for offline friends, participants reported struggling to maintain these friendships. Many of their friends had ended the relationships, perceived as caused by the challenges of relating to their experiences and the emotional burden of worrying about their well-being.

After I got my diagnoses and after I started self-harming, I lost a close group of friends because they are in a completely different place now. They have cut me out.

Participant 8

The participants’ experiences suggest that a feeling of social isolation often accompanies a life with mental illness. They saw the challenges in finding and maintaining supportive friendships as rooted in a sense of living in parallel worlds. While friends who understand and share similar struggles provide crucial support and empathy, non-mentally ill friends offer stability, different perspectives, and a sense of normalcy; participants viewed both forms of friendship as beneficial.

Theme 4: when support, relations, and motivation change over time, one must ask, ‘should I stay, or should I go?’

Participants describe peer support within these online communities as initially being reciprocal, but after years of engagement, they experienced a shift to a more one-dimensional form of support, with older users taking on mentorship roles for newer or younger users.

Maybe it's also because those of us who are a little older know what it was like to be young and have no one to look up to. So, you need someone who can say ‘Hey, you can have a good life’.

Participant 6

Participants expressed their desire to be role models and their awareness of representing a mirror for younger users and offering inspiration and hope. They also explained that there is a clear difference among users of the networks based on age and experience, emphasising that those younger than 18 often post very explicit and concerning self-harm and suicidal content. Conversely, participants described a tendency to follow profiles that offer relatable content and meaningful engagement. They seek connections with those who have similar experiences or challenges, and some take an active role in promoting what they call a recovery-oriented culture within the community.

Many sees the network as… I can only be here if I'm feeling bad. No, hopefully not. Because then you're stuck. Because you'll miss this community that you were such a big part of, while you were feeling really bad. If you think that you can't be a part of it if you grow and get better, then you'll be stuck. You don't want to get better. And that's not how it is supposed to be.

Participant 5

They argued that feeling unable to participate when doing well can trap some individuals in a place where they cannot get better. By actively sharing both highs and lows, they hope to make the network a place where one can seek and provide peer support without romanticising life with mental illness. However, some participants noted that sharing negative experiences garners more engagement than positive updates, suggesting that the community is more interested in shared struggles than successes. This imbalance can make those who are doing well feel overlooked, with some even feeling prompted to share self-harm or suicidal content, despite being in a place of recovery.

This is going to sound weird. But the better you are doing, the less reactions you'll get. That's just my experience. I don't think they notice me in the same way anymore.

Participant 12

Participants outlined that everyone is responsible for who they follow and how they use the communities while acknowledging the potential difficulty younger and more inexperienced users face in navigating them. They emphasised that younger users often share more explicit content and seem less reflective about what is shared.

All the participants had considered leaving the communities, acknowledging that this decision is deeply personal and influenced by various factors requiring much contemplation. Some use the community more during some periods than others, depending on their need, often engaging less during periods in which they are feeling better.

The community is a major validation machine. None of us know how to self-validate. We all hate ourselves. We validate each other, trying to give some love that we don't know how to give ourselves. So, if you're in a period where you can actually do that, then you kind of don't need it.

Participant 13

Some participants chose to stay in the community because they continue to derive meaning from the exchange of peer support and sharing their experiences. They feel a sense of duty and fulfilment in being a role model and offering hope based on their experiences. Conversely, some participants described leaving the community because they felt it hindered their recovery, exacerbating their self-harming behaviour and suicidal thoughts, leading to the realisation they needed to distance themselves for their well-being.

I don't think I could take care of myself in that environment. It always ended up in more self-harm and massive suicidal thoughts and some suicide attempts. I couldn't be in that environment. I see that today.

Participant 9

Others described having opted out because they finally found real-world community and support. Some participants had not yet decided about leaving, describing the decision as fraught with uncertainty and fear. They expressed that their needs have changed, yet they fear the repercussions of severing ties. Some grapple with a sense of responsibility towards others still active in the community and conveyed fear for their well-being.

How the hell do you delete your profile? Because what about the people I follow… What if they commit suicide? They're just not strangers anymore. These people mean the world to me.

Participant 4

Participants expressed being torn between the desire to disconnect and the emotional bonds formed with other users. The relationships within the community have become significant as a result of the intimacy and confidentiality embedded in their communication. Participants were distressed by the thought of leaving due to the potential loss of meaningful connections and concerns about the well-being of others, especially since most participants narrated having lost friends from the online communities to suicide.

Discussion

This study sheds further light on the seemingly paradoxical nature of the unique intimacy present in online friendships despite their anonymity. These online platforms provide a context in which individuals can openly express and discuss self-harming and suicidal thoughts and behaviours, ⁶ thereby challenging conventional assumptions regarding secrecy and authenticity. Anonymity plays a critical role in enabling these disclosures, underscoring the complex processes involved in online self-presentation and identity negotiation.^6,34 Additionally, it addresses how exchanging peer support within these communities can evolve into friendships, which are known to be vital for mental health and overall well-being. ³⁵

Our analysis of the experience of peer support exchange in online communities centred around self-harm and suicidal content reveals that the anonymity these platforms offer supports participants’ sense of autonomy, helping to manage the stigma experienced within the mental health system. Negative reactions from family and the mental health system prompt participants to seek non-judgemental support online. Despite the anonymous nature of participation, shared struggles foster deep bonds and empathy, forming strong, supportive, intimate friendships. However, balancing these online connections with offline friendships is essential to maintain normalcy and avoid a cycle of illness. The motivation for engaging in peer support exchange online evolves over time, with more experienced users taking on the position of mentor and role model for younger users. Deciding whether to leave the community or stay is influenced by the need for validation, fear of isolation, and desire for recovery. These findings align with existing evidence suggesting that engaging in mental health peer-supportive relations in online communities centred around self-harm and suicidal behaviour is a double-edged sword.^16,17,19,36

Our findings contribute directly to an ongoing, increasingly polarised debate. ³⁷ On the one hand, proponents are in favour of strict social media regulation, with some even calling for an outright ban for social media use for children and young people. On the other hand, opponents argue that such measures would amount to censorship, violating young people's right to freedom of expression and potentially cause more loneliness among already vulnerable people who are finding a community online. ³⁸ Our findings underscore the complexity of young people's engagement with online communities related to self-harm and suicidal behaviour. Participants described these spaces as both helpful and, at times, problematic, hence illustrating that online peer support can offer meaningful connection, emotional relief, and a sense of belonging, while also posing potential risks. These insights point to the need for a more nuanced approach to regulation, that is, one that recognises the diversity of user experiences and balances protection with the preservation of supportive peer networks for those who may not otherwise have access to care or understanding within their physical network or mental health services.

It is well established that exchanging peer support online is valuable.^6,18,19,39 We argue that peer support in these online communities can be seen as returning to fundamental peer support principles, notwithstanding its professionalisation in contemporary mental health settings.^40,41 This form of peer support, characterised as anarchistic and autonomous, is highly valued by participants. They perceive the mental health system as paternalistic, under-resourced, and unsupportive of autonomy, contradicting the historical origins of recovery and peer support, which emphasised mutual aid and reciprocity.^42,43 Thus, the perceived psychiatrisation of recovery ⁴³ strengthens affiliation to these online communities. Our findings indicate that participants seek and maintain their relationships with these networks in opposition to the stigmatisation and lack of understanding they experience in the mental health system.

The nature of the peer support explored in this study reflects the principles highlighted by Mead et al., which emphasise reciprocal respect and empathic understanding through shared experiences. ¹ This allows for authentic relationships without the constraints of an expert/patient dynamic, ¹ simply fostering friendships and providing a context in which needs are met. ³⁵ Therefore, it is essential to understand how and why peer support exchange in online communities centred around self-harming and suicidal behaviour can potentially aid recovery, and further research is warranted.

Collectively, the participants’ narratives illuminate the unique, close, and supportive nature of peer support within these communities, which develops into friendships for most. Despite differing from conventional offline friendships, they share the same characteristics of intimacy, confidentiality, and a sense of significance. ³⁵ Online friendships appear to be founded on honesty, shared experiences of suffering, and an absence of pretence, allowing for a deeper connection and enabling participants to take ownership of their narrative. Our participants underscored the critical need for understanding, non-judgemental support, and emotional safety for individuals facing mental health challenges, calling for a more person-centred approach within the mental health system.^36,44 This approach should incorporate the patient's perspective, acknowledging their autonomy and expert knowledge about their own lives, aligning with a recovery-oriented approach.^43,45,46 It is also crucial for mental health professionals to become familiar with the nature of engagement in these online communities so that they can access the behaviour and intervene if necessary, ³⁶ underscoring the need to advance knowledge about participation in online communities centred around self-harming and suicidal content and training for staff in mental health settings.

The #chatsafe guidelines developed by Orygen ⁴⁷ represent an important framework for promoting safe and supportive conversations about mental health and suicide prevention in online spaces. These guidelines encourage peer-to-peer communication that is respectful, non-judgemental, and informed by best practices, ⁴⁷ which aligns with many participants’ desires for supportive peer relationships within social media-hosted online communities. However, our findings highlight a crucial caveat: most participants expressed significant mistrust towards the mental health system and professional interventions, feeling stigmatised rather than supported by such approaches. Therefore, while initiatives like #chatsafe guidelines can play a valuable role in fostering safer peer dialogue, it is essential that their implementation be sensitive to community dynamics and participant concerns. Overly clinical or professionally framed interventions risk alienating users who prefer peer-led support and who may reject traditional mental health services. Future work should explore ways to integrate evidence-based guidelines like #chatsafe guidelines in a manner that respects participant autonomy and addresses the stigma and mistrust reported by participants.

However, in some cases, engagement can also increase self-harming urges and suicidal ideation,^16,17,19 thus hindering recovery. Our findings indicate that older members of online communities often promote recovery and act as role models, whereas younger users tend to focus more on explicit, harmful content related to their struggles. This exposure may increase vulnerability to triggers, exacerbate self-harming behaviours, and warrants careful attention. Future research should explore this, acknowledging that the informal nature of online peer support also presents unique challenges that require careful consideration, ¹⁸ which may be influenced by users’ age and functional level.

Media coverage often highlights the negative aspects of participation in online communities focused on self-harm and suicidal behaviours, overlooking the potential benefits of peer support and friendship-building. Addressing both positive and negative elements is crucial for a balanced narrative to inform policymakers and clinical practice guidelines in this under-exposed yet vital area for mental health. It is hoped that additional research will add further nuance to the current debate about social media, focusing on both risks and opportunities.

Methodological considerations

Most studies on online self-harming and suicidal behaviour represent findings from females below the age of 25, ²⁹ and 12 out of the 13 participants in the present study also identified as female. Based on current evidence and supported by our findings, participation in these communities seems to be most prevalent among females. However, future studies should try to ensure a gender-diverse study population, if possible, as other gender identities may have differing experiences and needs related to peer support.

Although all participants reported current or previous contact with mental health services, several described negative experiences and expressed distrust or avoidance of formal care. Such complex relationships with mental health services are well-documented, with stigma, dissatisfaction, and perceived paternalism contributing to disengagement or reluctance to seek professional support.^48,49 Consequently, some individuals turn to peer support and online communities as alternative sources of understanding and help, especially when traditional services are perceived as inadequate or stigmatising. ³ While our sample reflects this diversity in engagement with clinical care, it may still under-represent individuals who are entirely outside the mental health care system. Future research should aim to include such perspectives to further broaden understanding of peer support in online communities.

Despite a broad recruitment strategy, only one participant was recruited from mental health settings. Professionals reported being reluctant to ask potential candidates to participate, deeming the potential candidate too sick to participate or too close to recovery, and not wanting to trigger a relapse by asking. This is a well-known challenge when recruiting within vulnerable groups, ²³ a term that people suffering from mental health conditions are often associated with. It should, however, not automatically be an exclusive term, ⁵⁰ remembering that the recovery approach, a guiding principle in mental health practice ⁵¹ should also apply in research, emphasising the capacity of individuals to self-determine participation when given sufficient information. ²³ The remaining 12 participants were recruited via social media in collaboration with various NGOs; this represents a promising way to recruit particularly hard-to-reach populations, with the ability to reach a wider segment of the population. ⁵²

The data collected in this study are rigorous, embodying a nuanced description of experiences central to our objectives. This detailed exploration not only reinforces the validity of our findings but also deepens our understanding of the complex dynamics at play, providing valuable insights into the phenomenon. While we were unable to include patients under 18, the sample remains diverse.

The use of peer support theory strengthened information power, thus resulting in a need for fewer participants, ²¹ and we believe that the 13 participants provided a comprehensive set of data. Further, MEH has prior experience of conducting interviews and both theoretical and clinical experience with the target group and the phenomenon, which heightened the quality of dialogue and interviews. ²¹ The feedback from patient representatives was also fruitful in ensuring relevance and quality.

However, anonymity is crucial for gaining insight into a closed world in which shame, secrecy, and stigma have an impact. According to Danish law, participants under 18 years can only participate in research if their parents or legal guardians give permission. Since being part of these online communities is largely rooted in anonymity and confidentiality requirements, we strongly suspect that the legal requirement for parental permission is a significant reason for our inability to recruit participants under the age of 18, as many adolescents do not necessarily share their engagement with their parents. Our sample likely reflects a subset of participants who are relatively experienced and resourceful in navigating in online communities centred around self-harming and suicidal behaviour, which may limit the generalisability of our findings. Less experienced or more vulnerable users might face challenges not fully captured in this study. However, despite missing perspectives from participants under 18 years, we believe we have provided an analysis that can aid in understanding the exchange of peer support in other contexts in which people connected by similar life experiences can meet and relate.

Conclusion

This study examined the subjective experiences of peer support of participants in online communities centred around self-harm and suicidal content. Anonymity on these platforms supports autonomy and reduces mental health stigma, motivating individuals who seek non-judgemental support to find it online. Participating in these online communities can foster deep, empathetic bonds that develop into friendships; however, balancing online and offline relationships is essential to prevent isolation and maintain a sense of normalcy. The roles of network users evolve, with more experienced participants often mentoring younger ones and serving as role models. The decision to stay in or leave the communities is influenced by validation needs, fear of isolation, and recovery goals. Despite offering crucial support, these communities can also heighten urges to self-harm and suicidal thoughts, necessitating further research into their impact on mental health. Ultimately, online communities can generate deep, meaningful friendships that support recovery, but the associated risks should not be overlooked.

Appendix A: Interview guide

Introduction

Who am I?

- My name is Malene, I'm 36 years old and I'm a nurse

What is this about?

- I research online self-harming and suicidal behaviour and would love to learn more about how friendships and relationships are formed.

- That's why I'm really happy that you wanted to write/talk to me about your experiences with forming friendships online.

How?

- I've prepared some questions but I'm not looking for anything specific; anything can turn out to be relevant. Therefore, please just tell me what you feel like and what you think is important.

- If it's okay with you, I will record the interview. The recording will not be shared with anyone outside the project and will be deleted when the study is completed.

- We may use some statements from you in the final study, but you will not be able to tell that the statements come from you. You will be completely anonymous and unrecognisable.

- Is there anything you would like to ask me before we get started? You can also ask me along the way.

Themes	Introductory questions	Additional prompts
Demographics	First of all, I'm going to ask you some very factual questions. If there's anything you don't want to answer, just let me know. Is that okay?	What gender do you identify with? How old are you? What region do you live in? Are you currently studying or working? Do you have a partner/are you married/single? Do you live at home/alone/with someone/residential or other?
Mental and physical health		Do you have a physical illness? (e.g. asthma or diabetes?) Do you have a psychiatric diagnosis? (which one(s))
Self-harming behaviour	Since you agreed to talk to me after seeing my post, I assume that you have either currently or previously injured yourself. Can you tell me a bit about when and how you have physically harmed yourself? I understand that you have also used social media as part of self-harming. Can you explain a little about how you have used social media as part of self-harm? (Ask how and when)	Have you ever physically injured yourself? (Ask when and how) How did you learn about this/how did you know it existed? (Ask if it was through google, through friends, through social media or something else?) What do you typically do on social media when it comes to self-harm-suicidal behaviour? Ask about; 1) Have you tried posting things that could intentionally harm yourself (i.e. cause bad publicity), e.g. pictures, etc. (Ask how and when) 2) Have you ever made posts to deliberately get someone to say something bad about you or to you? (Ask how and when) 3) Have you ever written to yourself from a fake profile? (Ask how and when)
Social media	How old were you when you got your first social media profile?	Where do you have (or have had) profiles on social media? Can you tell me how you use the platforms and what you do? (Ask about frequency, usage patterns, interactions)
Motivations	What are some of the best things about having social media? (ask for specific examples)	What are some of the most challenging or difficult parts of being part of these forums? Do you find it difficult not to visit these forums and maybe worried about losing touch?
Relations	Have you ever seen content on social media that made you concerned for their safety or the safety of others? (Elaborate with; What was it about the post that made you concerned?)	What do you think about what you encounter? How does it affect you? Who do you think sees your posts? What do you think they think about what you post? What do you think about what you post? How do you get to know each other? How do you experience your relationships through these forums? How do you feel the trust between you? How much do you share about yourself? How much do you feel others share? What would you call your relationship? What do you find on social media that you don't find in real life? Ask for concrete examples of the unique things the participant finds in online forums. How do you find that you support each other? Do you sometimes find that you're not so good for each other? Can you give some examples of when you feel you are good for each other? Can you give some examples of when you feel that you are less good for each other? (Ask for inspiration for new methods/feeling that seeing posts can worsen your own condition, etc.)
Impacts	How do you feel being on social media affects you when it is used for self-harm purposes? (Ask for elaboration if necessary)	Do you ever experience suicidal thoughts in connection with this? (Please elaborate and ask about any previous suicide attempts) Do you ever find that the use of social media affects your physical self-harm for better or worse? Negative (e.g. choice of method/use of method/encouragement) Positive (e.g. deterrent/someone to talk to/relief of feelings/self-help/referral to help)
The word of the participant	What do you believe people need to know about taking part in online forums? (What is it we do not seem to know or seem to understand?)	Is there anything I haven’t asked you? Is there anything you want to add?

The end of the interview

Thank you so much for helping me learn more.

I am very grateful for that.

How do you feel now that we've sat down and talked about some difficult things?

(Follow up with the participant, supportive conversation, possibly give the number of the psychiatric emergency room)

Please feel free to contact me if you need to add more or if you don't want to join anyway – it is always a possibility.

Appendix B: Consolidated criteria for reporting qualitative studies (COREQ): 32-item checklist

Developed from:

Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care. 2007. Volume 19, Number 6: pp. 349–357

No. item	Guide questions/description	Reported on page no.
Domain 1: Research team and reflexivity
Personal characteristics
1. Inter viewer/facilitator	Which author/s conducted the interview or focus group?	P. 5
2. Credentials	What were the researcher's credentials? E.g. PhD, MD	P. 5
3. Occupation	What was their occupation at the time of the study?	P. 5
4. Gender	Was the researcher male or female?	P. 5
5. Experience and training	What experience or training did the researcher have?	P. 5
Relationship with participants
6. Relationship established	Was a relationship established prior to study commencement?	P. 5
7. Participant knowledge of the interviewer	What did the participants know about the researcher? e.g. personal goals, reasons for doing the research	P. 5
8. Interviewer characteristics	What characteristics were reported about the interviewer/facilitator? e.g. bias, assumptions, reasons, and interests in the research topic	P. 5
Domain 2: Study design
Theoretical framework
9. Methodological orientation and Theory	What methodological orientation was stated to underpin the study? e.g. grounded theory, discourse analysis, ethnography, phenomenology, content analysis	N/A
Participant selection
10. Sampling	How were participants selected? e.g. purposive, convenience, consecutive, snowball	P. 3
11. Method of approach	How were participants approached? e.g. face-to-face, telephone, mail, email	P. 3
12. Sample size	How many participants were in the study?	P. 4
13. Non-participation	How many people refused to participate or dropped out? Reasons?	P. 4
Setting
14. Setting of data collection	Where was the data collected? e.g. home, clinic, workplace	P. 5
15. Presence of non-participants	Was anyone else present besides the participants and researchers?	P. 5
16. Description of sample	What are the important characteristics of the sample? e.g. demographic data, date	P. 4
Data collection
17. Interview guide	Were questions, prompts, guides provided by the authors? Was it pilot tested?	P. 5
18. Repeat interviews	Were repeat inter views carried out? If yes, how many?	N/A
19. Audio/visual recording	Did the research use audio or visual recording to collect the data?	P. 5
20. Field notes	Were field notes made during and/or after the interview or focus group?	N/A
21. Duration	What was the duration of the inter views or focus group?	P. 5
22. Data saturation	Was data saturation discussed?	P. 16
23. Transcripts returned	Were transcripts returned to participants for comment and/or correction?	N/A
Domain 3: Analysis and findings
Data analysis
24. Number of data coders	How many data coders coded the data?	P. 5
25. Description of the coding tree	Did authors provide a description of the coding tree?	P. 6
26. Derivation of themes	Were themes identified in advance or derived from the data?	P. 5
27. Software	What software, if applicable, was used to manage the data?	P. 5
28. Participant checking	Did participants provide feedback on the findings?	N/A
Reporting
29. Quotations presented	Were participant quotations presented to illustrate the themes/findings? Was each quotation identified? e.g. participant number	Pp. 7–13
30. Data and findings consistent	Was there consistency between the data presented and the findings?	P. 13
31. Clarity of major themes	Were major themes clearly presented in the findings?	P. 7
32. Clarity of minor themes	Is there a description of diverse cases or discussion of minor themes?	N/A