Autism and Suicide: Why the Risk Is So High, and Why It Is Preventable

Content noteThis article discusses suicide and suicidal thinking, and it includes my own experience of losing a parent to suicide. Please take care of yourself as you read. You do not have to read it all in one go. If you are struggling, the support lines at the end of this article are there for you at any time, wherever you are in the world.

Autistic people are far more likely to think about suicide, to attempt it, and to die by it than non-autistic people are. That finding has repeated across every large study of the past decade, in country after country, and it holds even after researchers account for other mental health conditions. It is one of the clearest and most urgent facts in autism research, and most people, including many clinicians, have never been told it.

A second finding matters just as much. A large part of this risk is driven by things we can change. It grows out of what autistic people so often meet in the world around them, including trauma, exhaustion, being misunderstood, and support systems that were never built for autistic minds. Risk that is built by circumstances can be lowered by changing those circumstances. That is the ground for hope in this piece, and it is the reason I am writing it.

This year, World Suicide Prevention Day on 10 September sits close to something I carry with me all the time. My mum died by suicide on the 6th September. I am an autistic and ADHD researcher, and I write about neurodiversity for a living, and even so, her death is the thing that made this subject impossible to look away from. I am not going to describe how she died, because that detail helps no one and can harm people who are already struggling. What I want to do instead is tell you what the research says, carefully and accurately, and what I wish the people around her had known.

A note before I begin. I hold a Master of Laws and I work as a senior safety and risk professional, and I have published peer-reviewed research on neurodiversity. I am not a clinician, and nothing here is medical advice. Every figure in this article comes from a named, published study, and where a claim rests on a source that has not yet been peer-reviewed, I say so plainly.

How much higher is the risk of suicide for autistic people?

The short answer is that it is high enough to change how mental health care should be done. The numbers are consistent, and they are large.

In one of the earliest large studies, Cassidy et al. (2014) surveyed adults who had been diagnosed as autistic at a specialist clinic in the United Kingdom. Two thirds of them, 66 percent, said they had thought about ending their life, and 35 percent had made a plan or an attempt. The rate of suicidal thinking was more than nine times higher than in the general population. These were adults diagnosed later in life, which describes a great many autistic people, including me.

The most reliable overall picture comes from Newell et al. (2023), a meta-analysis that pooled 36 studies and almost 48,200 autistic and possibly autistic people. Across all of them, around a third had experienced suicidal thoughts (34.2 percent), roughly one in five had made a plan (21.9 percent), and close to one in four had attempted suicide or engaged in suicidal behaviour (24.3 percent). One in four means a quarter of a whole community.

For deaths rather than thoughts, the largest recent analysis is Santomauro et al. (2024), which combined studies with global health data. Autistic people were almost three times as likely to die by suicide as non-autistic people, and autistic people without a co-occurring intellectual disability were more than five times as likely. The authors estimated that close to two percent of all suicide deaths worldwide in 2021 would not have happened if autistic people did not carry this raised risk. That is a number of lives, every single year, that the world could keep.

A 2026 study by Chikaura et al. tested whether this was really about autism or simply about the hard experiences autistic people tend to have. After accounting for trauma and for other co-occurring conditions, being autistic still predicted self-harm, suicide attempts and suicide plans. Autism did not explain everything, and it did not vanish once other factors were controlled for.

One caution, because accuracy matters to me and to my work. You will sometimes see it claimed that autistic people die 16 years earlier than everyone else. That specific figure is a misreading of the research it is usually attached to, as Hand et al. (2025) documented when they traced how the original study has been cited over the years. Life expectancy for autistic people does appear to be reduced, with the better estimates putting it at around six years for autistic people without an intellectual disability, and with wide uncertainty around that (O’Nions et al., 2023). The real figures are serious enough without inflating them.

For organisationsIf a quarter of a community has attempted suicide, then suicide risk is not a rare edge case to screen out. It is a baseline consideration for any workplace or service that includes autistic people, which is every workplace. Building awareness of this into manager training and wellbeing provision is a reasonable and proportionate step, and a powerful one for the people it reaches.

Why are autistic people more likely to experience suicidality?

The research points, again and again, towards what happens to autistic people, more than towards anything broken inside them. Several threads run through the evidence, and they braid together.

The first thread is trauma. Autistic people experience far more of it than non-autistic people do. A meta-analysis found that around 44 percent of autistic people had experienced trauma, a figure that climbed much higher among those who had been bullied or abused (Trundle et al., 2023, as reported in Chikaura et al., 2026). Chikaura et al. (2026) then found that autistic adults carrying the heaviest lifetime load of trauma were dramatically more likely to self-harm, to plan or attempt suicide, and to live with a mental health condition that affected their daily life. Childhood victimisation and difficult experiences at work came up most often alongside those outcomes.

The second thread is about how distress works on the inside. Tsypes et al. (2026) studied autistic adults whose autism had been confirmed by a clinician, and found that the most consistent marker of suicide risk across the whole group was anhedonia, which is a dulled ability to feel pleasure or reward. Emotional lability, meaning rapid and hard-to-settle shifts in feeling, was strongly linked to self-harm and to how severe a person’s suicidal thinking became. These are things that can be noticed and supported, if anyone knows to look for them.

The third thread is the support that never arrives. In a study whose title came straight from a participant, Camm-Crosbie et al. (2019) reported autistic adults saying, in effect, that people like me do not get support. They described reaching out and finding services that did not fit them, then giving up. When help is exhausting to access and does not work once you get there, distress has nowhere to go.

Underneath all of this sits a pattern that suicide researchers describe with two ideas: feeling like a burden to others, and feeling that you do not belong. A 2026 systematic review by Tubío-Fungueiriño et al. found that these two feelings ran through the risk factors for autistic people especially strongly. Social isolation, bullying, loneliness and rejection are common in autistic lives, and they are exactly the experiences that theory predicts will raise suicide risk.

Add to this the daily work of camouflaging, which is the effort of masking autistic traits to get through school, work and relationships, and the autistic burnout that can follow years of it, and a picture forms. It is a picture of people spending enormous energy to appear fine, often while being anything but.

For organisationsThe drivers above are workplace-shaped. Bullying, exclusion, the exhaustion of masking through back-to-back meetings and open-plan noise, and support processes that are hard to navigate all appear in this research. A neuroinclusive workplace, with predictable communication, sensory-considered spaces and support that is genuinely easy to reach, is a suicide-prevention measure, even though it is rarely described as one.

What about AuDHD? Does having ADHD alongside autism change the risk?

It appears to raise the risk, and part of the reason may be that the ADHD side goes unrecognised for years. This is my own profile, so I hold this section close.

Autism and ADHD co-occur often. Around 38 to 40 percent of autistic people also meet the criteria for ADHD (Rong et al., 2021). When people talk about autistic suicide risk, they are very often talking about AuDHD people without naming them.

The systematic review by Tubío-Fungueiriño et al. (2026) found ADHD named as a risk factor for suicide or suicidal behaviour across several of the studies it examined. At the same time, and this matters, it found signs that early identification of ADHD, and even hyperactivity itself, could be protective in some groups. A very large recent analysis of health records, which is still a preprint and has not yet been peer-reviewed, points towards ADHD-linked impulsivity and emotional dysregulation as one contributor to risk during adolescence and young adulthood (Baker et al., 2026, preprint).

The thread I take from this is that the greatest danger often comes from having one of the two conditions missed entirely. When someone is recognised as autistic but their ADHD is never picked up, or the reverse happens, they receive support that fits only half of them. For late-diagnosed AuDHD adults, and I am one, that mismatch can run for decades before anyone notices.

For organisationsDo not assume a single diagnosis tells you what someone needs. An autistic employee may also have ADHD, and the accommodations that suit one profile can clash with the other. Timers and scaffolding help the ADHD side; predictability and advance notice help the autistic side. Asking the person, rather than guessing from a label, is the practical move and the respectful one.

Are autistic women and girls at particular risk?

Yes, in specific ways, and being diagnosed late is a large part of why. The pattern here is real, and it is easy to miss.

Santomauro et al. (2024) found that the raised risk of dying by suicide was proportionally larger for autistic women, compared with non-autistic women, than the equivalent gap was for men. Kirby et al. (2024) went as far as issuing a formal call for clarity on the question, arguing that autistic women’s suicide risk has been under-recognised and that the field needs to take it seriously for the sake of the whole community.

Autistic women and girls tend to camouflage more, and camouflaging is linked to suicidal thoughts and behaviours. They also tend to be diagnosed later, sometimes not until adulthood, which means years of struggling without a name for it, of being told they are anxious, or difficult, or too sensitive. My mum’s generation of women were rarely recognised as autistic at all.

For organisationsThe autistic woman on your team may be the one who seems to be coping best, because she has spent her whole life learning to appear that way. Coping on the surface is a very different thing from being well. Making support easy to reach without a diagnosis or a disclosure, and normalising its use, reaches people who would never otherwise put their hand up.

Why do mental health services so often fail autistic people?

Because they were built for non-autistic minds, and the small mismatches add up into large barriers. This is where prevention most often breaks down.

Autistic people describe crisis lines that rely on reading tone, waiting rooms that overwhelm the senses, and assessments that ask for exactly the kind of on-the-spot emotional articulation that many autistic people find hardest (Camm-Crosbie et al., 2019). There is also diagnostic overshadowing, where a person’s distress is put down to their autism and then not treated as the emergency it actually is.

The support that is designed specifically to prevent suicide in autistic people barely exists yet. A 2026 review protocol from Russell et al. set out to map suicide-prevention supports for the autistic community, and it noted at the outset both how much higher the risk is and how little tailored provision there is to meet it. Clinicians themselves report feeling less confident assessing suicide risk in autistic clients than in others. When the tools were never made for you, and the people meant to help feel unsure, falling through the gaps says nothing about you and everything about the design of the system.

For organisationsEmployee assistance programmes and wellbeing services are not automatically accessible to autistic staff. Phone-first crisis support, long verbal intake processes and neurotypical assumptions about how distress looks can all shut autistic people out. Offering text and written options, and briefing providers on how distress can present in autistic people, closes some of that gap at very little cost.

What actually helps, and what does prevention look like?

Connection, being understood, early recognition, and support shaped around autistic people rather than expecting them to reshape themselves. The evidence on what protects is thinner than the evidence on what harms, and it still points somewhere clear and hopeful.

The 2026 systematic review by Tubío-Fungueiriño et al. found that the most consistent protective factors were social ones. Supportive relationships, a sense of belonging, and a positive, trusting relationship with a practitioner all appeared to lower risk. Early identification of co-occurring conditions helped too. Being known, and being known early, matters a great deal.

When researchers asked the autistic community directly what would help, in the largest survey of its kind, the answer reached well beyond crisis response. Moseley et al. (2026) gathered the views of thousands of autistic people and those who love them, and a central theme was making lives worth living, alongside faster diagnosis and support that arrives before the point of crisis. Prevention, in other words, is upstream. It lives in the shape of a whole life, well before any moment of emergency.

Practical tools designed for autistic people are beginning to appear and to show promise, including safety plans adapted specifically for autistic adults (Rodgers et al., 2024). That is the direction of travel, and it is worth being hopeful about.

On a human level, the things that help are often smaller than a service. They are the friend who texts instead of phoning. The manager who says take the time you need and means it. The person who asks a second time. If you are autistic or AuDHD and you are struggling right now, please know that I have needed help myself, and reaching for it has been worth it, even when it was hard, and even when it had not worked before.

For organisationsPrevention is mostly ordinary. Belonging, being understood, and support that is easy to reach are protective, and all three are within an employer’s gift. The B.R.A.I.N. framework I use in my work, standing for Belonging, Respect, Awareness, Inclusion and Neurodiversity, is one way to build these on purpose rather than leaving them to chance.

What I want you to take from my mum’s story

I cannot tell you exactly why my mum died. Suicide is almost never about one thing, and reaching for a single, tidy reason does a disservice to how complicated a human life is.

I have learned that the risk is real, that it is not shared evenly, and that autistic and AuDHD people carry more of it than almost anyone talks about. I have also learned that risk built by trauma, by isolation, by being misunderstood and unsupported, can be lowered by the opposite of those things. The research itself supports that hope.

If you take one thing from this article, let it be that suicide is preventable, that help exists and is worth reaching for, and that the way we treat autistic people, at home, at school, at work and in health care, is itself a form of prevention. My mum is the reason I do this work. If sharing her, even a little, means that one family has a different September to the one mine had, then it is worth every word.

Be gentle with yourself today, and with the neurodivergent people in your life. You rarely know what someone is carrying underneath appearing fine.

If you need supportIn New Zealand, free call or text 1737 any time to talk with a trained counsellor. You can also call the Suicide Crisis Helpline on 0508 828 865 (0508 TAUTOKO), or Lifeline Aotearoa on 0800 543 354, or free text 4357. In an emergency, call 111.Outside New Zealand, you can find a crisis line in your country at findahelpline.com.If you have been bereaved by suicide, you deserve support too. Please reach out to one of the services above, or talk to your GP.

References

Camm-Crosbie, L., Bradley, L., Shaw, R., Baron-Cohen, S., & Cassidy, S. (2019). “People like me don’t get support”: Autistic adults’ experiences of support and treatment for mental health difficulties, self-injury and suicidality. Autism, 23(6), 1431–1441. https://doi.org/10.1177/1362361318816053

Cassidy, S., Bradley, P., Robinson, J., Allison, C., McHugh, M., & Baron-Cohen, S. (2014). Suicidal ideation and suicide plans or attempts in adults with Asperger’s syndrome attending a specialist diagnostic clinic: A clinical cohort study. The Lancet Psychiatry, 1(2), 142–147. https://doi.org/10.1016/S2215-0366(14)70248-2

Chikaura, T. A., Weir, E., Griffiths, S., Procyshyn, T., Pelton, M., Allison, C., Hodges, H., White, S. R., Ford, T., & Baron-Cohen, S. (2026). Traumatic experiences, psychological distress and suicide-related behaviors in autistic adults. Autism Research, 19(1), e70137. https://doi.org/10.1002/aur.70137

Hand, B. N., Nikahd, M., Wolf, B. J., Hyer, J. M., Longo, A., Gilmore, D., & Bishop, L. (2025). Citation context analysis of autism mortality and suicide findings from Hirvikoski’s landmark study. JAMA Network Open, 8(2), e2461953. https://doi.org/10.1001/jamanetworkopen.2024.61953

Kirby, A. V., Conner, C. M., & Mazefsky, C. A. (2024). Are autistic females at greater risk of suicide? A call for clarity to advance suicide prevention for the whole community. Autism Research, 17(5), 898–905. https://doi.org/10.1002/aur.3115

Moseley, R. L., Marsden, S. J., Pelton, M., Weir, E., Procyshyn, T., Allison, C., Parsons, T. A., Cassidy, S., Chikaura, T., Hodges, H., Mosse, D., Rodgers, J., Hall, I., Owens, L., Cheyette, J., Crichton, D., Hedley, D., & Baron-Cohen, S. (2026). “The best way we can stop suicides is by making lives worth living”: A mixed-methods survey in the UK of perspectives on suicide prevention from the autism community. eClinicalMedicine, 93, 103793. https://doi.org/10.1016/j.eclinm.2026.103793

Newell, V., Phillips, L., Jones, C., Townsend, E., Richards, C., & Cassidy, S. (2023). A systematic review and meta-analysis of suicidality in autistic and possibly autistic people without co-occurring intellectual disability. Molecular Autism, 14(1), 12. https://doi.org/10.1186/s13229-023-00544-7

O’Nions, E., Lewer, D., Petersen, I., Brown, J., Buckman, J. E. J., Charlton, R., Cooper, C., Corbett, A., Happé, F., Manthorpe, J., Richards, M., Saunders, R., Zanker, C., Mandy, W., & Stott, J. (2023). Estimating life expectancy and years of life lost for autistic people in the UK: A matched cohort study. The Lancet Regional Health Europe, 36, 100776. https://doi.org/10.1016/j.lanepe.2023.100776

Rodgers, J., Cassidy, S., Pelton, M., Goodwin, J., Wagnild, J., Bhattarai, N., et al. (2024). Feasibility and acceptability of autism adapted safety plans: An external pilot randomised controlled trial. eClinicalMedicine, 73, 102662. https://doi.org/10.1016/j.eclinm.2024.102662

Rong, Y., Yang, C. J., Jin, Y., & Wang, Y. (2021). Prevalence of attention-deficit/hyperactivity disorder in individuals with autism spectrum disorder: A meta-analysis. Research in Autism Spectrum Disorders, 83, 101759. https://doi.org/10.1016/j.rasd.2021.101759

Russell, A., Cremen, C., Rainbow, E., & Melia, R. (2026). Suicide prevention interventions and supports for the Autistic community: A scoping review protocol. BMJ Open, 16, e099614. https://doi.org/10.1136/bmjopen-2025-099614

Santomauro, D. F., Hedley, D., Sahin, E., Brugha, T. S., Naghavi, M., Vos, T., Whiteford, H. A., Ferrari, A. J., & Stokes, M. A. (2024). The global burden of suicide mortality among people on the autism spectrum: A systematic review, meta-analysis, and extension of estimates from the Global Burden of Disease Study 2021. Psychiatry Research, 341, 116150. https://doi.org/10.1016/j.psychres.2024.116150

Trundle, G., Jones, K. A., Ropar, D., & Egan, V. (2023). Prevalence of victimisation in autistic individuals: A systematic review and meta-analysis. Trauma, Violence & Abuse, 24(4), 2282–2296. https://doi.org/10.1177/15248380221093689

Tsypes, A., Allen, T. A., Antezana, L., Beck, K. B., Conner, C. M., Scott, L. N., & Mazefsky, C. A. (2026). Integrating dimensional personality and autistic traits to predict suicidal ideation, suicide attempts, and nonsuicidal self-injury in autistic adults. Autism Research, 19(4), e70202. https://doi.org/10.1002/aur.70202

Tubío-Fungueiriño, M., Cruz, S., Conde-Pumpido-Zubizarreta, S., Pozo-Rodríguez, M., Hervás, A., Carracedo, A., & Fernández Prieto, M. (2026). A systematic review of the risk and protective factors for suicide in autism spectrum disorder. The Spanish Journal of Psychology, 29, e30. https://doi.org/10.1017/SJP.2026.10034

Baker, M., Lam, W., Virtosu, M., & de Lacy, N. (2026). Suicide attempt risk in autism: A national EHR study of 2.3 million individuals [Preprint]. medRxiv. https://doi.org/10.64898/2026.07.15.26358168