The hidden scale of unmet need: why many pupils with dyslexia, autism and ADHD are identified late, or not at all

The numbers are hard to absorb. The Department for Education (2025) records over 1.7 million pupils in England with special educational needs: 19.6%, or about one in five. The Education Policy Institute adds important context: while around a fifth of pupils have SEN in any given year, close to four in ten children (37%) are identified with SEND at some point between ages 5 and 16 (around 44% of boys and 30% of girls). Those, though, are only the children who are eventually identified.

For dyslexia, identification is widely thought to lag well behind need. A 2019 report by the All-Party Parliamentary Group for Dyslexia and other Specific Learning Difficulties, for which the British Dyslexia Association provides the secretariat, estimated that over 80% of people with dyslexia leave school without a diagnosis, noting that, of an estimated 870,000 dyslexic schoolchildren in England, fewer than 150,000 had been formally identified. This is an advocacy estimate from 2019 rather than an official statistic, and it depends on a contested prevalence assumption (often put at around 10%); even so, it reflects a widely reported gap between likely need and formal identification. Plain-language definitions of these terms are in our glossary.

For ADHD, the NHS England independent ADHD Taskforce (2025) describes the condition as "under-recognised, under-diagnosed and under-treated," estimating a population prevalence of around 3-5% (about 5% in children and 2-3% in adults). It notes that women and girls are much less likely to be diagnosed, and that the underlying prevalence has not risen. Instead, rising referrals reflect greater awareness against a backdrop of historic under-recognition.

Demand for assessment is high. The Children's Commissioner for England (2024) estimated that around 400,000 children, roughly 3% of all children in England, were waiting at the end of 2022-23 for a first appointment for a suspected neurodevelopmental condition, with many waiting years for an initial assessment.

None of the signs below means a child definitely has dyslexia, ADHD or autism. But because identification is so often delayed, it can help to notice patterns over time rather than one-off difficulties.

Does your child find writing harder than it should be? Do they avoid homework, not out of laziness, but because it feels overwhelming? Do they lose focus, fidget, or take much longer than classmates to finish tasks? Do they seem anxious about the work itself rather than about friendships? Do they work hard but get results that don't reflect their effort?

Any one of these can have many explanations. But the evidence above shows that a large number of children with unmet needs go through school without anyone recognising the pattern. If something doesn't quite add up, it is reasonable to look more closely, and you can begin gathering useful information before any formal assessment. Our guides on how parents can support schools in identifying needs and on advocating for school support set out practical next steps.

Qwixl:Milo is a Chrome extension that works inside Google Docs, where most students do their homework. It was built with this gap in mind: the space between suspecting a need and getting formal support.

As a child writes, Milo quietly observes typing patterns: speed, consistency, pauses, how often they correct and delete, and how their flow changes across a session. Over time these can surface four processing indicators: literacy and learning, attention and focus, writing and motor skills, and processing and working memory (screening signals, not diagnoses). They are the kinds of everyday signals professionals look for, gathered naturally from homework rather than a one-off test.

Milo also offers adaptive writing support: it reads the assignment, helps with structure and planning, and includes a chat assistant that adjusts to a child's writing level. For students who need simpler language, shorter prompts or help getting started, it adapts; for confident writers, it stretches them.

This is not a diagnosis. Milo's indicators are screening-level signals, clearly presented as such. But they give families a way to understand how their child approaches writing, whether they already have a diagnosis, are waiting for one, or are simply wondering whether the difficulty is more than "not trying hard enough."

Crucially, Milo does not require a referral, a waiting list or an assessment fee. From the first homework session it begins building a picture, and the more a child writes, the richer that picture becomes.