Every September, around World Suicide Prevention Day, attention turns to suicide prevention. Self-harm usually appears somewhere in that conversation, and it is usually in the wrong place. It gets treated as a warning light on the dashboard, a sign that someone is on their way to ending their life, and the response that follows is built around stopping the behaviour as quickly as possible.
That response is well meant; however, it frequently makes the situation worse. Self-harm and suicide are related, but not in the way most people assume, and the assumption itself does damage. It drives people into secrecy, it makes them dread the appointment they need, and it means the distress underneath goes unaddressed while everyone concentrates on the wound.
What We Now Mean by Self-Harm
Start with the definition, because it changed in 2022 and the change matters. Self-harm used to be defined as hurting yourself without intending to die. That is still how the term is often used in the United States, where the label is non-suicidal self-injury. UK guidance no longer defines self-harm by intent. NICE guidelines now define it as intentional self-poisoning or injury, irrespective of the apparent purpose of the act.
The removal was deliberate rather than careless. Sorting people by stated intent turns out not to work. It shifts between episodes, and it shifts within a single episode. Sizeable numbers of people report that during an act of self-harm they simply did not care whether they lived or died. Asking someone to place themselves on one side of a line that they are not experiencing as a line produces an answer that looks tidy and predicts very little.
When this piece says self-harm, it means the behaviour, whatever was going through the person’s mind at the time.
Myth 1: People Who Self-Harm Are Trying to Kill Themselves
Fact: For most people, most of the time, self-harm is how they stay alive rather than how they try to die.
This is the myth that matters most, and there is now direct evidence rather than only self-report.
A 2022 study monitored people who self-injure in real time, capturing their state before and after episodes rather than asking them to remember afterwards. Suicidal thoughts rose in the period leading up to an act of self-harm and dropped away afterwards. The researchers’ own conclusion was that self-injury can work as an effective, if damaging, way of coping with a suicidal state.
That is the opposite of a rehearsal. It is a release valve.
Two qualifications belong alongside that claim, because the picture is not all in one direction. Across a lifetime, a history of self-harm remains among the strongest predictors of eventual suicide. And when researchers compared people who said they wanted to die with people who said they did not, subsequent suicide deaths did not differ between the groups.
Both are true at once, and they are answering different questions. In the moment, self-harm mostly is not an attempt to die and often makes dying feel unnecessary. Over years, the same mechanism that brings relief also wears down the barrier, because pain becomes familiar and the body learns not to react.
The practical consequence is that treating every act of self-harm as an attempted suicide is a category error, and treating it as harmless because nobody meant to die is a second error in the other direction. Self-harm should be taken seriously without being mistaken for a suicide attempt.
Myth 2: People Who Self-Harm Are Attention-Seeking
Fact: Self-harm sometimes communicates distress, and communicating distress is not a fault.
When people are asked why they self-harm, private reasons dominate. Around two-thirds to four-fifths give reasons that have nothing to do with anyone else: managing feelings, stopping a spiral, punishing themselves, feeling real again. But between a third and more than half also endorse reasons involving other people, most often making distress visible when words have failed.
That is not a shameful admission. Communicating pain is what humans do. Someone who has learned that saying “I am not coping” gets them nothing has found another way to say it.
The stigma does not sit in noticing that self-harm can communicate. It sits in the leap from communication to manipulation, and from manipulation to withholding care. That leap is still made routinely in healthcare settings, where people who self-harm report being dismissed, treated coldly or made to feel they have wasted someone’s time. The Lancet Commission on self-harm in 2024 documented this as a global pattern.
If someone is asking for help in the only way available to them, the answer is to make an easier form of communication available rather than to punish them for the one they used.
Myth 3: Don’t Approach Someone Who Self-Harms, Leave It to the Professionals
Fact: Listening without judgement is the first useful step anyone can take, and it needs no training.
You do not have to discuss the self-harm at all. You do not have to inspect wounds, count episodes or extract a promise. Ask about the distress underneath. Stay in the room while they answer. The self-harm is the visible part, but it is not the part that needs your attention first.
What people fear when they consider telling someone is disgust, panic and being taken over. Any of the three closes the conversation and makes the next one less likely.
Myth 4: You Will Know Someone Self-Harms Because of Cuts on Their Arms
Fact: Cutting is one form among many, and a great deal of self-harm leaves nothing to see.
Self-harm includes burning, scalding, hitting, banging the head or other parts of the body, hair-pulling, biting, swallowing objects, inserting objects and self-poisoning. People choose sites that clothing covers, and methods that leave no mark at all when concealment matters.
There is a further category that often goes unrecognised. Some autistic people injure themselves during sensory or emotional overload: head-banging, hitting, biting, skin-picking that has gone past the point of harm. Research published in 2025 found autistic adults describing this as a reset switch, where the pain shuts off channels that have become unbearable. It is unpremeditated, it does not arise from emotional pain in the conventional sense, and it does not fit the standard picture at all.
Myth 5: Self-Harm Is a Fashion and Young People Grow Out of It
Fact: Self-harm is far more common than it was, it is not confined to the young, and the numbers have quadrupled in a generation.
In England, lifetime self-harm among adults stood at 2.4 per cent in 2000, 6.4 per cent in 2014 and 10.3 per cent in 2023 to 2024. Among people aged 16 to 24, the current figure is 31.7 per cent of young women and 15.4 per cent of young men. Roughly five million adults in England have self-harmed at some point.
The gender pattern deserves a note of its own, because it runs in an unexpected direction. Women self-harm more than men, and the gap is widest among the young. Suicide runs the other way: in England in 2024, the male suicide rate was 17.1 per 100,000 against 5.6 for women, with the highest rate of all among men in their early fifties. The people most likely to self-harm and the people most likely to die by suicide are not the same population. That divergence is one of the strongest arguments against reading self-harm as a straightforward step towards suicide.
Myth 6: If It Were Serious, They Would Have Done Something Worse
Fact: The severity of an injury tells you very little about the danger a person faces.
This assumption is everywhere, including among clinicians. A superficial cut gets read as a gesture. An overdose gets read as the serious attempt.
The evidence does not support the ranking. Self-cutting, routinely treated as the lesser presentation, is associated with greater eventual suicide risk than overdose. And because intent shifts within episodes as well as between them, the state of mind that produced a minor injury Today is not fixed in place for the following month.
Judging a person by the size of the wound means judging them by the least informative signal available.
Myth 7: The Goal Is to Get Them to Stop
Fact: Removing a coping strategy without replacing it usually leaves the person less able to cope, not more.
If self-harm is doing a job, taking it away leaves the job undone.
People who self-harm describe this clearly. When others restrict their access to methods, or respond with criticism and alarm, the urge intensifies rather than fading. What actually changes is the secrecy: they hide it better, stop treating wounds properly and pull away from the services that might have helped.
None of this means self-harm is acceptable and should be left alone. Stopping is an outcome of recovery rather than the route into it. The work is on what the self-harm achieves, and on building a genuine coping tool that does the same job without the damage. Once that tool is in place, the behaviour tends to recede on its own.
Where treatment is concerned, the strongest evidence in the most recent Cochrane review sits with mentalisation-based therapy, which reduced repetition of self-harm with high-certainty evidence. Group-based emotion regulation therapy also performed well. Case management, phone contact and information alone showed no benefit at all.
Myth 8: A Risk Assessment Will Tell You Who Is in Danger
Fact: Risk prediction at the level of an individual person barely works, which is why NICE now instructs clinicians not to rely on it.
A meta-analysis of 172 longitudinal studies found that prior self-harm predicts later suicide at a level only marginally better than chance. Even the strongest predictor available raised the absolute probability to a number the authors described as still nearly zero.
NICE guidance now states directly that risk assessment tools and scales should not be used to predict future suicide or repetition of self-harm, and that people should not be sorted into low, medium and high risk to decide who gets treatment and who is discharged.
For anyone who has been through an assessment that felt like a form being completed rather than a conversation being had, this is the evidence catching up with the experience. The clinical task is understanding what the self-harm achieves, not scoring it.
Myth 9: Self-Harm Means Borderline Personality Disorder
Fact: Self-harm cuts across diagnoses, and it appears at high rates in populations that have nothing to do with personality disorder.
Recurrent self-harm is one of the criteria used in diagnosing what is variously called borderline or emotionally unstable personality disorder, and the association is real. But the inference runs one way only, and people make it in both directions.
Autistic people have more than three times the odds of self-harm compared with non-autistic people. People with ADHD have roughly twice the odds. Self-harm is also common alongside depression, eating disorders, trauma histories and, frequently, no diagnosis at all.
The cost of the assumption is misdiagnosis, and it falls hardest on autistic women and girls, who have spent years being handed a personality disorder label for distress that had a different origin all along.
Taking It Seriously Without Getting It Wrong
The message underneath all nine myths is one idea stated from different angles. Self-harm is always a signal that something is seriously wrong, and it is almost never an attempt to die.
Holding both halves is harder than holding either one, which is why people keep dropping one. Drop the first and self-harm becomes a phase, a habit, a behaviour to be managed rather than understood. Drop the second and every episode becomes an emergency, every conversation becomes a risk assessment, and the person in front of you learns to say less next time.
The international theme for World Suicide Prevention Day has been changing the narrative on suicide. The narrative most in need of changing where self-harm is concerned is the one that reads every wound as a rehearsal. That narrative frightens families, exhausts services and obscures the question that would actually help: “What is the self-harm doing for the person?”.
Self-harm is a solution to a problem the person has not been able to name. Naming it is where the help begins. The assumptions in this list do a great deal of the damage, and if you would like help working out which of them are shaping your own situation, you are welcome to get in touch with Storm’s Edge Therapy.
Is self-harm different from other self-destructive behaviour, such as heavy drinking?
The distinction most people draw is between direct injury and slower routes to the same relief, and it matters less than it appears. Drinking to blackout, starving, and driving badly on purpose often do the same regulatory job over a longer timescale. Mapping them together usually tells you more about what a person is managing than arguing about which one counts as self-harm.
How do I know whether these assumptions have got into how I respond to someone?
The clearest marker is not what you say but what the other person tells you. If they have started disclosing less than they used to, or you are hearing about episodes only once they are over, your response has probably registered as alarm rather than interest. Noticing that shift is more informative than reviewing your own intentions.
Does someone have to want to stop before therapy can help?
No, and setting willingness to stop as a precondition tends to delay the work rather than focus it. Ambivalence is the ordinary starting position, and the early work is usually about understanding what the self-harm achieves rather than negotiating its removal. Wanting to stop tends to arrive later, once there is a replacement worth wanting.
