Assessment & Research

'Picking the best of a bad bunch': Exploring stakeholder perspectives of self-harm assessment tools for autistic adults.

Newell et al. (2025) · Autism : the international journal of research and practice 2025
★ The Verdict

Off-the-shelf self-harm forms are rejected by autistic adults—partner with them to build something new.

✓ Read this if BCBAs who assess self-harm or suicidal thoughts in verbal autistic adults.
✗ Skip if Clinicians working only with young children or non-speaking clients.

01Research in Context

01

What this study did

The team asked autistic adults, clinicians, and carers to judge three common self-harm questionnaires.

They ran long interviews so people could explain what felt wrong or missing.

No scores were taken; the goal was to hear why the tools felt unusable.

02

What they found

Every group said the same thing: all three tools fail autistic adults.

People hated the vague wording, the long Likert scales, and the lack of autism-specific triggers like sensory overload or social burnout.

Because the forms felt impossible, many said they would simply skip them, leaving real risk hidden.

03

How this fits with other research

Hedley et al. (2023) already built a suicide-thought scale, the SIDAS-M, with autistic input and got hopeful early data. Victoria’s team shows why that work was needed: the older generic tools are rejected.

Davidovitch et al. (2018) found that generic quality-of-life scales also miss autism-specific life domains. The same problem repeats here—standard mental-health forms leave out what matters to autistic adults.

Adams et al. (2022) saw parents rate the school-refusal scale as poor for autistic children. Across ages and topics, one-size-fits-all checklists keep failing; individualized or co-built tools are the emerging fix.

04

Why it matters

If you screen for self-harm, don’t hand an autistic client a generic form and call it done. Expect to co-create short, concrete questions that map onto real-life triggers like changes in routine or sensory pain. Start by asking the client which words or scales make sense to them, then pilot two or three items next session instead of dumping a full packet.

Free CEUs

Want CEUs on This Topic?

The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.

Join Free →
→ Action — try this Monday

Show your client one Likert scale, ask what the numbers mean to them, and adjust the wording together.

02At a glance

Intervention
not applicable
Design
qualitative
Sample size
9
Population
autism spectrum disorder
Finding
negative

03Original abstract

Autistic people are at greater risk of self-harm than non-autistic people, yet no tools exist specifically for assessing self-harm in this population. We therefore conducted two focus groups with autistic adults (n = 5) and professionals (n = 4) to examine their perspectives on three existing self-harm assessment tools (Non-Suicidal Self-Injury - Assessment Tool, Inventory of Statements About Self-Injury and Questionnaire for Non-Suicidal Self-Injury). Using a reflexive thematic analysis, we found one overarching theme - 'picking the best of a bad bunch' - where participants acknowledged some strengths of the tools, but multiple limitations outweighed these. Subthemes included cognitive considerations (e.g. complexity, length, working memory demands, introspection), missing elements (e.g. broader self-harm behaviours, functions, context, support) and challenges in conceptualising self-harm, particularly around intentionality and stimming. Participants also stressed the importance of addressing stigma, considering co-occurring conditions like attention-deficit hyperactivity disorder and intellectual disabilities and upholding a duty of care. Overall, findings show that existing self-harm assessment tools are not appropriate or acceptable for autistic adults who self-harm, indicating a clear need for a new self-harm assessment tool developed with and for autistic people.Lay abstractThe perspectives of autistic adults and professionals on existing self-harm assessment toolsSelf-harm is defined as when someone intentionally hurts or poisons themselves, regardless of the reason. This can include suicide attempts as well as behaviours such as cutting, hitting or burning as a way to express or manage difficult feelings. Concerningly, autistic people are more likely to self-harm than non-autistic people, but there are currently no tools specifically designed to assess self-harm in this group. This makes it harder for researchers and service providers to identify autistic people who self-harm and offer the right support or treatment. Previous studies have suggested that three existing self-harm assessment tools could be adapted for autistic people, but it is unclear what autistic adults and professionals think about these tools. We conducted two focus groups: one with autistic adults who have self-harmed and another with professionals who work with autistic people who self-harm. Both groups felt that none of the three tools were suitable for autistic people. They explained that some questions might be harder for autistic people to answer, such as those requiring them to remember specific details or identify their emotions. They also noted the tools missed out on important aspects of the autistic experience, like managing sensory and social overload. Both groups highlighted that self-harm is complex and discussed the overlap with behaviours such as stimming (e.g. repetitive movements or sounds). They emphasised the importance of a supportive approach to self-harm that considers the role of co-occurring conditions, like attention-deficit hyperactivity disorder (ADHD) or intellectual disabilities, and ensures researchers and professionals prioritise participant safety. These findings suggest that instead of adapting existing tools, a new one should be developed specifically with and for autistic people. A tailored tool could help identify self-harm earlier and lead to better support for autistic adults.

Autism : the international journal of research and practice, 2025 · doi:10.1177/13623613251348555