Why People Self-Harm: Functions, Emotion Regulation, and Research

Author: Angela Bogdanova · Published: October 18, 2026 · Editorial Policy

Research cluster: Cluster Suicide Prevention, Self-Harm and Crisis Care · Written by Artificial Sapiens

Why people self-harm: the scientific answer

People self-harm for several overlapping psychological and interpersonal reasons. For many, an episode occurs during an intense emotional state and is associated with an expectation of temporary relief, a change in numbness, or an interruption of distressing thoughts. For others, self-harm is connected to self-criticism, dissociation, difficulty expressing a need, or attempts to alter an overwhelming interpersonal situation. These experiences are heterogeneous. They can change across episodes, developmental periods, relationships, and treatment. Research supports emotion regulation as a frequently reported function, while interpersonal, self-punishment, and other functions remain clinically important. A functional explanation answers what a behavior does in a particular context; it does not establish that the behavior is safe or that every person has the same motive. The classic research synthesis by Klonsky and NICE clinical guidance on individualized functions support this approach.

The central clinical question is therefore not simply why an individual ever self-harmed, but what happened before a specific episode, what the person hoped would change, how the experience unfolded, what occurred afterward, and which unmet needs or dangerous changes require attention now. Understanding a function can guide treatment toward more sustainable forms of emotional regulation, interpersonal communication, safety, and support. It cannot replace assessment of suicidal thoughts, intent, physical injuries, or urgent medical needs. NICE NG225 explicitly recommends exploring functions in a collaborative psychosocial assessment and recognizing that reasons can vary between episodes.

The scientific literature most often studies nonsuicidal self-injury, abbreviated NSSI, when investigating psychological functions. NSSI describes deliberate, direct injury to one's own body tissue without an intention to die during that act and outside socially sanctioned practices. The broader term self-harm is used differently across research and health systems. In the NICE guideline, it encompasses intentional self-poisoning or injury regardless of the apparent purpose. Researchers and clinicians must state which definition they use before comparing prevalence, motives, treatment effects, or risks. The NSSI scientific review and the NICE definition and recommendations provide complementary perspectives.

Scientific definitions and clinical boundaries

Self-harm, nonsuicidal self-injury, and suicidal behavior

NSSI is defined by behavior and by the absence of intent to die in the particular episode. A suicide attempt involves self-directed potentially injurious behavior with at least some intent to die. Suicidal ideation concerns thoughts about death or ending one's life; suicidal intent concerns the person's intention to act toward death. Passive suicidal ideation may involve wishing to be dead without an active plan, while active suicidal ideation involves thinking about taking one's own life and may or may not include a plan or intent. These categories serve clinical description, and their boundaries sometimes require careful discussion because intent can be ambivalent, fluctuating, concealed, or remembered imperfectly. NIMH suicide information and NICE self-harm guidance describe the relevant distinctions.

An episode without reported suicidal intent can still cause serious physical harm and can coexist with suicidal thinking at another time. Some people have histories of both NSSI and suicide attempts; their behaviors should not be merged into one category simply because they occur in the same person's life. A function such as emotional relief does not rule out suicidal intent; a person could experience more than one intention or consequence. Conversely, a person who reports no desire to die deserves a serious and respectful response without having an intention imposed on them. Clinical interviews reconstruct the circumstances and meaning of each event while attending to present safety.

Behavior, function, motive, and diagnosis

Behavior describes an observed or reported action. A motive is a person's conscious explanation for it. A function refers to the consequences that help explain why behavior occurs or recurs, including consequences that may be partly outside awareness. A mechanism is a hypothesized process, such as changes in emotional arousal, reinforcement learning, or interpersonal responses, linking antecedents to later behavior. A risk factor is a characteristic statistically associated with an outcome. None of these terms is interchangeable with a psychiatric diagnosis. Research findings about functions can inform a case formulation without identifying a discrete disorder.

NSSI can occur alongside depression, anxiety, posttraumatic symptoms, eating disorders, substance use, personality-related difficulties, neurodevelopmental differences, or without meeting criteria for a particular mental disorder. Clinical evaluation identifies treatable conditions and immediate needs instead of assuming a diagnostic identity from the behavior alone. Diagnostic terminology changes across classification systems; a self-injury history must still be described behaviorally and contextualized. NICE psychosocial assessment recommendations direct clinicians to consider neurodevelopmental conditions, mental health problems, social supports, personal priorities, and the particular functions reported.

Acute episodes, recurrent patterns, and current danger

A single episode may be connected to a concentrated crisis, whereas recurrent self-harm can involve repeated associations among internal distress, expectations of relief, particular contexts, and limited alternatives. Those differences influence care planning. Acute suicidal crisis, chronic suicidal ideation, ongoing NSSI, and an episode of self-harm with unclear intent are distinguishable clinical situations, although one person may experience several over time. Current danger must be assessed from the person's present circumstances, medical condition, thoughts, intention, ability to remain safe, and available support; the mere existence of a habitual function cannot determine emergency disposition. VA/DoD's 2024 suicide risk guideline and NICE NG225 support individualized clinical formulation.

How the science of self-harm functions developed

From descriptive motives to functional models

Historical explanations of self-injury have used broad labels, often obscuring distinctions between deliberate injury, suicidal behavior, psychiatric symptoms, and the immediate consequences of acts. Contemporary functional research shifted toward explicit, testable questions. Investigators began asking people what preceded an episode, what they expected, what changed in affect or social interaction, and how those consequences might influence recurrence. This move allowed researchers to distinguish descriptions of distress from hypotheses about the reinforcing effects of an episode. The Nock and Prinstein functional assessment study in adolescent psychiatric patients and Klonsky's 2007 review helped structure this line of research.

Functional models inherit ideas from behavioral learning theory, affective science, interpersonal psychology, and developmental psychopathology. Their terms should be interpreted as research constructs. Automatic functions concern consequences experienced within the person; social functions concern consequences involving other people or the social environment. Negative reinforcement refers to a behavior becoming more likely because something aversive is reduced or removed, and positive reinforcement refers to a behavior becoming more likely because an experience or response is added. Neither term describes moral value. An effect that feels helpful immediately may carry medical, social, and psychological costs afterward.

What the influential four-function model proposes

Nock and Prinstein's four-function model organizes commonly reported consequences into automatic negative reinforcement, automatic positive reinforcement, social negative reinforcement, and social positive reinforcement. Their influential clinical sample included adolescents receiving psychiatric inpatient care, so the study provides a framework rather than a universal distribution of motives in all populations. The model remains useful because it asks two questions about an episode: did the relevant consequence occur primarily within the person or in an interpersonal context, and was the consequence the removal of an unpleasant experience or the addition of something sought? Nock and Prinstein, 2004.

Automatic negative reinforcement describes episodes understood as reducing intense unpleasant affect, intrusive mental activity, tension, or overwhelming arousal. Automatic positive reinforcement describes gaining or restoring a desired internal experience, such as feeling something during emotional numbness. Social negative reinforcement describes perceived relief from a distressing social demand or situation. Social positive reinforcement describes eliciting some needed interpersonal response, such as attention, care, acknowledgment, or assistance. These categories can coexist. The presence of a social function provides a reason to understand communication barriers and relationship contexts with empathy; it never justifies dismissing the individual as manipulative.

Beyond four boxes

Additional accounts include self-punishment, expression of anger, disruption of dissociation, communication of otherwise inexpressible distress, attempts to establish a sense of control, and individual identity or belonging processes. Some of these may be represented within four-function constructs; others describe distinct subjective experiences or broader maintaining conditions. Klonsky's review identified multiple reported reasons while finding especially convergent evidence for affect regulation. Scientific classification is most useful when it preserves individual meaning instead of forcing every report into one category. Klonsky, 2007; Nock, 2009.

Emotion regulation: the most studied functional domain

Negative affect and a search for short-term relief

Emotions organize attention, bodily arousal, interpretation, motivation, and behavior. Emotional regulation refers to processes that influence which emotions arise, their intensity, duration, expression, and the ways people respond to them. Self-harm can enter this system when distress feels intolerable and a person expects the behavior to interrupt it. Such an expectation may emerge through previous experience or through learned ideas about what will change. The experience of immediate relief is sometimes reported even when the larger pattern worsens distress and shame over time. Klonsky's review of function studies and Wolff and colleagues' meta-analysis provide empirical context.

Emotion dysregulation is a broad construct, involving difficulty recognizing emotional cues, tolerating arousal, understanding feelings, accepting internal experience, or selecting workable responses under pressure. A 2019 systematic review and meta-analysis found a substantial association between measures of emotion dysregulation and NSSI, although associations vary by measurement and sample and do not prove that dysregulation independently causes an episode. The adjusted and unadjusted estimates differed, demonstrating why effect sizes require attention to bias and study design. Emotion-regulation difficulties are clinically relevant across diagnoses; they do not imply a uniform mechanism or a uniquely defining trait in every person who self-harms. Wolff et al., 2019.

Why immediate relief can maintain a costly behavior

In a reinforcement account, a person experiences intense discomfort, self-harms, and then reports reduced discomfort or believes a feared state has stopped. The temporal association may strengthen expectations that the action will work again. Under stress, access to flexible strategies may narrow, especially if the person has little social support or feels ashamed of disclosing distress. The result can be a recurring behavioral pathway even though the relief is transient, future distress persists, and injuries accumulate. This is a conceptual account consistent with many reports; identifying actual learning contingencies in an individual requires longitudinal data and collaborative assessment rather than a presumed universal pattern. Nock, 2009.

The phrase 'emotion regulation' describes an attempted change in an internal state, not an endorsement of the means used. Clinically responsible care seeks to understand what need the behavior has been meeting and to build safer, accessible alternatives with the patient. Effective alternatives may require changes in treatment, environment, family communication, sleep, trauma care, social connection, and material security as well as individual coping skills. A narrow list of distractions cannot address every function or every level of distress. NICE NG225 recommendations.

Does distress always fall after self-harm?

Retrospective self-report and laboratory studies historically gave support to the proposition that negative affect can precede and decline after NSSI. Yet immediate experience has been harder to establish reliably in natural settings. Ecological momentary assessment studies ask participants about mood, urges, and behavior repeatedly during everyday life, minimizing some recall delay. A systematic review found negative affect often rises before NSSI thoughts and behaviors, while observed changes after episodes are less consistent across studies and analytic choices. Sampling intervals, the rarity of observed events, reporting omissions, and variation in emotional measures complicate interpretation. Brown and colleagues' systematic review.

This nuance changes the language clinicians should use. Temporary relief is a frequently reported experience and a useful hypothesis to examine, but it should not be asserted as inevitable. An episode may be followed by shame, increased distress, regret, exhaustion, social response, or a mixture of feelings. A person may expect relief and instead experience no reduction in affect; another may experience a delayed shift. A careful case formulation records actual experiences before and after episodes and revises the model when observations differ from expectations.

Emotion numbness, dissociation, and altered states

Some people describe experiences of emptiness, emotional blunting, derealization, or dissociation rather than high arousal. In functional models, a person may seek the return of subjective sensation, a feeling of reality, or a sense of connection with their body. Those reports help distinguish relief from an aversive state from restoration of a desired experience. The two categories may appear together as distress, numbness, and arousal fluctuate. Clinicians should examine dissociative symptoms, trauma-related processes, substances, medical conditions, and possible safety risks without assuming the explanation in advance. Nock and Prinstein's four-function formulation provides a conceptual starting point.

Dissociation varies greatly in definition and severity. The presence of self-harm does not establish a dissociative disorder, and the subjective aim of ending numbness should not be confused with a desire for death. Nevertheless, people may find their intention difficult to describe while highly distressed or dissociated. Assessment should therefore allow time, avoid coercive assumptions, and return to uncertain questions after the person is stabilized.

Interpersonal functions and the social environment

Distress that cannot yet be communicated in words

Self-harm sometimes occurs in the context of attachment insecurity, loneliness, conflict, perceived rejection, invalidation, coercion, bullying, or an inability to ask directly for support. Some people report that the behavior expresses a distress that words have failed to convey. Others describe no interpersonal motive at all. Social consequences may affect recurrence even when the person did not consciously intend a particular response. An interpersonal function is best examined through the person's experience and the actual sequence of interactions, not through stereotypes about motivation. Nock and Prinstein, 2004; NICE NG225.

Seeking recognition, reassurance, or care reflects human needs for attachment and support. The term 'attention-seeking' has often been used dismissively and may shut down honest disclosure. A more precise clinical question asks whether the person was trying to communicate suffering, obtain help, escape a threat, stop an interaction, or restore connection. This framing creates opportunities for safer communication and changes how others respond, while preserving responsibility for immediate medical care and safety.

Escape from overwhelming social demands

Situations involving pressure, humiliation, conflict, rigid expectations, or lack of perceived escape can function as antecedents. An episode may be followed by a temporary cessation of a demand, a change in conflict, or removal from an environment. This does not mean that the person intentionally engineered an outcome or that the social situation alone caused the behavior. It suggests that individual treatment and environmental intervention may both be relevant. For adolescents, school, home, peer relationships, and online interactions can be central contexts. NICE NG225 adolescent assessment guidance.

Family and clinician responses matter. Punishment, humiliation, moralizing, or withholding necessary care can intensify shame and discourage disclosure. Compassionate listening, clear expectations, predictable access to support, and coordinated professional help create conditions in which needs may be expressed more safely. When abuse or coercion is present, safeguarding and protection take priority over a simplistic discussion of communication skills.

Reinforcement is not blame

Behavioral descriptions of social reinforcement are probabilistic scientific explanations of consequences, not judgments of character. A family member who provides compassionate attention after a disclosure is doing something important; the solution is to establish reliable access to support before a crisis, alongside clinically appropriate care. Withholding comfort to avoid 'reinforcing' behavior misunderstands both the complexity of causation and the ethical duty to respond to distress. Supporters should respond to danger immediately while collaborating with professionals to develop safer ways to recognize and address needs.

Self-criticism, punishment, shame, and beliefs about the self

Some individuals describe self-harm in language of punishment, undeservingness, guilt, self-disgust, or intense anger turned inward. These are reported psychological functions requiring individualized exploration. Self-criticism may follow interpersonal rejection, trauma, perfectionistic demands, perceived failure, or a broader negative self-schema; such associations differ across people. Relief and self-punishment can appear together, and the person may experience punishment as an expected consequence rather than a separate conscious goal. Klonsky's functional review.

Shame can shape both the behavior and whether it becomes visible to others. A person may hide injuries, avoid medical care, fear family reactions, or feel that disclosure confirms a negative identity. Stigma can therefore obstruct measurement and treatment. Clinicians should invite disclosure without requiring a public confession or equating scars with personality. Therapeutic work may address self-directed hostility, rigid beliefs, emotion regulation, and a safer relationship with the body in a manner suited to the person's needs and developmental context.

Self-punishment is a reported motive, not evidence that suffering is deserved. Assessing it requires attention to suicidal intent, severe depression, trauma symptoms, obsessive guilt, and other clinical experiences that can produce superficially similar descriptions. The relevant formulation clarifies what thoughts are present, how strongly they are believed, what they lead to, and what may interrupt them safely.

Learning, habit, and repetition across time

Antecedents, consequences, and learned expectations

Functional analyses examine an episode's antecedents, behavior, and consequences. Antecedents may include a mood change, conflict, sensory overload, isolation, memory, substance use, sleep loss, or sudden perception of threat. Consequences can involve changes in arousal, attention, interactions, avoidance, regret, or support received. After repeated episodes, the expectation that an act will produce a desired change can become established even when the observed benefits are inconsistent. Such an account is clinically testable through careful history and prospective monitoring; it must not be treated as proof of an individual's mechanism merely from frequency. Nock, 2009.

A learned association may become easier to activate in familiar contexts. Under stress, habitual responding sometimes displaces reflective choices, but direct evidence that a specific individual's NSSI is a habit in a mechanistic sense requires more than a history of repetition. Researchers distinguish habit claims, reinforcement hypotheses, coping expectations, and intentional action because their predictions differ. Therapies can target early cues, expand response options, and support the person in acting on a collaboratively developed safety plan without assuming a fixed character trait.

Benefits-and-barriers accounts

Integrative models propose that self-harm becomes more likely when anticipated immediate benefits are salient and barriers to acting are reduced by distress, exposure, learned expectations, or repeated experience. The benefits-and-barriers model is a useful theoretical organization of such hypotheses, but it cannot establish a single verified causal pathway across people. Observational reports, laboratory proxies, and naturalistic designs provide different kinds of evidence and each leaves uncertainties. Hooley and Franklin's theoretical formulation.

An integrative explanation also has to account for why many people experiencing severe distress never self-harm. Broad emotion dysregulation or exposure alone is insufficient. Personal history, social support, alternatives, access to treatment, developmental factors, beliefs, context, and chance all shape behavior. A scientifically defensible model needs to explain both occurrence and nonoccurrence, as well as why change becomes possible during recovery.

Why frequency does not reveal function

Two people with similar episode frequency may have different motives, degrees of suicidal ideation, medical complications, and environmental stressors. The same person may have one episode linked to numbness and another linked to conflict or self-criticism. Frequency, timing, and severity are important clinical facts but do not assign a motive. Care proceeds by asking rather than guessing. NICE's recommendation to consider each episode individually is particularly important where professionals encounter repeated crises.

What epidemiological studies can and cannot tell us

Prevalence depends on definition and method

NSSI research has often found particularly high lifetime self-report prevalence during adolescence and emerging adulthood. An influential 2014 meta-analysis of community studies reported pooled lifetime estimates of roughly 17% among adolescents, 13% among young adults, and 6% among older adults, alongside major variation in measurement and study design. These figures describe historical samples and definitions, rather than a definitive current probability for a particular person or country. Studies disagree about what behaviors count, whether a single episode qualifies, the relevant recall interval, and how willing participants are to disclose sensitive events. Swannell and colleagues' systematic review.

Prevalence estimates should not be confused with rates of severe injury, emergency presentation, clinically assessed NSSI, suicide attempts, or suicide mortality. Hospital statistics disproportionately reflect episodes that reach medical services, while community questionnaires may capture undisclosed behavior. Survey privacy, cultural meaning, translation of terms, sampling frame, age composition, and changing disclosure norms all affect observed rates. A cluster article about functions therefore uses prevalence for context rather than treating a number as an explanation of why the behavior occurred.

Clinical samples and community samples differ

Many classic functional studies recruited psychiatric inpatients, outpatient patients, or selected adolescents. They provide crucial information about clinically distressed populations, yet their distributions of motives may not generalize to community adults, older individuals, culturally diverse settings, or people who have never disclosed NSSI to a professional. Convenience sampling and recruitment through online self-injury communities raise further issues of selection. Self-report measures commonly ask whether a behavior 'helps' with emotions; those answers capture beliefs and perceived experiences as well as possible consequences. Nock and Prinstein, 2004; Klonsky, 2007.

Differences in function estimates across studies may reflect true differences in the lives of participants or differences in the questionnaires presented to them. Research should report the exact sample, instruments, reference period, and how ambiguous cases were classified. Comparisons without these details may overstate population contrasts.

Associations cannot predict an individual episode

Emotional dysregulation, previous self-harm, trauma exposure, social isolation, mental health difficulties, and other variables may correlate with self-injurious behavior across groups. Such associations cannot yield a deterministic story about an individual. Risk factors explain some variation across populations, while warning signs reflect changes requiring clinical attention; neither automatically specifies a person's immediate likelihood of a future event. The broader suicide literature demonstrates limitations in predicting future behavior from isolated risk factors, reinforcing the importance of contextual assessment over numeric certainty. Franklin and colleagues' longitudinal meta-analysis and NICE guidance on risk-scale limitations.

Psychological, biological, and social mechanisms

Affective reactivity and inhibitory control

A person's emotional state emerges from interactions among appraisal, attention, physiological arousal, memory, social context, and opportunities to act. During a crisis, high arousal or emotional narrowing may impair access to previously learned skills. Some functional theories propose that intense affect and limited regulatory resources make a familiar response more accessible. These processes are plausible and consistent with reports of distress preceding episodes; establishing their causal role requires prospective designs, experimental tasks that safely approximate component processes, and careful attention to confounding. Wolff et al., 2019; Brown et al..

The same observation may have multiple explanations. High negative affect preceding self-harm may signal a specific regulation failure, the severity of an ongoing psychiatric condition, an interpersonal event, sleep deprivation, or a mixture. Neurocognitive and physiological measurements may reveal group-level differences without identifying a unique biological signature of NSSI. Clinical care therefore integrates psychological and medical information instead of using speculative biomarkers as diagnostic shortcuts.

Pain processing and neurobiological hypotheses

Research has investigated pain perception, stress-related physiology, endogenous opioids, reward processing, and other biological pathways as potential contributors to self-injury. These hypotheses are neither interchangeable with motives nor established single causes. Laboratory differences in pain sensitivity or physiological response may reflect diverse factors, including history, current state, measurement conditions, medication, dissociation, and other confounders. Findings in a selected sample do not justify predicting who will self-harm or using biological explanations to override a person's account. The broader biopsychosocial research program is summarized in Nock's review and Hooley and Franklin's integrative model.

Clinically, the important biological questions often concern immediate injuries, medical consequences, sleep, pain disorders, substance effects, and coexisting conditions that warrant care. No clinician should infer the safety of an episode from reports that injury felt unusually painless or from a hypothetical neurochemical process.

Trauma, adversity, and developmental context

Adverse experiences can affect threat perception, self-evaluation, trust, access to support, and emotion regulation. Some people link self-harm to traumatic reminders, dissociation, or learned beliefs about themselves. Others self-harm without a known trauma history. A trauma-informed formulation asks about exposure, ongoing safety, posttraumatic symptoms, dissociation, and what helps the person maintain agency. It preserves the difference between an association, a proposed mechanism, and evidence of causation in any particular life. NICE NG225 psychosocial assessment guidance.

The presence of adversity also raises practical questions about housing, coercion, violence, discrimination, poverty, healthcare access, and family security. Treatment may need to address ongoing conditions that perpetuate distress. Describing these conditions as 'background factors' while focusing exclusively on coping skills can understate their causal plausibility and clinical importance.

Social learning, exposure, and online contexts

Young people and adults can learn language, expectations, and behavioral scripts through peers, family narratives, social media, or online communities. Social exposure may alter how an individual interprets distress or seeks belonging, but observed links between online activity and self-harm are complex. People experiencing distress may also seek certain spaces because they already need support, producing reverse causation. Content moderation, social isolation, community support, online stigma, and exposure to harmful material must be analyzed separately. NICE's adolescent assessment guidance recommends discussing online experiences in context rather than assuming their meaning.

Responsible public communication avoids sensational descriptions and detailed depictions of injury. Online spaces can provide compassion, recovery narratives, or referral to services; they can also normalize harmful behavior or expose users to distressing information. Evaluation needs study designs capable of distinguishing these mechanisms and identifying which platform practices improve safety.

Why functions vary between adolescents and adults

Adolescence, development, and changing support systems

Adolescence involves rapidly changing relationships, autonomy, emotional development, school demands, and family roles. These shifts influence both opportunities for coping and meanings attached to behavior. Some adolescents describe affect regulation, others interpersonal communication, self-punishment, numbness, or overlapping motives. Care should evaluate the young person's developmental abilities, emerging independence, privacy needs, and sources of safety, rather than simply applying adult assumptions. NICE NG225 guidance for children and young people.

A parent or caregiver may first discover injuries, a school staff member may hear a disclosure, or the adolescent may seek care independently. The response should be calm, medically appropriate, and nonpunitive. Clinicians need to explain confidentiality and its limits, involve caregivers where helpful and safe, investigate abuse or exploitation when indicated, and coordinate with age-appropriate mental health care. Even when a young person says they have never intended to die, a direct, developmentally appropriate evaluation of suicidal thinking and immediate safety remains important.

Caregiver involvement and safeguarding

Family participation is useful when caregivers can provide emotional containment, reliable communication, transport to care, and collaborative safety support. Participation should be shaped by the adolescent's wishes, developmental competence, legal requirements, and safeguarding context. Where a caregiver is a source of violence, coercion, or serious harm, protection may require alternative support arrangements. Clinical decisions about information sharing must follow applicable law and professional standards. NICE clinical recommendations explicitly address family participation and safeguarding.

Young people should not be made solely responsible for creating safe circumstances when their environment is overwhelming. Effective help can involve family work, adjustments at school, access to a trusted adult, treatment of concurrent mental health conditions, and follow-up after emergency attendance. Research comparing youth treatments must also separate adolescents with NSSI from those with broader self-harm or suicidal behaviors, because enrolled populations and outcomes differ.

Adult life circumstances and hidden behavior

Adult self-harm may occur amid occupational stress, caregiving responsibilities, grief, intimate relationships, mental illness, trauma, social isolation, and medical difficulties. Adults may be especially concerned about disclosure affecting employment, parenting, relationships, or professional reputation. Some have a long history; others begin later in life. An adult-focused assessment attends to responsibilities, access to private care, community resources, functional impairment, and the possibility that scars or episodes have been concealed for years. The individual's motive cannot be inferred from their age or history alone.

Older adults experiencing self-harm warrant particularly careful medical, cognitive, psychiatric, and social assessment. Frailty, physical illness, bereavement, isolation, and suicidal intent may have different clinical significance from the distributions usually studied in adolescent NSSI research. NICE NG225 includes age-sensitive recommendations; researchers should avoid extrapolating youth functional prevalence to older adults.

Coexisting conditions and diagnostic differentiation

Depression, bipolar disorders, and mood disturbances

Depression can involve hopelessness, self-criticism, emotional pain, suicidal ideation, and diminished pleasure. Bipolar disorders may involve mood episodes, agitation, changes in sleep and energy, and context-dependent changes in judgment or safety. These are relevant clinical states, but self-harm does not by itself establish either diagnosis, and the presence of a mood diagnosis does not identify the function of a particular episode. Assessment must examine episode course, suicidal intent, sleep and energy change, medications, substance use, and safety. For broader risk-specific questions, the project's separately owned Bipolar Disorder and Suicide Risk study addresses that canonical intent.

A person who self-harms to reduce tension while depressed may have different treatment needs from someone experiencing intrusive thoughts, trauma-related dissociation, or intoxication. The immediate clinical formulation should describe the behavior and its function while treating concurrent diagnoses according to their own evidence base. NICE NG225 recommends assessing coexisting conditions in treatment planning.

Personality functioning and relational difficulties

Emotion dysregulation and self-injury have been studied in relation to borderline personality disorder and other patterns of personality functioning. Associations require careful interpretation because the behavior can occur without the disorder, and experiences of shame or invalidation may be magnified by diagnostic stigma. Clinicians should evaluate longstanding patterns of relationships, self-concept, impulsivity, dissociation, and symptoms using appropriate diagnostic criteria and consider trauma, developmental context, and alternative explanations. The distinct canonical article BPD and Self-Harm: Risk, Safety Planning, and Treatment addresses condition-specific risk and treatment.

The function-centered approach protects against reductive assumptions. A person may use self-injury to regulate acute distress and also experience relational difficulties, but neither fact authorizes assigning a personality disorder without full assessment. Treatment planning should respect reported experiences and improve access to evidence-based care.

Autism, neurodevelopmental differences, and sensory experiences

Autistic people and people with other neurodevelopmental differences may experience sensory overload, communication challenges, social exclusion, mental health difficulties, or distress in inaccessible environments. Clinicians must distinguish NSSI from behaviors with different developmental and functional meanings and avoid automatically translating an observed act into either suicidal intent or an NSSI-specific emotional motive. Assessment adaptations may include direct language, time to process questions, accessible communication, and exploration of sensory and environmental conditions. NICE NG225 explicitly calls for neurodevelopmental considerations.

The project's independently published Suicide Risk in Autistic Adults: Evidence, Assessment, and Prevention examines suicide-related risk in autistic adults. The present article's remit is the broader science of functions; distinctions between repetitive behavior, sensory coping, NSSI, and suicidal actions depend on behavior and intent as assessed in the individual.

Obsessive intrusions, psychosis, and altered interpretation

Intrusive thoughts about self-harm may occur in obsessive-compulsive disorder without a desire to act on them. Their unwanted, fear-provoking character and associated compulsions may call for a different treatment formulation from urges to self-injure. By contrast, psychotic symptoms, severe disorganization, or impaired reality testing may change immediate clinical needs, especially when accompanied by commands, fear, severe agitation, or medical risk. The published Suicidal OCD and Schizophrenia and Suicide Risk articles retain their separate canonical topics.

No single word, such as 'urge,' 'intrusion,' or 'voices,' can substitute for clinical interviewing. The practitioner asks what the experience means, whether it is wanted, feared, resisted, accompanied by intention, or altering behavior, and whether urgent medical or psychiatric attention is required.

Eating disorders, substances, and trauma-related symptoms

Self-harm can coexist with disordered eating, substance use, posttraumatic symptoms, and other conditions involving distress or difficulties with regulation. Similar functional language—such as control, relief, punishment, or numbness—may appear across behaviors, yet different actions have different medical risks, treatment pathways, and outcome measures. Alcohol or other intoxication can alter judgment and increase immediate danger. A clinician assesses both the behavior and the full clinical context, including signs requiring emergency care. NICE NG225 advises attention to coexisting conditions and not delaying appropriate care because of intoxication.

How researchers study functions

Retrospective questionnaires and interviews

Many investigations use structured questionnaires asking participants to endorse reasons for NSSI or interviews reconstructing episodes. These tools are practical for assessing groups and generating clinical hypotheses. Their limitations include recall bias, the influence of present mood, changes in self-understanding, socially desirable answers, and the temptation to interpret a reported reason as a verified consequence. Instruments may bundle different experiences in one item or assume categories not meaningful to the participant. Researchers should report the instrument's validation, age group, and definition of self-harm. Nock and Prinstein's original functional assessment and Klonsky's evidence review illustrate this literature.

The person who reports a motive provides indispensable evidence about subjective meaning. Respect for that evidence can coexist with scientific precision: an interview can establish what someone remembers and believes without directly proving a causal chain. Clinicians can revisit hypotheses collaboratively over time.

Ecological momentary assessment and temporal sequences

Ecological momentary assessment, experience sampling, and diary designs examine thoughts, feelings, and events closer to the time they occur. These approaches are particularly valuable when a behavior is episodic and affect changes quickly. They can distinguish between-person correlations from within-person variation and analyze sequences of emotional changes around reported urges or acts. Yet they encounter missing reports, participant burden, small numbers of observed episodes, uncertain timing, and possible behavior change induced by monitoring. Brown and colleagues' ESM review.

A time sequence is stronger evidence about temporal ordering than a retrospective summary, but temporal precedence does not alone prove causality. For example, distress may decline after an episode because the precipitating event passed, a person received help, or the assessment captured a later moment. The most defensible conclusions integrate multiple methods rather than assigning definitive reinforcement from a single graph.

Experimental proxies and their boundaries

Laboratory studies have investigated emotion, stress responses, and experimental proxies for component experiences, avoiding dangerous behavioral experiments. These designs can improve control over a narrow variable, but a proxy is not equivalent to actual self-harm in everyday life. Ethical constraints appropriately limit direct causal manipulations of dangerous behavior. Researchers must therefore translate proxy findings cautiously and evaluate whether results replicate across naturalistic samples. Klonsky's review.

Longitudinal cohorts and intervention studies

Longitudinal studies can examine whether baseline characteristics precede later self-harm, while intervention trials can test whether a treatment changes prespecified outcomes. Neither design is interchangeable with functional assessment: a factor that predicts later episodes may not be the reason for an individual event, and a therapy that reduces recurrence may act through several pathways. Attrition, treatment selection, confounding, baseline severity, and inconsistent outcome definitions affect conclusions. Evaluating mechanisms ideally requires prospective measurement of both hypothesized mediators and behavior, sufficient temporal separation, and replicated analyses.

What would constitute stronger mechanistic evidence?

Stronger evidence would combine precise definitions, repeated within-person measurement, credible temporal sequencing, well-measured contextual confounders, and tested changes in proposed processes during interventions. Replication across adolescents and adults, clinical and community samples, cultural settings, and different reporting methods would improve generalizability. Outcomes should include changes in function, urge intensity, self-harm frequency, injury-related healthcare, suicidal ideation, suicide attempts, quality of life, and engagement in care, with each reported separately. These are research priorities, not assumptions about established efficacy.

Clinical assessment: understanding the person and maintaining safety

Start with immediate medical and psychological needs

An individual presenting after self-harm requires an assessment of injuries and medical urgency along with an opportunity for compassionate psychosocial evaluation. The practitioner should ask directly and respectfully about current suicidal thoughts, intent, plans, recent behavior, intoxication, access to potentially dangerous means, ability to remain safe, and available support. Any acute medical complication or immediate threat of suicide warrants urgent emergency evaluation. The person should receive needed medical attention regardless of whether the episode was described as suicidal. NICE NG225 recommendations; VA/DoD, 2024.

Medical stabilization and psychosocial care should be coordinated. A person's fear of judgment can make disclosure difficult, so a calm explanation of confidentiality and the purpose of questions is clinically important. Clinicians must avoid assuming that a person who has repeated episodes needs less attention than someone presenting for the first time. Changes in behavior, severity, intention, or circumstances may require a different response.

Screening, psychosocial assessment, formulation, and diagnosis

Screening is a brief method used to identify people who need further evaluation. It is not a diagnosis and is not an individual prediction of suicide. Psychosocial assessment explores the person's current state, context, behavior, intent, developmental and mental health factors, strengths, support, and care preferences. Risk formulation integrates these findings into a contextual professional account of present safety and actions needed. Diagnosis, when clinically appropriate, evaluates specific conditions using accepted criteria. The functions of self-harm are one component within this larger assessment process. NICE NG225.

NICE advises against using self-harm or suicide risk scales to predict future episodes, to decide who should receive treatment or be discharged, or to assign global low, medium, or high risk bands as if they determined future events. Clinicians should instead assess needs and immediate and long-term safety, alongside a documented formulation. A positive screen should prompt appropriate clinical engagement, not an automatic diagnostic label. NICE recommendations 1.6.1–1.6.6.

Reconstructing an episode collaboratively

A functional interview can begin with the preceding hours or days: important events, sleep, physical discomfort, social encounters, trauma reminders, emotional intensity, unwanted thoughts, sensory experiences, substances, and access to help. The clinician asks what the person hoped might change, whether they considered other responses, what they experienced immediately and later, and whether their interpretation has changed. Questions remain open-ended so the assessment does not supply the answer. The discussion should also explore individual strengths and the situations in which urges passed without injury. NICE recommendations on exploring functions.

The history should cover frequency, patterns over time, medical consequences, earlier suicide attempts, current suicidal ideation, ambivalent intent, mental health conditions, family and community context, barriers to treatment, and the person's goals. Clinicians should record uncertainties honestly. A function inferred from one episode should be revisited rather than copied automatically into all future encounters.

How a useful clinical formulation is organized

A formulation describes predisposing vulnerabilities, current precipitating circumstances, mechanisms that may maintain distress or recurrent episodes, strengths and protective resources, and a practical safety and treatment plan. For example, a clinician may observe that intense self-criticism rises after a recurring interpersonal conflict, that urges intensify during isolation, and that access to a trusted supporter helps the person choose another response. Those observations suggest testable treatment targets, without claiming certainty about causal mechanisms. The formulation remains collaborative and changeable.

Immediate safety needs override theoretical completeness. If a person is at risk of imminent self-injury or suicide, cannot maintain safety, is severely intoxicated, or needs urgent medical treatment, emergency care and supervision take precedence. A detailed account of psychological functions can be developed once the person is safe enough to participate.

Protective factors and accessible resources

Protective resources may include trusted relationships, meaningful responsibilities, culturally appropriate community support, accessible mental health care, skills for emotional regulation, safe housing, and collaborative plans for crisis periods. Their significance depends on whether they are available to the individual in the present situation. A theoretical protective factor on a questionnaire does not guarantee immediate safety. Clinicians should ask who can help today, what care is feasible, what obstacles exist, and how follow-up will occur. WHO's LIVE LIFE implementation guide places individual care within wider public health systems.

From understanding functions to evidence-based care

Function-informed treatment has a broader purpose

An effective care plan addresses both safety and the distress that self-harm has been associated with managing. If the person identifies overwhelming affect before episodes, treatment may emphasize recognizing early signals, naming emotions, tolerating intense arousal, and finding sustainable ways to regulate it. If interpersonal conflict is central, therapy can address communication, boundary-setting, access to supportive relationships, and the surrounding environment. If dissociation or trauma reminders are important, treatment may require trauma-informed approaches and careful pacing. A functional formulation helps set therapeutic priorities, but it does not establish that a specific technique will prevent future episodes. NICE NG225.

Good treatment also recognizes that existing options may be unavailable during an acute crisis. An intervention that works when a person is calm may be difficult to use when frightened, intoxicated, dissociated, or isolated. Care planning should consider rehearsal, timing, accessibility, available people, and the circumstances in which the person has previously remained safe. Treatment targets can change as family circumstances, clinical symptoms, and self-harm functions change.

Dialectical behavior therapy and skills-based approaches

Dialectical behavior therapy (DBT) combines strategies for emotion regulation, distress tolerance, mindfulness, interpersonal effectiveness, and behavior change. It has been studied in selected adults and adolescents with self-harm and suicidality, often with substantial differences in inclusion criteria, treatment intensity, comparison care, and endpoints. Research on adolescent self-harm provides support for comprehensive DBT-A in reducing repeated self-harm in some trials and systematic reviews. These findings should be attributed to the treatment packages and populations actually evaluated, rather than to a single skills worksheet or a generic 'DBT technique.' The Cochrane review of interventions for young people who self-harm and 2025 youth treatment network meta-analysis inform this evidence assessment.

Adult findings are more nuanced. The Cochrane review of psychosocial interventions for adults who self-harm distinguishes repetition of self-harm from the frequency of acts, changes in suicidal ideation, and broader clinical outcomes. The extent and certainty of evidence differ by comparison and endpoint. Demonstrated improvement in emotional regulation, therapy engagement, or reported frequency cannot automatically be translated into a claim about the prevention of suicide deaths. The separately published Dialectical Behavior Therapy for Self-Harm and Suicidal Behavior article examines the clinical intervention in its own right.

Cognitive behavioral and problem-solving approaches

Cognitive behavioral approaches can help people examine rigid beliefs, recognize early cues, develop alternatives to a recurrent response, and address problems that maintain distress. The components, duration, and target population vary across studies. In adults presenting with self-harm, the Cochrane synthesis suggests that cognitive behavioral approaches may reduce subsequent repetition at some follow-up intervals, but certainty and relevance differ across trials. Such conclusions concern the observed study outcome, not a guarantee that functional mechanisms have been corrected or that an individual is no longer at risk. Witt and colleagues' adult Cochrane review.

Problem-solving, communication work, and treatment of coexisting conditions often complement targeted self-harm interventions. When a person faces unsafe housing, abuse, financial instability, or severe isolation, clinical care may also require social support and changes outside psychotherapy. Function-informed treatment gains meaning when it connects psychological skills to the conditions in which the behavior occurs.

Suicide-focused treatments and distinct outcome categories

When suicidal thinking or behavior accompanies NSSI, clinicians need suicide-specific assessment and care as well as attention to the functions of self-injury. Collaborative Assessment and Management of Suicidality (CAMS), suicide-focused cognitive behavioral interventions, safety planning, and structured follow-up contacts belong to a related but distinct clinical evidence base. Evidence concerning suicidal ideation, suicide attempts, self-harm recurrence, emergency utilization, and mortality must be analyzed as different endpoints. VA/DoD's 2024 guideline reviews suicide-focused clinical management, while NICE NG225 addresses self-harm across intent categories.

A collaboratively developed safety plan identifies personalized warning signs, internal coping steps, supportive people and settings, professional services, and steps to make the environment safer. It should be practical, accessible, and revised as needs change. For adolescents, safety planning is ordinarily one component of comprehensive and family-aware care. Its existence alone does not establish an independently effective treatment for every youth outcome. A functional formulation can help identify warning signs and alternatives, but it cannot replace crisis intervention when safety deteriorates.

Medication and treatment of coexisting disorders

There is no established medication whose indication is simply to eliminate the psychological functions of NSSI. NICE advises against offering drug treatment specifically to reduce self-harm. Medication may nevertheless be indicated for a properly assessed concurrent disorder, using that disorder's evidence base, risks, monitoring, and clinical context. Medication decisions belong in individualized professional care and should not be inferred from a function questionnaire. NICE NG225, intervention recommendations.

Why treatment studies often disagree

Trials may enroll people after any self-harm episode, people with documented NSSI, adolescents with repeated behavior, patients with borderline personality disorder, or adults seen in emergency services. Their endpoints may concern whether any new episode occurred, frequency of episodes, suicidal ideation, suicide attempts, depressive symptoms, distress, or emergency use. Treatments vary in dose, training, fidelity, follow-up length, and whether standard care already includes substantial support. Small samples and missing outcomes further widen uncertainty. Cochrane adult review; Cochrane youth review.

A treatment could reduce repetition through several mechanisms at once: increased ability to tolerate emotions, more responsive support, improved treatment of depression, reduced family conflict, or earlier access to emergency care. A favorable clinical result alone cannot identify which hypothesized function changed. Mechanism-specific evidence must measure those proposed processes directly and establish appropriate temporal relationships.

What to do when someone self-harms or reports urges

Respond to the immediate situation

If an injury may require medical attention, seek appropriate urgent healthcare. Ask about current safety and suicidal thoughts in direct, calm language; allow the person to describe their experience without arguing about motives. When the person reports imminent intent to die, an immediate plan, an inability to remain safe, severe impairment, or a dangerous medical condition, emergency services or an emergency department are appropriate. In the United States, call 911 for an immediate life-threatening emergency. For suicidal or emotional crises in the United States, a person can call or text 988 or use the 988 Suicide and Crisis Lifeline, which offers crisis support around the clock. Outside the United States, use the country's established emergency and crisis services rather than assuming that 988 operates internationally. 988 Lifeline; NICE NG225.

A person with nonimminent urges may benefit from contacting a trusted person, moving toward a safer supportive environment, using previously agreed coping steps, and arranging prompt contact with their clinician. These measures fit within an individualized plan and do not replace medical evaluation when injury or danger is present. Asking for help early can make the person's available options wider before distress escalates.

Speak without assigning a motive

Supportive language recognizes distress and communicates a willingness to help: 'I care about what is happening to you'; 'What was going on before this felt overwhelming?'; 'Are you safe right now?'; 'Would you like me to help you find medical or psychological support?' An individual may not yet know their reasons, and uncertainty is an acceptable answer. Avoid accusations, threats, shaming, demands for promises that cannot be kept, and graphic questioning about acts. Family members should obtain professional advice when they are unsure whether physical or psychiatric emergency evaluation is needed.

Responses should preserve privacy and dignity, while sharing information with appropriate professionals when safety or safeguarding duties require it. For children and adolescents, caregiver involvement needs to respect developmental needs, confidentiality boundaries, and any risk that the home setting itself is unsafe. These considerations are consistent with NICE recommendations for family participation and safeguarding.

Build support that exists between crises

Long-term support can include predictable contact with trusted people, access to therapy, family and relationship work when appropriate, school or workplace accommodations, treatment for coexisting conditions, and a plan for noticing early signs of increasing distress. A person may want help with loneliness, trauma, practical hardship, or stigma as much as with the immediate urge. Services should offer continuity after a crisis and coordinate responsibilities clearly. NICE NG225 recommends coordinated care and timely aftercare where safety concerns persist.

For people with repeated episodes, coordinated multidisciplinary review can identify whether the existing plan is inaccessible, insufficient, or mismatched to the person's actual needs. Changes in motive, suicidal thoughts, medical severity, or living conditions should lead to a revised assessment and care plan. Repetition is a reason for renewed attention, not reduced eligibility for care.

Recovery and the changing meaning of self-harm

Recovery may involve fewer episodes, improved capacity to respond to urges, greater emotional flexibility, strengthened social relationships, better control of a coexisting disorder, reduced shame, or a larger sense of agency. These developments need not occur at the same pace. The reduction of one behavior does not automatically resolve trauma, loneliness, depression, or other difficulties. Conversely, meaningful progress in relationships and functioning can appear before a person's self-harm behavior has completely stopped. Goals should be developed collaboratively and monitored with measures suited to the individual.

A history of self-harm does not fix a person's identity or determine their future. Some people experience a single episode, some recurrent periods, and others patterns that change markedly with treatment and environment. A person may reinterpret the reasons for an earlier episode after gaining language for emotion, receiving a diagnosis, or leaving an unsafe situation. These evolving narratives are part of clinical understanding, and they show why one-time questionnaires should not be treated as final explanations.

Follow-up supports the transition from crisis-focused help to sustainable care. Where self-harm follows periods of acute distress, clinicians should make treatment pathways understandable, facilitate access, communicate across services with appropriate consent, and identify who is responsible for contact if circumstances worsen. NICE NG225 emphasizes continuity, collaborative care plans, and prompt aftercare when ongoing safety concerns exist.

Ethical questions, stigma, and public-health implications

The language of function can be explanatory, but it can also be misused. Describing social consequences must never become an excuse to dismiss suffering, minimize injury, punish disclosure, or withhold treatment. Clinicians and researchers should attend to power differences, cultural expectations, confidentiality, and the person's own interpretation. A respectful formulation is transparent about what is established, what is hypothesized, and what the person wants to change. NICE NG225 rejects punitive approaches and recommends compassionate, collaborative assessment.

Scientific writing should avoid graphic detail, claims that a given form of self-harm is harmless, and narratives that imply an inevitable progression from NSSI to suicide. Their overlap and statistical associations are clinically important, but individual outcomes cannot be deduced from a single category or symptom. Responsible public communication uses precise definitions, preserves differences between self-harm and suicidal behavior, and directs readers to appropriate help. The related Self-Harm and Suicide Attempts: Differences, Overlap, and Risk study examines that distinction in depth.

Population-level prevention complements individual treatment. Access to appropriate healthcare, supportive school and workplace environments, protection from violence, reduced discrimination, community connectedness, responsible communication, and safer environments can alter the circumstances in which crises develop. The WHO LIVE LIFE implementation guide offers a multisector public-health framework for suicide prevention. Its population strategies address important adjacent risks but do not independently establish the psychological function of NSSI in any one person.

Digital mental health interventions and general-purpose chatbots introduce additional questions. Support quality, confidentiality, escalation pathways, and empirical evidence differ across regulated clinical technologies, structured digital treatment programs, crisis services, and informal AI conversations. A tool that encourages reflection about emotions is not thereby validated as a treatment for self-harm or as a substitute for clinical safety assessment. Research should evaluate measurable outcomes, adverse events, and appropriate human oversight for each intervention category.

Research limitations and open questions

The principal methodological limits

The strongest conclusion in the functional literature is that motives and consequences are heterogeneous, with substantial recurring evidence for affect regulation. Greater uncertainty concerns how often short-term relief truly occurs in daily life, which neurobiological mechanisms are necessary, how social contingencies operate, and whether one pathway explains change during treatment. Self-report, selected psychiatric samples, retrospective recall, temporal gaps in diary methods, and laboratory proxies limit causal inference. Klonsky, 2007; Brown and colleagues' systematic review.

Important future scientific tests

Research would benefit from direct comparison of multiple motives in the same individuals across time, examination of episodes that were successfully interrupted, and better representation of adults, older people, and populations outside high-income Western settings. Studies should distinguish urges from actions, NSSI from broad self-harm and suicide attempts, and immediate affective changes from longer-term outcomes. Prospective evidence about transitions, remission, recurrence, treatment mediators, and social context would improve both theory and practice.

Ethical research should not require participants to reproduce harmful acts or expose them to avoidable danger. High-quality observational design, appropriate consent and safeguarding, meaningful participant involvement, and transparent reporting can improve evidence while protecting participants. The aim is to understand how people move toward safer choices and more supportive environments, as well as how episodes arise.

Frequently asked questions

Is self-harm always an attempt to die?

No. Nonsuicidal self-injury is defined by the absence of an intention to die during the relevant act, whereas a suicide attempt involves at least some intent to die. Broader self-harm terminology may include both, and a person's intent can be mixed or difficult to establish. Every presentation warrants assessment of current safety and medical needs. NICE NG225.

Does self-harm always make someone feel better?

No. Temporary relief is commonly reported, but naturalistic research finds variable postepisode emotional patterns. Some people experience shame, greater distress, or no relief. The question should be explored empirically and individually rather than presumed. Brown et al..

Can someone have several reasons for the same episode?

Yes. Someone might experience strong emotion, self-criticism, numbness, and interpersonal strain together. Their own understanding may shift over time, and functions can change across episodes. Nock and Prinstein's functional model; NICE NG225.

Does an interpersonal function mean that someone is manipulating others?

No. An interpersonal function refers to the role of social contexts or responses in a behavioral sequence. Distress communication, need for connection, and relief from overwhelming demands deserve compassionate exploration. Blame undermines disclosure and can interfere with appropriate care. NICE NG225.

Is emotion dysregulation a diagnosis?

No. Emotion dysregulation describes a set of difficulties influencing emotional experience and response. It is studied across different conditions and contexts, and association with NSSI does not identify a specific disorder or a singular cause. Wolff et al., 2019.

Can screening scales tell whether a person will attempt suicide?

They cannot reliably predict an individual future act. Screening can identify the need for further evaluation, but psychosocial assessment and professional formulation are required to determine appropriate support and safety decisions. NICE explicitly advises against using risk scales or global categorical risk bands to predict suicide or determine treatment access. NICE NG225.

Can treatment help when self-harm has become recurrent?

Yes. Evidence-based psychological interventions and coordinated care can reduce certain self-harm outcomes in specified populations, although effects and certainty vary. Treatment should address current safety, functions, coexisting disorders, and barriers to support, with appropriate follow-up. Results for recurrence, urges, ideation, attempts, and mortality must be distinguished. Cochrane adult review; Cochrane youth review.

When is urgent help needed?

Seek emergency medical assistance for serious injury, a potentially life-threatening medical problem, imminent suicidal intent, an immediate plan or inability to stay safe. In the United States, emergency services are reached by calling 911; crisis counseling is available by calling or texting 988 through the 988 Lifeline. In other countries, use local emergency and crisis resources. Prompt professional evaluation is appropriate for self-harm even when a person reports no intention to die. NICE NG225.

Related Articles

Nonsuicidal Self-Injury: Definition, Patterns, and Clinical Care — clinical definitions, observed patterns, and care across settings.

Self-Harm and Suicide Attempts: Differences, Overlap, and Risk — behavioral definitions, intent, co-occurrence, and methodological limitations.

Self-Harm Urges: Coping, Support, and When to Seek Urgent Care — practical responses and thresholds for urgent care.

Dialectical Behavior Therapy for Self-Harm and Suicidal Behavior — evidence and implementation of a structured clinical treatment.

BPD and Self-Harm: Risk, Safety Planning, and Treatment — condition-specific evidence and care.

Suicide Risk in Autistic Adults: Evidence, Assessment, and Prevention — neurodevelopmental context and suicide-related assessment.

Suicidal OCD: Intrusive Self-Harm Thoughts, Compulsions, and the Difference from Suicidal Intent — differential evaluation of unwanted intrusive thoughts.

References

988 Suicide and Crisis Lifeline. (2026). Crisis counseling and support in the United States. Official service.

Brown, A. C., Dhingra, K., Brown, T. D., Danquah, A. N., and Taylor, P. J. (2022). A systematic review of the relationship between momentary emotional states and nonsuicidal self-injurious thoughts and behaviours. Psychology and Psychotherapy: Theory, Research and Practice, 95(3), 754–780. DOI.

Chen, X., Dong, Y., Ye, M., et al. (2025). Comparative efficacy and acceptability of psychotherapeutic, pharmacological, and combination treatments for non-suicidal self-injury in children and adolescents: A systematic review and network meta-analysis. BMC Psychiatry, 25, 328. DOI.

Franklin, J. C., et al. (2017). Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychological Bulletin. DOI.

Hooley, J. M., and Franklin, J. C. (2018). Why do people hurt themselves? A new conceptual model of nonsuicidal self-injury. Clinical Psychological Science. DOI.

Klonsky, E. D. (2007). The functions of deliberate self-injury: A review of the evidence. Clinical Psychology Review, 27(2), 226–239. DOI.

National Institute for Health and Care Excellence. (2022). Self-harm: Assessment, management and preventing recurrence (NG225). Official guideline.

National Institute of Mental Health. Suicide statistics and definitions. Official resource.

Nock, M. K. (2009). Why do people hurt themselves? New insights into the nature and functions of self-injury. Current Directions in Psychological Science, 18(2), 78–83. DOI.

Nock, M. K., and Prinstein, M. J. (2004). A functional approach to the assessment of self-mutilative behavior. Journal of Consulting and Clinical Psychology, 72(5), 885–890. DOI.

Swannell, S. V., Martin, G. E., Page, A., Hasking, P., and St John, N. J. (2014). Prevalence of nonsuicidal self-injury in nonclinical samples: Systematic review, meta-analysis and meta-regression. Suicide and Life-Threatening Behavior. DOI.

U.S. Department of Veterans Affairs and U.S. Department of Defense. (2024). VA/DoD Clinical Practice Guideline for Assessment and Management of Patients at Risk for Suicide. Official guideline.

Witt, K. G., et al. (2021). Interventions for children and adolescents who self-harm. Cochrane Database of Systematic Reviews. Cochrane review.

Witt, K. G., et al. (2021). Psychosocial interventions for self-harm in adults. Cochrane Database of Systematic Reviews. DOI.

Wolff, J. C., Thompson, E., Thomas, S. A., Nesi, J., Bettis, A. H., Ransford, B., Scopelliti, K., Frazier, E. A., and Liu, R. T. (2019). Emotion dysregulation and nonsuicidal self-injury: A systematic review and meta-analysis. European Psychiatry, 59, 25–36. DOI.

World Health Organization. (2021). LIVE LIFE: An implementation guide for suicide prevention in countries. Official guide.