What the Intervention Evidence Supports
Article three of three. Minds in Session.
This article discusses suicide and suicidal thinking in general terms. It contains no description of methods and no individual accounts. If you are struggling as you read, the sources of help listed at the end are there for you.
The first article in this series set out the evidence on suicide risk among autistic people and located that risk was substantially in the environment and in service design. The second examined why the risk is so often missed. This article addresses what helps.
The evidence base for autism-specific suicide prevention is considerably less developed than the evidence describing the problem. Very little has reached the standard of a fully powered efficacy trial. What follows therefore distinguishes between what has been tested, what is supported by adjacent evidence, and what constitutes reasonable clinical inference not yet subjected to trial.
Adapted safety planning
A safety plan is a written, individualised sequence of steps a person follows when suicidal thinking intensifies, covering their own early warning signs, strategies that help them stabilise, people they can contact, and how to make the immediate environment safer. Safety planning has a reasonable evidence base in general populations. A meta-analysis by Nuij and colleagues published in the British Journal of Psychiatry in 2021 found a reduction in suicidal behaviour across six studies, though four of the six were assessed at high risk of bias and no significant effect on ideation was demonstrated.
Standard safety plans rest on assumptions that do not hold uniformly. They assume the person can detect a crisis developing, which alexithymia may preclude. They assume that telephoning a friend is stabilising rather than a source of acute anxiety. They assume that generic wording conveys the same meaning to all readers. Autistic adults involved in developing an adapted version described being presented with what amounted to a blank form and asked to populate it from an inner life they could not readily recognise and/or articulate.
Jane Goodwin and colleagues at Newcastle University, working with Cassidy and with autistic advisers, published the adaptation work in Autism in Adulthood in 2025. The adapted approach alters the process as much as the document published. The plan is completed with a trusted supporter, in advance and while the person is calm, rather than during a crisis. Wording is plain, concrete and literal. Visual aids, including a feelings wheel, are offered for identifying emotional states. The coping strategies are proposed by the supporter rather than solicited from a blank page. The format, colour, layout and medium are the person's own choice.
The approach was tested in an external pilot randomised controlled trial led by Jacqui Rodgers at Newcastle with Cassidy and colleagues, published in eClinicalMedicine in 2024, which randomised 49 autistic adults in the United Kingdom. The trial established feasibility and acceptability, which is the necessary preliminary step. It was not designed to establish efficacy in reducing suicidal behaviour, and the authors state this directly and call for a definitive trial. On present evidence this is the most developed autism-specific intervention available, and its efficacy continues to be an open question.
Psychological therapy that has been adapted
Autistic people are frequently offered standard psychological therapy, find it unhelpful and are recorded as not having benefited. The more accurate account is often that a therapy designed for one cognitive and communicative profile was delivered without modification to a person with another.
Adaptation is a matter of substance rather than pace. It involves working concretely and reducing reliance on metaphor and abstraction; making structure and expectations explicit rather than leaving them to be inferred; using written and visual material alongside conversation; attending to the sensory properties of the room, since capacity spent managing lighting and background noise is capacity unavailable for therapeutic work?; allowing longer for processing and for the working alliance to form; and undertaking emotional literacy work first where a person cannot yet identify and name internal states, rather than beginning an intervention that presupposes they can.
The 200 autistic adults surveyed by Camm-Crosbie and colleagues described consistent requirements: a clinician who understood autism, continuity of practitioner rather than a rotating team, sufficient sessions for a relationship to form, and flexibility in the mode and timing of contact. The National Institute for Health and Care Research has since issued a commissioning call for research on cognitive behavioural therapy adapted for autistic adults with a mental health problem, which indicates that the gap is recognised at national level and that the definitive evidence is still being generated.
Distinguishing burnout from depression
Autistic burnout, as defined by Raymaker and colleagues, denotes pervasive and prolonged exhaustion, loss of function and reduced tolerance of sensory input arising from chronic stress and a mismatch between demand and capacity without adequate support. It is a common context for suicidal thinking and is regularly treated as depression.
The distinction has direct treatment implications. Depression may respond to behavioural activation, to medication and to structured psychological work. Burnout responds primarily to reduced demand: fewer obligations, lower sensory load, permission to stop masking, and a recovery period measured in months rather than days. Encouraging increased activity in a person who is in burnout may extend the state it was intended to relieve.
Where a person has become progressively less able to manage tasks they previously handled, has become less tolerant of ordinary sensory input and has been under sustained pressure to perform; burnout warrants consideration before a conclusion of treatment-resistant depression is reached.
Reducing the requirement to camouflage
Camouflaging independently predicted suicidality in the 2018 study by Cassidy and colleagues, after adjustment for depression, anxiety and employment status. It follows that reducing the necessity for camouflaging is a plausible preventive strategy at organisational level.
In practice this means environments in which a person is not required to expend capacity, managing sensory conditions before beginning work; in which communication preferences are accommodated rather than read as social difficulty; in which written instructions and explicit expectations are standard; in which declining eye contact or a social event is unremarkable; and in which disclosure does not carry professional cost.
This falls within ordinary reasonable adjustment practice under the Equality Act 2010 and is generally inexpensive to implement. The evidential caveat should be stated plainly: the association between camouflaging and suicidality is established, but no trial has tested reduced camouflaging as a suicide prevention intervention. This is reasonable inference from robust correlational findings, not demonstrated causation.
Earlier and more accessible diagnosis
Given the proportion of people who died by suicide in whom Cassidy and colleagues found evidence of an unidentified autistic profile, improving access to diagnostic assessment functions as a preventive measure.
Diagnosis is not itself an intervention. Its value lies in what it makes available. It reframes a history the person may have understood as a record of personal failure. It brings adjustments within reach. It changes how clinicians interpret presentation. One point warrants emphasis: the period following a diagnosis received in adulthood involves a substantial reappraisal of one's own history and is demanding in its own right; and post-diagnostic support at that stage forms part of the intervention rather than an optional addition to it.
Practical guidance
If you are autistic and struggling. Distress need not resemble anyone else's to be real and to warrant help. Where a service has indicated that you are too articulate or too capable to require support, that reflects the service's assessment criteria rather than your need. It is reasonable to request adjustments to how you are assessed, to bring someone with you, to respond in writing, and to state that you find it difficult to identify or describe how you feel. Completing a safety plan with someone you trust while circumstances are stable is worth doing in advance of need.
If you are close to an autistic person. Presentation is an unreliable guide. A composed, factual account of a serious difficulty remains an account of a serious difficulty. Ask directly and in plain terms rather than by implication and expect a literal answer to a literal question. Attend to functional change, meaning what the person is no longer able to manage, as closely as to emotional expression, since functional change is frequently the more reliable indicator.? Asking directly about suicidal thinking does not increase it. The systematic reviews of Dazzi and colleagues in Psychological Medicine in 2014 and of DeCou and Schumann in Suicide and Life-Threatening Behavior in 2018 found no evidence of an iatrogenic effect, although the available studies concern research and clinical screening settings rather than conversations between family members, and few have included autistic participants.
If you commission or run services. The implications are specific. Consider autism rather than assume its absence. Adopt adapted assessment approaches. Do not treat articulacy as a proxy for wellbeing. Close the gap between mental health and autism services into which people currently fall. Provide continuity of practitioner. Train staff substantively rather than nominally.
Conclusion
A risk driven substantially by unmet need, inaccessible services and assessment instruments developed for a different population is a risk that can be reduced. Each of those elements reflects decisions that are made and can be made differently.
Most autistic people are not suicidal. Among those who reach that point, the research record repeatedly describes people who sought help and received help that did not fit. That is the element within reach.
Sources of support
If you are struggling, you do not have to wait until things are worse to ask for help.
• Samaritans. Free, day or night, 365 days a year. Call 116 123, at any hour. The Welsh Language Line, 0808 164 0123, is open from 7 pm to 11 pm every day. A Relay UK service is available for d/Deaf and hard of hearing callers, and details are on the Samaritans website. Branches also offer face-to-face support; it is worth telephoning ahead to arrange a time.
• Shout. Free, confidential text support, 24 hours a day, which may suit you better if speaking on the telephone is difficult. Text SHOUT to 85258.
• PAPYRUS HOPELINE247, for anyone under 35 experiencing thoughts of suicide, and for anyone concerned about a young person. Call 0800 068 4141, text 88247, or email pat@papyrus-uk.org.
• Urgent NHS mental health support. In England and Wales, call 111 and select the mental health option. In Scotland, call NHS 24 on 111. In Northern Ireland, Lifeline is available on 0808 808 8000.
• Your GP, who can refer you into local services.
• In an emergency, call 999 or attend your nearest A&E department.
About the author
Esther John , Clinical and Forensic Psychologist, Minds in Session.
Minds in Session provides psychological assessment, therapy and organisational consultancy, with particular expertise in neurodiversity, leadership and workplace mental health.
References. Rodgers J, Cassidy S, Pelton M, Goodwin J et al. (2024) Feasibility and acceptability of autism adapted safety plans: an external pilot randomised controlled trial. eClinicalMedicine 73:102662. Goodwin J, Gordon I, O'Keeffe S et al. (2025) Adapting safety plans for autistic adults with involvement from the autism community. Autism in Adulthood 7(3):293-302. Nuij C et al. (2021) Safety planning-type interventions for suicide prevention: meta-analysis. British Journal of Psychiatry 219(2):419-426. Cassidy S, Bradley L, Shaw R, Baron-Cohen S (2018) Risk markers for suicidality in autistic adults. Molecular Autism 9:42. Camm-Crosbie L, Bradley L, Shaw R, Baron-Cohen S, Cassidy S (2019) 'People like me don't get support'. Autism 23(6):1431-1441. Raymaker DM et al. (2020) Defining autistic burnout. Autism in Adulthood 2(2):132-143. Cassidy S, Au-Yeung S, Robertson A, et al. (2022) Autism and autistic traits in those who died by suicide in England. British Journal of Psychiatry 221(5):683-691. Dazzi T, Gribble R, Wessely S, Fear NT (2014) Does asking about suicide and related behaviours induce suicidal ideation? Psychological Medicine 44(16):3361-3363. DeCou CR, Schumann ME (2018) On the iatrogenic risk of assessing suicidality: a meta-analysis. Suicide and Life-Threatening Behavior 48(5):531-543.
