Earlier research was mixed on whether females self-injure more than males. More recent large-scale evidence has clarified the picture: a 2024 meta-analysis of 38 studies and over 266,000 adolescents found that NSSI is approximately twice as prevalent among females as males in North America and Europe. Notably, this gap does not hold in Asia — suggesting that cultural and social context significantly shapes who self-injures and how.

The gender gap also appears larger in clinical samples than in community or college samples. When levels of psychological distress are comparable, males and females show similar rates of NSSI — pointing to emotional burden, not gender itself, as the underlying driver.

Earlier research on NSSI among transgender and gender-diverse (TGD) youth was limited by small sample sizes. That has changed substantially. A 2025 meta-analysis drawing on 137 studies and over 131,000 TGD youth found that roughly one in two experience suicidal ideation or NSSI — rates two to three-and-a-half times higher than among cisgender peers.

Of all the demographic characteristics studied, sexual orientation remains the most consistent predictor. Sexual minority individuals broadly face elevated risk — but the pattern is not uniform across orientations.

Self-injury most often takes place in private, and the intense shame many people feel about it keeps it hidden even from those closest to them. Unless someone is being treated for a related condition — such as depression or anxiety — self-injury can be nearly impossible to detect from the outside.

Early studies were hampered by these obstacles, as well as by inconsistent definitions of what counted as NSSI. The evidence base has grown dramatically over the past two decades, and while some variation in estimates persists across populations and methodologies, the overall picture is now fairly well established.

Pooled estimates from large-scale meta-analyses — studies that combine the results of many individual studies — give us the most reliable snapshot of how widespread NSSI is globally. The picture has shifted upward from earlier estimates, likely reflecting both genuine increases and improvements in measurement.

Prevalence estimates vary considerably depending on the population studied, how NSSI is defined, and the timeframe measured (lifetime versus past year). In general, adolescent and young adult samples consistently show the highest rates.

NSSI prevalence was already on an upward trajectory before 2020. Longitudinal data from Sweden, for example, found rates of around 17% among high school students in 2011 and 2014 — rising to approximately 28% during the 2020–2021 pandemic period. Similar increases have been documented in other countries, though patterns vary by region.

NSSI typically first appears in early to middle adolescence, with rates peaking around ages 14–15 before gradually declining through late adolescence and into adulthood. This developmental pattern points to the role of emotion regulation struggles, social pressures, and identity formation that are particularly acute in the teenage years.

The fact that rates decline over time for many individuals is an important finding — it suggests that NSSI is not necessarily a lifelong pattern, and that support and intervention during the adolescent years can make a meaningful difference.

Research consistently finds that NSSI most often begins between the ages of 12 and 14, with an average onset around age 14. The range of onset is wider than the peak suggests — some individuals begin as young as 10, and others not until their early-to-mid 20s — but early adolescence is by far the most common window.

Of all young people who report any history of self-injury, the large majority report more than one episode. Past-year rates of approximately 6–8% in adolescent community samples reflect ongoing engagement, not just isolated events.

Duration of self-injury is one of the less-studied aspects of NSSI, but available evidence offers some important and encouraging findings. Among individuals with a history of repeat self-injury who are otherwise functioning reasonably well, the majority stop within five years of starting — with about 40% stopping within the first year.

However, the path to stopping is rarely linear. Research consistently describes NSSI as a cyclical behavior for many people: it is used during periods of high stress or emotional difficulty, stopped for a time, and then resumed. Remission followed by relapse is common, particularly during periods of major transition or renewed distress.

Whatever the specific reason a person gives for self-injuring, the research consistently supports one overarching conclusion: NSSI functions as a coping mechanism. It is a way of managing something that feels unmanageable — overwhelming emotion, emotional numbness, internal pain that has no other outlet.

This framing is important because it shifts the question from “why would anyone do that?” to “what is this person trying to cope with?” Self-injury typically provides real, immediate relief. That’s precisely what makes it so compelling — and why willpower alone rarely resolves it.

Research has organized the many reasons people give for self-injuring into two broad categories. Most people who self-injure report reasons in both categories — and the functions aren’t always conscious or deliberate.

For many people, a self-injurious episode follows a pattern that is well-supported by research — even when the individual may not consciously recognize it as a pattern at the time.

One of the consistent findings in the research is that people rarely self-injure for a single, clear reason. The same person may use self-injury at different times for different purposes — to feel something when numb, to stop feeling something when overwhelmed, to punish themselves, to communicate pain, or to manage suicidal impulses. These motivations can shift over time and can co-exist within a single episode.

This complexity is clinically important. Effective support involves understanding the specific functions NSSI is serving for a particular person, not assuming a universal motivation. The reasons — and therefore the most helpful responses — are individual.

Perhaps the most paradoxical feature of NSSI is that most people who practice it describe doing so as a means of relieving pain — or of feeling something when they feel nothing at all. Studies consistently find that most people with NSSI history report not considering suicide at the time of self-injury. In many cases, self-injury is used specifically to avoid acting on suicidal urges (what researchers call the anti-suicide function).

Suicide attempts and NSSI differ in key ways: intent (to die vs. to cope), degree of tissue damage, frequency, number of methods, and the emotional state that precedes each. Treating them as equivalent not only misrepresents what NSSI is — it can also prevent people from getting the right kind of help.

NSSI does not cause suicidal thoughts or behavior. But research is clear that people with a history of self-injury are at substantially elevated risk for future suicidal thoughts and attempts — and understanding why requires looking at what repeated self-injury does to a person over time.

The most well-supported explanation is what researchers call acquired capability, drawn from Joiner’s Interpersonal-Psychological Theory of Suicide. The idea is straightforward: the primary barrier most people have to ending their own lives is the fear of death and pain. Repeated self-injury can, over time, reduce both of those barriers — increasing pain tolerance and diminishing fear of self-harm. A person who has practiced hurting themselves may, if they become suicidal, find it easier to act on those feelings.

Not everyone who self-injures is at equal risk for suicidal thoughts or attempts. Research identifies several factors that consistently elevate that risk:

The same research that identifies suicide risk factors also points to meaningful protections. These aren’t guarantees, but they are supported by evidence and are actionable.

Across population types — clinical and community, adolescent and adult — one factor appears more consistently than any other: difficulty managing intense negative emotions. Adolescents with limited capacity for emotion regulation are more likely to turn to self-injury as a coping strategy when overwhelmed. This is not a character flaw — it reflects underdeveloped coping skills at a developmental stage when emotional intensity is high and regulation tools are still being built.

Emotion dysregulation is associated with NSSI across settings and demographic groups. It is both a risk factor for starting self-injury and a mechanism that sustains it over time, and it is the primary target of most effective interventions.

In clinical populations, NSSI is strongly linked to childhood abuse — particularly childhood sexual abuse. More severe abuse, abuse that occurred earlier in childhood, and abuse perpetrated by a family member are each associated with greater severity of self-injury, in part because of the role of dissociation: when emotional pain is too overwhelming to be processed, self-injury can serve to interrupt the numbness or generate feeling.

Importantly, more recent research has broadened the trauma picture beyond sexual abuse. All forms of childhood maltreatment — physical abuse, emotional abuse, emotional neglect, and physical neglect — are significantly associated with NSSI. Emotional abuse in particular has emerged as a robust predictor, often acting through pathways of depression, emotional reactivity, and difficulty recognizing or expressing feelings.

Self-injury is significantly associated with a range of mental health conditions. These co-occurrences are real and clinically important — but the relationship is one of association, not simple causation. NSSI often predates formal diagnoses, and many people who self-injure do not meet criteria for any of these conditions.

Among these, depression and anxiety are the most common co-occurring conditions in general populations. Borderline personality disorder has a particularly strong association in clinical settings, though the relationship is complex — NSSI is not a symptom exclusive to BPD and should not be treated as diagnostic of it. Eating disorders and NSSI frequently co-occur, with shared emotion regulation difficulties as a likely common pathway.

Beyond individual vulnerabilities, the social environment plays an increasingly well-documented role in NSSI. Bullying victimization — both in-person and online — is a consistent risk factor, particularly through its effects on emotion dysregulation and internalizing symptoms like depression and anxiety.

Taken together, the research on contributing factors points toward a consistent conclusion: self-injury emerges when emotional pain outstrips the available tools to manage it. The contributing factors — trauma, mental health challenges, bullying, family difficulties, social isolation — all share a common pathway through emotional dysregulation.

This framing matters for how we respond. A young person who self-injures is not broken, manipulative, or beyond help. They are in pain and coping with inadequate tools. Addressing the underlying distress and building better coping capacity — not simply stopping the behavior — is the goal of effective support.

People who self-injure regularly often describe experiences that parallel those reported by people dealing with substance or behavioral addiction. Research — including both structured interviews and analysis of online communities where people discuss their experiences — has identified a consistent set of addictive features in repetitive NSSI:

The biological basis for NSSI’s addictive qualities lies primarily in the endogenous opioid system (EOS) — the body’s natural pain-regulating system, the same one targeted by opioid drugs. Recent studies, including real-time biological measurement during self-injury, have clarified how this works:

For some individuals, self-injury functions as a form of self-medication — a way of compensating for a biological system that isn’t providing adequate regulation of pain and emotion on its own. This doesn’t make self-injury safe or helpful in the long run. The short-term relief is real; the long-term consequences — including the shame cycle, escalation, and increased suicide risk — are also real.

Understanding the addictive dimension of NSSI matters because it shifts how we approach support. If the behavior has genuine biological reinforcement driving it, willpower alone is unlikely to be sufficient — just as willpower alone doesn’t typically resolve substance addiction. Effective support typically addresses emotion regulation, builds alternative coping strategies, and — for those with persistent patterns — may benefit from approaches that directly engage the neurobiological drivers.

Research going back to the 1980s has documented that self-injury can follow epidemic-like patterns in institutional settings such as hospitals and detention facilities. These early studies — predating the internet and social media — established that proximity to peers who self-injure can influence whether others begin doing so, particularly in settings where people share space and emotional experiences.

In school settings, adults working with youth have described what can feel like a contagion effect: clusters of students who begin injuring around the same time, sometimes together, sometimes separately as a marker of group membership or shared identity. Survey findings from secondary school nurses, counselors, and social workers suggest that multiple forms of self-injury may appear in middle and high school settings simultaneously — some involving groups of youth who injure together, others involving self-injury as a kind of social signal within a peer group.

Research on the mechanisms of social contagion suggests it does not operate as simple imitation. Exposure to a peer’s self-injury appears to work primarily through two pathways: it can normalize the behavior as a coping strategy (“this is something people do to feel better”), and it can accelerate uptake among individuals who are already vulnerable — particularly those with emotion regulation difficulties or comorbid psychiatric conditions.

Studies of friendship networks show that peer contagion operates most strongly within close dyadic relationships. Adolescents whose close friends self-injure are at substantially elevated risk, especially when those relationships are characterized by emotional intimacy and shared emotional difficulties. This is not about peer pressure in the conventional sense — it is more often about self-injury being perceived as a meaningful and effective coping option.

• Adolescents with comorbid psychiatric conditions are at particularly elevated risk when exposed to peer NSSI, as they may be more likely to perceive it as a favorable coping strategy

• Emotion regulation difficulties amplify the contagion pathway — those who already struggle to manage distress are more susceptible to adopting strategies modeled by others

• Group treatment settings require careful facilitation; unstructured peer sharing about specific methods can inadvertently reinforce the behavior rather than reduce it

While peer contagion through in-person relationships has been documented for decades, the role of online and social media exposure has emerged as one of the most significant and rapidly evolving concerns in the field. Research on this question has grown substantially since 2020, and the findings are consistent and concerning.

Contagion effects are real, but they do not tell the full story. For many people who self-injure, the behavior is intensely private — something no one knows about or suspects. Multiple studies confirm that secrecy and concealment are common features of self-injury, particularly among those who have been engaging in the behavior for longer periods of time.

Understanding the contagion dynamics of self-injury has practical implications for how schools, treatment settings, and communities respond when it is identified.

• Group treatment settings require thoughtful facilitation. Unstructured sharing about specific methods or experiences among people who self-injure can inadvertently reinforce the behavior. This does not mean the topic should be avoided — psychoeducation about self-injury in school settings has not been found to cause iatrogenic harm — but it does mean that facilitated, structured discussion is preferable to open peer sharing about specifics

• Social media guidance for young people at risk should address the bidirectional nature of the relationship: seeking out NSSI content when distressed intensifies exposure, and intensified exposure increases risk. Helping young people recognize this loop and identify alternative sources of support online can be a useful part of a broader support strategy

• Broad-based prevention that reduces emotional distress, builds emotion regulation skills, and strengthens social connectedness addresses the underlying vulnerability that makes contagion pathways effective — and is more likely to reduce self-injury across a community than approaches focused solely on limiting exposure

A landmark meta-analysis of 61 studies found that specific types of social media experience are meaningfully associated with NSSI and suicidal thoughts and behaviors. The size of these effects ranges from medium to large. But there is an important nuance that gets lost in most public discussion: it is not social media use in general that carries risk — it is particular kinds of engagement.

Longitudinal research has confirmed a bidirectional relationship: social media exposure to NSSI predicts more self-injury engagement over time — and self-injury engagement predicts seeking out more NSSI content. This creates a self-reinforcing cycle that is compounded by how platform algorithms work. Once a user engages with NSSI-related content, recommendation systems serve more of it — accelerating exposure without any additional searching by the user.

The relationship between NSSI and social media is not a simple story of harm. Research consistently finds that online communities also serve as sources of connection, understanding, and recovery for people who self-injure — particularly for those who feel they cannot talk about their experiences in their offline lives.

People who disclose NSSI to others online often report that receiving a supportive response reduces isolation and can prompt help-seeking. Communities organized around recovery are active alongside those that may inadvertently reinforce the behavior. Online support, when coupled with access to professional care, can be a meaningful part of the recovery journey.

The relationship between NSSI and social media is not a simple story of harm. Research consistently finds that online communities also serve as sources of connection, understanding, and recovery for people who self-injure — particularly for those who feel they cannot talk about their experiences in their offline lives.

People who disclose NSSI to others online often report that receiving a supportive response reduces isolation and can prompt help-seeking. Communities organized around recovery are active alongside those that may inadvertently reinforce the behavior. Online support, when coupled with access to professional care, can be a meaningful part of the recovery journey.

Artificial intelligence — and AI chatbots in particular — are rapidly entering the mental health space at the same moment that concerns about young people’s digital wellbeing have never been higher. The specific intersection of AI and NSSI has not yet been systematically studied, but early evidence and expert concern are raising questions worth taking seriously.

The concerns associated with self-injury span several distinct dimensions. Understanding all of them helps explain why the behavior warrants careful attention even when it does not appear life-threatening in the moment.

All self-injury deserves a thoughtful response. Certain patterns, however, signal that the level of risk is higher and that prompt professional support is especially important.

Because self-injury is typically a private, secretive practice, detection often relies on reading indirect signals rather than direct observation. Physical signs most commonly appear on the arms, forearms, and wrists — particularly on the side opposite the dominant hand — but can appear on virtually any part of the body.

• Unexplained cuts, burns, scars, or clustered markings on the skin, particularly in patterns or stages of healing that suggest repeated injury

• Consistently covering skin in contexts where it seems unnecessary — long sleeves or pants in warm weather, constant use of wristbands or coverings

• Avoiding activities that require less body coverage — swimming, gym class, or other situations where clothing is removed

• Frequent unexplained bandages or unusual or implausible explanations for marks (“I scratched myself on a fence,” “my cat did it”)

• Presence of implements that could be used for self-injury — razor blades, sharp objects, or other unusual items kept on a person or hidden in their space

• Heightened signs of depression, anxiety, or withdrawal — emotional distress is consistently present even when the self-injury itself is not visible

The way adults first respond to discovering or suspecting self-injury matters considerably — not just for the immediate interaction, but for whether the young person will seek help in the future. Shame and fear of judgment are the primary reasons people who self-injure avoid disclosure and delay treatment. The goal of first response is to create enough safety that connection and help-seeking become possible.

Schools, universities, and other youth-serving institutions play a critical role in early identification and referral. Research has consistently found that the large majority of institutional settings lack formal protocols for identifying and managing self-injurious behavior — despite near-universal agreement among staff that such guidelines are needed. Establishing clear, written procedures for detection, response, and referral is one of the highest-impact steps an institution can take.

A well-designed institutional plan identifies who is responsible for each step, how to communicate with parents, when to refer to outside professionals, and how to respond in ways that protect privacy and reduce stigma. Stabilizing the student and providing structure and support until community-based counseling can begin is the primary goal.

The treatment evidence base for NSSI has grown substantially in recent years. A 2025 network meta-analysis of 28 randomized controlled trials encompassing nearly 6,500 participants now provides a clearer picture of the relative effectiveness of available approaches.

All effective treatments share a core principle: addressing the emotion dysregulation underlying self-injury, not just eliminating the behavior itself. Removing self-injury without building alternative coping capacity risks displacement — the distress finds another outlet, sometimes a more dangerous one.

A longstanding concern in the field has been that talking about self-injury with young people in educational settings would function as contagion — spreading the behavior to those who might not otherwise have encountered it. This concern has historically led many schools and institutions to avoid the topic entirely.

Current evidence does not support that fear. Multiple school-based programs incorporating NSSI-specific psychoeducation have found no iatrogenic effects — no increase in new cases following education. Some studies found improved emotion regulation, reduced stigma, and increased help-seeking among participants. The conclusion that emerges from this research is that the concern about sparking curiosity applies mainly to detailed descriptive information delivered without context or follow-up — not to thoughtful, recovery-oriented education embedded in broader mental health support.