A Response to Kathleen Stock On “If a Child Says ‘I Won’t’, Don’t Reach for a Label” (The Sunday Times, 22 February 2026)

On the 22nd February 2026, The Sunday Times published a piece by Kathleen Stock dismissing pathological demand avoidance as, in her own framing, a "pseudo-clinical label" that harms children by excusing behaviour that simply needs firmer parenting. The article was illustrated with anecdotes about her own son, who fussed about toothbrushing, turned out to be fine, and never needed professional involvement.

The very next morning the Measure What Matters investigation published its forensic review of 25 neurodivergent children whose deaths between 2020 and 2024 were preceded by documented, repeated failures in SEND provision. Children who had EHCPs, children whose distress had been visible, recorded, and ignored, children who had been treated, in case after case, as defiant rather than dysregulated, as behavioural problems rather than as people whose nervous systems were failing to survive the environment they were placed in. That same morning, the Government published its Schools White Paper, proposing to remove legally enforceable Education, Health and Care Plans from the majority of children who currently hold them.

The timing of Stock's article is important, not because she intended it to land alongside a report about dead children, she presumably did not, but because it is impossible, in February 2026, to write a piece arguing that demand avoidance is largely parental overconcern and professionally-driven label-seeking without that argument doing political work. Whether or not that work was intended, it is the work it does.

What the article actually argues

Stock's central claim is straightforward: PDA does not appear in any mainstream diagnostic manual, therefore clinicians and parenting coaches should treat it with caution; those who do not are probably keeping themselves in business. Her evidence for this is primarily that her own son hated brushing his teeth, grew out of it, and is now a charming young man.

She then makes a second argument: that PDA guidance - which often recommends reducing direct demands, offering autonomy and collaborative approaches - is counterproductive, because it prevents children from building tolerance and teaches them that resistance works.

And she concludes with a third: that labels like PDA harm children by offering them an identity built around a disorder, removing their agency over their own behaviour.

These are not uniformly stupid arguments. The concern about over-extension of clinical frameworks into ordinary developmental variation is legitimate. The question of what constitutes a meaningful threshold for diagnosis is a live one in the peer-reviewed literature, and researchers across the field continue to debate it. Professor Stock is not wrong to note that PDA does not currently appear in DSM-5 or ICD-11.

But the argument she constructs from these observations is riddled with errors of reasoning, and it rests on a complete absence of the evidence base that actually does exist. That evidence base does not support her conclusions. In several key respects, it directly contradicts them. 

What the research actually shows

Amelia Green's 2024 study - a systematic research project conducted through the University of Birmingham and one of the most comprehensive examinations of PDA children in the English education system to date - found that 86% of PDA children and young people struggled with school attendance, beginning from as early as Reception year.

86 per cent - not a small subset and not children with additional complicating factors - 86% of a sample of children identified with demand avoidant profiles of autism, across both primary and secondary settings, experienced significant difficulties attending school.

Gore Langton and Frederickson (2016) - published in the International Journal of Research in Special Educational Needs - found that 95% of PDA children started in mainstream schools, but experienced high rates of placement breakdown and exclusion. 71% had been unable to attendschool at some point. Brede et al. (2017) replicated these findings, with placement breakdowns attributed specifically to unmet needs, the demands of the school environment, and a systemic lack of understanding of the child's difficulties.

Truman, Crane, Howlin and Pellicano (2021), published in the International Journal of Inclusive Education, identified five consistent themes in the negative school experiences of children with demand avoidant profiles: being misunderstood by school staff; masking difficulties within school; lack of appropriate support; anxiety and mental health difficulties caused by those negative experiences; and parents feeling disenfranchised by the education system.

The question is not whether these children exist. They are visible in the data, in the case notes, and in 1,253 family testimonies submitted to the Measure What Matters investigation.

The question Stock wants us to ask - is this label meaningful? - is a reasonable scientific question in the abstract. But the way she asks it, through the lens of her son's toothbrushing, leaves the 86% entirely unaddressed. She offers no explanation for what is happening to them. She does not engage with why, if this is simply children who need firmer parenting, so many of them are not in school at all.

On the neuroscience of demand and the evidence for low-demand approaches

Stock's second argument - that low-demand approaches prevent children from building tolerance and teach them that refusal works - misunderstands the aetiology of demand avoidance in autistic children fundamentally.

Dr Georgie Siggers and Becky Day's 2024 retrospective audit, published in the British Journal of Psychiatry, examined the case notes of 20 autistic children who had been unable to attend school for at least three months. Every single child - 100% of the cohort - showed the same cluster of presentations: chronic exhaustion, loss of previously acquired skills, heightened sensory needs, social withdrawal, mood dysregulation, and physical complaints. The uniformity of findings was striking enough that the researchers called explicitly for an urgent shift away from potentially punitive approaches, and for a fundamental re-evaluation of educational practices for autistic children. It states that the response these children need is not firmer demands, but a different kind of understanding altogether.

Stuart et al. (2020), published in Child and Adolescent Mental Health, identified the relationship between intolerance of uncertainty -a well-documented feature of autistic neurology - and demand avoidance behaviour. The connection between anxiety and demand avoidance has since been replicated by Johnson and Saunderson (2023) and by White et al. (2023), both published in peer-reviewed journals. O'Nions and Eaton (2020) outline the evidence for low-demand, low-arousal approaches specifically.

The reason collaborative, low-demand approaches are recommended is not because clinicians are indulging children or keeping themselves in business. It is because the research shows these children's nervous systems are experiencing genuine threat, and that demand-escalating approaches increase rather than resolve that threat. The logic of "if we just push harder, they'll learn to cope" is not supported by the evidence. The Siggers and Day findings suggest that pushing harder produces burnout, not tolerance.

Stock invokes the image of "a four-year-old" as her reference point. But the children in the peer-reviewed literature - the children in school avoidance, placement breakdown, zero provision, inquest findings - are not primarily four-year-olds fussing about bathtime - they are children whose capacity to function within the demands of an institutional environment has collapsed, often after years of trying.

What 'no diagnostic manual listing' actually means

Stock argues that PDA's absence from DSM-5 and ICD-11 indicates clinicians should approach it with caution. This is a legitimate observation, and one that responsible practitioners already hold in mind - the peer-reviewed literature itself reflects ongoing debate about diagnostic boundaries, terminology, and whether PDA constitutes a distinct profile or a dimension of autistic presentation.

But absence from a diagnostic manual does not mean absence from clinical reality. Autistic burnout does not appear as a standalone category in DSM-5 either. The Siggers and Day findings - 100% concordance in symptom presentation across a cohort of children unable to attend school - describe something clinically real and clinically significant, regardless of how the diagnostic taxonomy eventually categorises it.

What the research consensus does support is that demand avoidant behaviour in autistic children is real, is anxiety-driven, is associated with severe educational outcomes, and responds to approaches that are fundamentally different from those that work for neurotypical opposition or typical developmental defiance. While the terminology used to describe it is genuinely contested, the existence of the children is not.

Measure What Matters

The Measure What Matters investigation reviewed the deaths of 25 neurodivergent children and young people between 2020 and 2024, drawing on Local Child Safeguarding Practice Reviews, inquest findings, Prevention of Future Deaths reports, and 1,253 family testimonies. Five of those 25 children left no trace in any public record - no safeguarding review, no inquest, no Prevention of Future Deaths report. The investigation notes plainly: it appears the system did not look.

In case after case, the pattern preceding crisis was the same. A child whose behaviour deteriorated. A child whose attendance collapsed. A child read as defiant rather than as dysregulated. A child for whom the system's consistent response was not understanding but punishment - removal, exclusion, disciplinary escalation. Children with EHCPs whose legally enforceable rights were not applied.

These were not children who fell through the gaps. There were no gaps. There was a legal framework, with clear duties, enforceable rights, and statutory timescales. It was simply not applied.

The Measure What Matters report states it plainly: more than one in six families who submitted testimony about SEND maladministration described their child reaching the point of wanting to die. These were not children whose parents needed firmer parenting techniques, these were children whose systems-generated trauma -documented in the Cerebra Systems Generated Trauma report and visible across the family testimony database - represents the human cost of a framework that existed but was never applied.

Professor Stock's article does not address this. It cannot address this, because its analytical frame - parental overconcern, label-seeking, therapeutic overreach - has no room for it. The logic of her argument requires these children to be the exception, the edge case, the genuine few for whom a label might be appropriate. The evidence suggests they are not exceptions, they are a pattern.

On the White Paper

On the same morning, the Government's Schools White Paper proposed removing legally enforceable EHCPs from most children who currently hold them, replacing them with Individual Support Plans whose legal enforceability remains unclear. Autism and ADHD - the diagnoses present in every single detailed case study in the Measure What Matters investigation - are framed in the White Paper's accompanying documents as "predictable" needs that mainstream schools should accommodate without statutory plans.

The Government's own Equalities Impact Assessment identified no significant adverse impacts for disabled children from these proposals.

In this political context, an article arguing that neurodivergent children's needs are largely parental overconcern and label-seeking is not a neutral academic contribution to a scientific debate. It is fuel for a policy argument that is already being made, at pace, by people who have read the Measure What Matters report and responded by proposing to remove the legal protections that existed - unenforced - when those children died.

A note on the argument about labelling

Stock's final concern - that labels harm children by offering them an identity built around a disorder - deserves a direct answer. The argument is that children who are told they have PDA will internalise this as a fixed identity and use it to avoid building adaptive capacity.

This concern is not without some legitimacy in how labelling can function in certain contexts, but it runs in precisely the wrong direction for the children we are discussing.

The families submitting testimony to the Measure What Matters investigation, the parents in the Gore Langton and Frederickson (2018) study, the carers described throughout the peer-reviewed literature - these are not families who sought a label and relaxed. They are families who sought understanding because their children were in crisis, who found that a framework for understanding demand avoidance behaviour led to approaches that reduced crisis rather than entrenching it, and who then discovered that the system did not recognise that framework and would not act on it.

The Amelia Green study found that having an EHCP correlated significantly with educational outcomes - but that 59% of parents felt their child's EHCP did not adequately describe their demand avoidant behaviours. What harmed these children was not having a label - what harmed them, in case after case, was being in institutions whose staff read their distress as defiance, whose responses escalated demands rather than adapting to them, and whose legal duties to provide appropriate education went unfulfilled.

Removing the label does not remove the child – it simply removes the framework that gives those who work with them a way to understand what is happening.

What this letter is asking

What this letter asks - of Professor Stock, of The Sunday Times, and of anyone who read that article and found it persuasive - is simply this: that when we write and speak about neurodivergent children, we do so with the rigour and care the subject demands. That we engage with the evidence that exists rather than the anecdote that is easier to tell. That we understand what political work our arguments do before we publish them. That we consider, seriously, who is harmed when we get this wrong - not in the abstract, but in the specific, documented, irrefutable reality of what happens to children whose distress is met with dismissal rather than understanding.

Professor Stock's son turned out fine - the children in the Measure What Matters investigation did not. That difference is not explained by parenting philosophy, or by the presence or absence of a label, it is explained by what the system was willing to understand about them - and what it chose, instead, to call defiance.

Amber Windsor, Canary SEND

References

Brede, J. et al. (2017). 'Inclusion of pupils with an autism spectrum condition at secondary school: what is the perspective of the pupils themselves?' Autism and Developmental Language Impairments.

Gore Langton, E. and Frederickson, N. (2016). 'Mapping the educational experiences of children with pathological demand avoidance'. International Journal of Research in Special Educational Needs, 64(4), pp. 254–263.

Gore Langton, E. and Frederickson, N. (2018). 'Parents' experiences of professionals' involvement for children with extreme demand avoidance'. International Journal of Developmental Disorders, 64(1), pp. 16–24.

Green, A. (2024). An exploration into the educational experiences of PDA children. University of Birmingham.

Johnson, M. and Saunderson, H. (2023). 'Examining the relationship between anxiety and pathological demand avoidance in adults: a mixed methods approach'. Frontiers in Education, 8.

Measure What Matters (2026). Children Died. The System Looked Away. Published 23 February 2026. measurewhatmatters.blog

O'Nions, E. and Eaton, J. (2020). 'Extreme/pathological demand avoidance: an overview'. Paediatrics and Child Health, 30(12), pp. 411–415.

PDA Society (2018). Being Misunderstood: Experiences of the Pathological Demand Avoidance Profile of ASD. pdasociety.org.uk

Siggers, G. and Day, B. (2024). 'Beyond School Avoidance: Recognising, Identifying, and Addressing Autistic Burnout in Children'. BJPsych Open, doi: 10.1192/bjo.2024.433

Stuart, L., Grahame, V., Honey, E. and Freeston, M. (2020). 'Intolerance of uncertainty and anxiety as explanatory frameworks for extreme demand avoidance in children and adolescents'. Child and Adolescent Mental Health, 25(2), pp. 59–67.

Truman, C., Crane, L., Howlin, P. and Pellicano, E. (2021). 'The educational experiences of autistic children with and without extreme demand avoidance behaviours'. International Journal of Inclusive Education.

White, R. et al. (2023). 'Understanding the Contributions of Trait Autism and Anxiety to Extreme Demand Avoidance in the Adult General Population'. Journal of Autism and Developmental Disorders, 53(7), pp. 2680–2688.

Cerebra (2022). Systems Generated Trauma Report. cerebraresearch.org

If you are affected by any of the issues raised in this letter, support is available. Samaritans: 116 123 (free, 24 hours). Papyrus HOPELINE247: 0800 068 4141 or text HOPE to 88247.