A recent study published by researchers at the University of Oxford’s Department of Psychiatry found that one in three young people had seen self-harm content online in the previous month. Among those exposed, nearly 55% had encountered it once or twice, 26.5% a few times and 18.5% had seen self-harm content several times.

The research team, led by our new Researcher of the Month, Dr Holly Bear, analysed data from more than 32,000 students aged 11–18 who took part in the 2025 OxWell Student Survey in England. This is a large school-based survey which runs every two years across England, with participating schools concentrated particularly in Oxfordshire and Liverpool. It asks school students directly about their mental health, wellbeing and their lives more broadly, both online and offline, and provides a rich snapshot of what is happening in young people's lives at that moment.

Nearly three-quarters (72.9%) of those exposed had encountered self-harm content through passive routes, without actively seeking it. Algorithmic feed suggestions were the most commonly reported route overall. Even many of the younger survey respondents had encountered content (around one in four 11-12 year olds in the previous month) despite most social media platforms currently having a minimum age of 13.

Summary

Previous research has raised concerns about young people's exposure to self-harm content online, but Dr Bear and colleagues wanted to move beyond the question of simply whether young people see it. Within the Oxwell Student Survey, they asked how common and frequent is recent exposure, and how are young people coming across this material?

The term ‘self-harm content’ was deliberately left broad. Rather than imposing a strict definition, the researchers were interested in what young people themselves understood to be self-harm content. Previous research suggests that ‘self-harm content’ encompasses a wide range of material, from graphic images of self-harm, to recovery stories or peer support. We also know that supportive or recovery-focused material may sit alongside more distressing content in the same feed. The researchers didn't collect information about the specific nature of the content encountered, so couldn't distinguish between potentially harmful material and content that may be more neutral or supportive.

Young people were given a list of ways that they might have encountered self-harm content and could select as many as applied to them. Some routes were more passive. For these young people, content might simply appear in their feed, be embedded within unrelated content or be encountered accidentally. Other routes to exposure were more active. This included searching for content themselves or having it sent to them by someone else.

Passive exposure was extremely common. Nearly three-quarters of the young people who had encountered self-harm content hadn't actively sought it out.

Algorithmic suggestions were the most commonly reported route to exposure, with 66.2% of young people who answered the question about how they encountered self-harm content reporting that it had appeared in their feed. We have very little concrete information from platforms about exactly how their personalised algorithms work. Feeds seem to be tailored according to numerous signals, including what each user likes and engages with, who they follow, and what people in their networks are watching and sharing. These systems are likely designed, at least in part, to encourage continued engagement with the platform.

Without access to platform data or greater transparency about how content is selected and recommended, it is extremely difficult for independent researchers to understand why particular content appears in a young person’s feed, how exposure to potentially harmful material develops over time or whether platform safety measures are working to reduce harm. However, one hypothesis in the paper is that whilst a young person might not have searched for self-harm content, interactions with related posts about mental health, body image or emotional distress could influence feed recommendations.

Oxwell Student Survey data also found that around 12% of young people had used social media for mental health support in the past month, making it the most commonly reported form of online mental health support in the survey. Around one in five young people had used at least one form of online support.

Actively seeking self-harm content was far less common. Among young people who had encountered this material, 2.7% had exclusively searched for it themselves and 3% had exclusively received it from others.

When those who had encountered content through multiple routes were included, 7.5% reported actively searching for self-harm content and 8.1% reported having it sent to them by someone else.

More than 40% of young people in both the active-search-only and mixed-exposure groups reported encountering self-harm content several times during the previous month, more frequently than young people with only passive exposure. These young people were also more likely to report struggling in other ways. This includes reporting higher levels of symptoms of depression and loneliness, exposure to other online harms, and bullying, suggesting that this much smaller group may include young people who are already experiencing significant distress.

Interestingly, the researchers also identified differences between young people who searched for self-harm content and those who had it sent to them. Searching was particularly strongly associated with symptoms of depression, whereas being sent content was more closely associated with experiences of online aggression and coercion. It's possible that, for some young people, receiving self-harm content from others may form part of a wider pattern of interpersonal harm or bullying, rather than being an isolated experience

For these young people, the priority should be to ensure that searching for or encountering self-harm content provides a route towards appropriate support, rather than simply more of the same. Platforms could, for example, signpost young people towards sources of help or recovery-focused content. Parents and schools also need to feel confident recognising when a young person may be struggling and starting supportive conversations, so that encountering this content can become an opportunity to connect them with help.

Strikingly, every pathway to exposure identified in the study (passive or active) was associated with greater psychosocial vulnerability than no exposure, although the patterns differed. Across all pathways, exposure was associated with factors including loneliness, anxiety and depression symptoms and experiences of other online harms.

It is important to note that this is a cross-sectional study, meaning that it can identify associations, but not causality. We don't know the direction of these associations. It may be that young people who are already struggling are more likely to encounter this content, that encountering it influences how they feel, or that both processes are operating together.

You can find out more about the study and findings in this video on YouTube.

Implications

Tips for parents and educators

“I think the most important thing is how we respond to that, so staying calm, leading with curiosity and care rather than judgement, ask how they're doing, not just what they've been looking at.”

Be curious about your child’s digital life in general. Self-harm content is only one type of potentially harmful content (others might include violent content or body image content, for example). Take a genuine interest in their digital diet. Ask what’s been showing up in their feeds lately. Has anything surprised them? Have they seen anything that they didn't want to see? How did that make them feel? If a young person tells you that they have encountered self-harm content, stay calm and lead with curiosity and care rather than judgement. Ask how they are doing and how they feel, rather than focusing solely on what they have been looking at.

Try to learn more together about how feeds work. Talk to young people about how content reaches them online and the ways they can influence what appears in their feeds. Remind them that much of what they see has been selected by recommendation systems, so if something upsetting or embarrassing appears, it isn't necessarily a reflection of what they have deliberately searched for or chosen to see. Learning about these systems together can make it easier to talk openly about uncomfortable content when it does appear.

Be open about your own feed. Chatting with your child about what’s on your own phone might help to open up conversations and encourage dialogue.

Coming across self-harm content is a useful prompt to stay curious about how a young person is doing and keep conversations open. We shouldn't assume that exposure is insignificant simply because they weren't actively looking for the content.

Try not to assume the worst. Actively seeking this content doesn’t necessarily mean looking at graphic content. It might be that your teen has looked for supportive content from others.

Avoid making restriction your immediate response for either active or passive exposure. Taking a young person’s phone away or completely restricting their access might feel like the natural thing to do, but it could also cut them off from sources of support and make them less likely to tell you about something worrying they encounter in future. Try first to understand what is happening and what the young person needs.

Help them access appropriate support. For a young person who has actively sought self-harm material, suggest alternative support pathways. Depending on what is happening and how the young person is feeling, this might involve support through school, their GP or mental health services. The priority is to understand what is going on, respond supportively and help them access further support where needed.

Don’t be afraid to ask directly about self-harm. If you are concerned that a young person may be thinking about self-harm or may already have harmed themselves, it is OK to ask. Research suggests that asking about self-harm does not put the idea into a young person’s head or encourage the behaviour. Ask openly, calmly and kindly. Knowing what is happening puts you in a better position to listen, offer reassurance and help them access appropriate support.

Take it seriously if someone else is sending content. If a young person tells you that another person has sent them self-harm content, try to understand the wider context. Ask who sent it, what their relationship with that person is like, how receiving the content made them feel and whether anything else concerning is happening online or at school. For schools, this may need to be considered as a safeguarding concern. Being sent self-harm content could form part of a wider picture involving peer relationships or cyberbullying, so it is important to consider what else may be going on alongside practical steps such as blocking and reporting the content.

Support young people to curate feeds that feel supportive to them. Platforms are introducing more tools that allow users to influence what appears in their feeds, although we don't yet know how effective these are. Most platforms allow users to indicate that particular posts don't interest them, report content, mute or block accounts, or filter certain words. Some also allow users to reset their recommendations and effectively start again. Encourage young people to use these tools proactively, rather than assuming that they simply have to accept whatever their feed serves them. These features can be hard to find and aren’t necessarily intuitive.

Remember that online spaces can be a real source of connection, identity and self-expression for young people. The aim shouldn't simply be to teach young people to fear online spaces or avoid them altogether. We can also help them learn how to navigate these environments safely, make informed choices and develop the skills they need to thrive online.

Digital literacy education needs to move beyond understanding. Most young people have a fairly good understanding of the digital world and how algorithms work. It’s important to build on this knowledge, to explore how we might be able to reshape our feeds, what to do if something is upsetting, who to talk to if support is needed.

Make use of helpful resources. The Molly Rose Foundation has some excellent resources for parents, educators and young people.

If you are worried about your child’s mental health, you must visit your GP as soon as possible. If you think that your child is at risk of harm, you must contact support services immediately. Remember that our resources are not a replacement for clinical support.