The personal independence payment (PIP) is a government scheme providing financial support to people with a long-term condition or disability who face extra costs due to difficulties arising as a result of these conditions. Recent reports have highlighted large rises in the number of young people recorded as having ADHD or autism, and in the disability benefits paid to support them. By April 2026, just over 205,400 people aged 16–24 were receiving PIP – more than three times the 2019 figure, at an estimated £1.36bn in 2025/26.
Although payments are higher in more deprived areas, where the functional impairment these conditions cause tends to be greater and harder to weather, this analysis of how PIP for 16- and 17-year-olds has changed across England since 2019 finds that the picture has become more complex in recent years.
PIP for ADHD and autism: a weakening link with deprivation?
A series of reports for the government in recent months have highlighted very large increases in the number of young people recorded as having, or suspected to have, autism and attention deficit hyperactivity disorder (ADHD). 1 The reports point to the role played by these conditions in increased disability benefit spending 2 ; pressures on the health and education systems 3 ; and in the concerning numbers of young people not in education, employment or training 4 .
The most recent Adult Psychiatric Morbidity Survey for England found that 25% of 16–24-year-olds screened positive for ADHD symptoms in 2023/24 – indicated through self-reported difficulties linked to symptoms – up from around 14% in 2007. Similar increases, albeit from lower baselines, were also seen in older working-age groups. 5
For autism, the Independent Review into Mental Health Conditions, ADHD and Autism noted that self-reported prevalence in the 16–24 age group reached 9% in the 2025 GP Patient Survey, up from around 5% in 2022, while parent-reported prevalence in children sat at just under 12% in 2024, up from 9.5% in 2022. 6
These figures for self-identification and symptom-reporting are higher than estimates of clinical prevalence, and are likely to at least in part reflect growing awareness of neurodiversity, as well as a broadening of the criteria used to identify associated conditions. 7 However, regardless of underlying cause, the figures are indicative of the growth in demand for clinical and social support 8 , and of increased distress, particularly within the younger population. 9
Indeed, by April 2026, just over 205,400 people aged between 16 and 24 were receiving the personal independence payment (PIP) benefit for disabilities linked to their ADHD or autism, indicating that their condition significantly impaired their daily living. These numbers represent more than a three-fold increase since 2019, at an estimated annual cost in 2025/26 of £1.36bn. PIP benefits are paid by the Department for Work and Pensions, but to give sense of scale, total estimated payments to 16–25-year-olds for autism and ADHD alone in 2025/26 were just less than the planned NHS budget for the entirety of child and adolescent mental health services that year. 10
This increased distress requires urgent reflection on a number of fronts, including exploring the extent to which the ways educational 11 and work 12 settings are organised play a causal role in a situation where up to a quarter of the young adult population experience difficulties that leave them feeling 'abnormal”. 13
For the health care system specifically, large increases in the numbers seeking formal clinical diagnosis and treatment have contributed to very significant pressures on NHS mental health and learning disability services. 14 This has prompted an interest in distinguishing – potentially in advance of clinical diagnosis – high-priority cases from lower-priority cases; partly to avoid crowding out the most severe-need cases 15 , but also to avoid the unnecessary and possibly harmful pathologisation of a sizeable proportion of the population. 16
The policy challenge: distinguishing differing levels of need or impairment
Recent reports from both the ADHD Taskforce 17 and the Independent Review into Mental Health Conditions 18 highlight the concept and measurement of functional impairment as potentially useful for distinguishing between different levels of need within the growing numbers of people identified as having ADHD and autism, and to appropriately prioritise care and support.
'Functional impairment' refers to difficulties a person experiences in their everyday life as a result of their symptoms. The concept points to a key difference between how neurodevelopmental conditions manifest and how physical health conditions are commonly understood, in that it distinguishes symptoms from impairments, and observes that not all who have symptoms necessarily experience the same level of impairments.
Whether or not impairments lead to a loss of function can be highly contextual, with functional impairments emerging in some settings but not others. 19 Indeed, in some activities and careers, traits linked to ADHD and autism are viewed as beneficial or advantageous. 20 The relevance of social context to functional impairment echoes elements of the social model of disability, whereby disability (or functional impairment) is seen to stem from design failures in work, education and broader society, which actively disable the individual through processes and features which make their inclusion problematic and limit their social, educational and employment outcomes. 21
Yet although clinical diagnoses of ADHD and autism involve some assessment of functional impairment across different settings (such as work and social life), both the Independent Review and ADHD Taskforce reports describe a lack of clarity and consistency in how these assessments inform diagnosis. 22 Furthermore, for ADHD in children and young people, clinicians are advised to consider discounting evidence of high functioning (for example, high academic outcomes) if it is felt that the person is still not achieving their 'potential', or that their achievements mask significant struggles, or are possible only as a result of family support. 23
This implies that while clinical diagnosis can be taken to confirm the presence of defined symptoms or traits, it may not consistently indicate that the individual has fallen below a pre-identified or objective level of functioning or outcome.
For some, the notion of an objective, universal level of functioning or outcome against which an individual is assessed as either impaired or non-impaired is deeply problematic. Objections are both moral – highlighting the harm that can be caused when some individuals are defined as 'abnormal' or some forms of living as 'suboptimal' – and epistemological – questioning how we can know what constitutes optional or adequate functioning for others. 24
However, when confronted with the practical challenge of allocating finite clinical and welfare resources, context and functioning become central to the answering the question 'who is in greatest or most pressing need?' This point is argued in the 'Young people and work' interim report on young people not in education, employment or training (NEET), which states: "A young person with… a stable home, a supportive school and an engaged parent is in a fundamentally different position from a young person with the same clinical presentation who is living in temporary accommodation, out of school and without an adult advocate". 25
As the NEET report suggests, at the broad population level, we might expect levels of functional impairment (disability) – as related to both neurodiversity and other conditions – to at least partly correlate with local area deprivation indicators for two reasons. First, in so far as functional impairments limit an individual's ability to progress in employment and education, it would be reasonable to expect areas with a relatively high proportion of people with functional impairment to have relatively low levels of secure employment and educational outcomes. Second, areas that have fewer educational or employment opportunities may exacerbate people's experience of symptoms by providing a less supportive environment and social context.
To explore whether this hypothesis holds true to functional impairment stemming from ADHD and autism, and to explore how support for such disabilities is currently distributed across the population, we turned to data collected as part of the personal independence payment (PIP) benefit system.
Personal independence payments – what the data can and cannot tell us
PIP is a non-means-tested and non-taxable benefit which is designed to help fund the additional costs a person faces due to their disability. Claimants are asked to provide evidence of their diagnosis or symptoms, and to describe how these lead to functional impairments in their daily lives, through an assessment based around set common daily activities (such as washing, dressing, eating and travel) with defined criteria designed to objectively assess the individual's ability to complete each activity. 26
Only applicants who are able to establish that their condition means they cannot reliably and safely complete tasks spanning a number of normal daily activities without support qualify to receive PIP payments. 27
Organisations and charities supporting people with conditions associated with neurodivergence highlight that impairments relevant to PIP assessments might include, for ADHD 28 : procrastination, leading to forgetting to prepare meals in a timely fashion; poor impulse control making social interactions difficult or travel on public transport unsafe; and forgetfulness around personal hygiene. For autism 29 , noted impairments include many of these 30 but also add: difficulties processing information, causing tasks and interactions to take longer and requiring the supervision of others; sensory sensitivities making some everyday tasks complicated or more stressful; anxiety completing some tasks; and difficulties communicating with and understanding other people.
While the fact someone is employed is not in itself permitted to disqualify them from receiving PIP, the assessment process will take into account activities completed in their work when considering an individual's impairments and ability to complete specific tasks. As a non-means-tested benefit, an individual's context – such as the availability of family support, income, or a financial cushion to fund the additional costs of disability – are ignored. On the face of it, however, data on who receives PIP for ADHD and autism might be expected to tell us more than diagnosis data alone, in that it gives an indication of who experiences functional impairment as a result of their condition, based on relatively objective or uniform criteria.
It is important to note that the recent interim report from the government's review of PIP (the Timms Review) highlighted widespread dissatisfaction with the system, including concerns that assessments are not consistent or fair and are particularly poor at capturing functional impairments (or disability) in fluctuating and context-specific conditions such as ADHD and autism. This corroborates other studies. 31
It should be assumed, therefore, that PIP data – like other administrative data – can give us only a partial view of how disability and the need for support is distributed across the population. The full picture of need is obscured first by inequalities in access to formal diagnosis and then further clouded by potential inconsistencies in PIP assessments themselves.
While PIP data can only give us a partial picture of assessed need, it does give us a definitive insight into whose disabilities are currently recognised and receiving formal support, in the form of the weekly benefit payment, worth the equivalent of just under £9,000 for a full year to 16–19-year-olds in 2026. While there has been much interest and concern at the rising cost of the PIP benefit bill over recent months – particularly regarding young people and those receiving the benefit for ADHD and autism 32 – much less attention has been given to how that support is distributed between different levels of deprivation.
Monthly data on PIP recipient numbers is available by individual years of age and can be broken down by the single 'main condition' for which the payment is given, as well by the recipient's local area. This allows recipient numbers to be explored alongside local area characteristics, such as age, deprivation and ethnicity. There are important limitations to this form of analysis, as there is no guarantee that individual recipients perfectly reflect the characteristics of the areas in which they live, so insights need to be triangulated with other approaches before firm conclusions can be drawn. 33
Our analysis focused on the 16–17-year-old age group. Focusing on this narrow age band reduces some (although not all) of the limitations inherent in analysis which infers individual characteristics from local areas, as the 16- and 17-year-old age group is more geographically static than the 18+ group – an age when significant numbers of young people move home to attend university, college or to start independent adult life. 34 Although recorded and identified cases of autism and ADHD have been increasing since before 2019, we only explored PIP data from 2019 onwards, when geographical differences are less likely to be affected by the phased roll-out of the PIP system which began in 2013. 35
PIP for all conditions other than ADHD and autism
In June 2019, a total of 34,879 16–17-year-olds were found eligible to receive PIP payments in England, of whom 16,045 (46%) had their main disability recorded as autism or ADHD. Of the remaining 18,834, the majority (65%) had their main disability recorded as other 'psychiatric disorders”, with the most significant being learning disability, depression and anxiety. In the chart below, we show the June 2019 recipient rate for all disabilities other than ADHD and autism in 296 local authorities in England. 36 The percentage of 16–17-year-olds receiving PIP for those disabilities that month is plotted against the local authority's employment deprivation score, which is based on the proportion of the working-age population in each authority who are unable to secure employment (see data notes at the end of this briefing for more details). 37
In 2019, more deprived areas had a larger proportion of young people receiving PIP for disabilities other than ADHD or autism
ChartThe grey best-fit line indicates the level of association in June 2019 between employment deprivation and the 16–17-year-old PIP recipient rate for all disabilities other than ADHD and autism. Although not all the blue and purple dots (local authorities) sit perfectly on the line, most are fairly close to it. A deprivation gradient is clearly visible, with higher PIP rates the further along the deprivation scale a local authority sits. Simple linear regression analysis suggests that employment deprivation explains 41% of the variation between each authority's recipient rate.
This is consistent with our expectation that functional impairment is correlated with employment deprivation, and we see a similar trend when we compare PIP recipient rates to local authority educational deprivation scores, which are predominantly based on low educational attainment and attendance within the child population.
Moving forward to June 2025, and the total number of PIP recipients in the 16–17-year-old age group in England had more than doubled to over 83,000, increasing the overall recipient rate from around 2.7% in 2019 to 5.6%. Because recipient numbers for ADHD and autism grew faster than for most other conditions over that period, all other conditions made up a smaller share of the total by 2025, reducing from 54% of all recipients in June 2019 to 39% in June 2025.
Plotting June 2025 local authority recipient rates for those other conditions against employment deprivation scores indicates that the deprivation gradient persisted. Indeed, the relationship with deprivation strengthened, with employment deprivation in June 2025 explaining 53% of the variation in local authority PIP recipient rates in the 16–17-year-old group, up from 41% in June 2019.
PIP for ADHD and autism
Turning now to ADHD and autism, the picture looks a little different. As the chart below shows, in June 2019 a deprivation gradient was also visible, with higher rates of 16–17-year-olds receiving PIP for ADHD and autism in local authority areas with higher levels of employment- and education-related deprivation.
However, as the dispersion of blue and purple dots around the grey trend line indicates, there is greater uncertainty in the association, with employment deprivation only explaining around 22% of the variation in rates seen across authorities, compared to 41% for all other disabilities in the same month.
In 2019, PIP recipient rates for ADHD or autism were higher in more deprived areas, although the pattern was less clear
ChartSome authorities do sit almost exactly on the trend line – such as Hartlepool, which has one of the highest levels of deprivation and a PIP recipient rate of 2.1% in June 2019. At the other end of the trend line, Wokingham has one of the lowest levels of employment deprivation and a PIP recipient rate for autism and ADHD of 0.9%. However, others have much higher-than-expected recipient rates, such as the relatively affluent Warwick, whose 1.7% PIP recipient rate is more than twice as high as that of the significantly more deprived Bradford.
Comparing PIP recipient rates with educational deprivation shows a similar picture: areas with the lowest levels of educational attainment and attendance tend to have higher PIP recipient rates for ADHD and autism, but the relationship is not as strong as we see for disabilities other than autism and ADHD.
The changing picture on ADHD, autism and deprivation
Moving forward six years to examine the PIP recipient rate for ADHD and autism in June 2025, and the relationship with deprivation weakens further. The trend line still shows a moderate tendency for the PIP receipt rate to rise with deprivation. However, with each year between 2019 and 2025, the variation between individual local authorities becomes more dispersed. PIP rates increase across all points on the employment deprivation scale (from an all-England average of 1.3% in 2019 to 3.4% in 2025) but increase fastest in the least deprived areas, resulting in rates in those areas in 2025 exceeding those seen in the most deprived areas just six years earlier.
While deprivation explained around 22% of the local authority level variation in ADHD and autism PIP rates in June 2019, it explained only around 15% in June 2025 – the reverse of the dynamic we saw for conditions other than ADHD and autism.
This greater uncertainty in the relationship between deprivation and PIP recipient rates by 2025 is illustrated by the larger number of local authorities sitting further away from the line of best fit. Relatively affluent Runnymede in Surrey, for instance, had a recipient rate of 5.4% in June 2025, up from 1.7% in the same month in 2019 and more than one-and-a-half times the rate in the most deprived authority of Blackpool (at 3.3%), which was much closer to the level that the trend line indicates would have been expected of Runnymede.
While the very highest rates in 2025 still occurred in authorities with some of the highest levels of deprivation, including Knowsley and Liverpool, the lowest rate (1.35%) was found in the London borough of Brent, despite employment deprivation levels in the authority being higher than average. Brent's PIP recipient rate was less than half the rate found in the two local authorities with the lowest levels of employment deprivation: Hart in Hampshire and Wokingham in Berkshire, at 3% and 3.6% respectively, both of which saw more than four-fold increases in PIP recipient rates between 2019 and 2025.
By 2025, there had been a large rise in PIP recipients for ADHD or autism, with the rate in the least deprived areas increasing particularly significantly
ChartThe change in relationship between local authority deprivation score and PIP recipient rates for ADHD and autism is interesting, as it points to a potential weakening in the association between disability as experienced by the individual – and recognised and supported through PIP – and aggregated area-based indicators of unemployment and low educational attainment, which we might reasonably expect to be more strongly correlated with rates of disability-related impairment.
One explanation for this could be the cost-of-living crisis that hit the UK from the end of 2021. 38 By June 2025, PIP payments for 16–17-year-olds with ADHD and autism averaged at close to £8,800 over a full year. 39 It is conceivable that rising inflation may have contributed to an increase in claims from better-off areas, as cost-of-living pressures prompted more people to take up benefits they were entitled to. This explanation would frame increased recipient rates in the least deprived areas as a form of correction, bringing rates closer in line with underlying experienced disability.
Another explanation could be that our measures of employment and educational deprivation – taken from the English Indices of Deprivation 2025 – are not relevant to disability as experienced by 16–17-year-olds, and as static single measures (based on data from 2022/23 for employment, and 2018/19–2022/23 for education) do not adequately reflect year-on-year changes in unemployment or education outcomes within local areas.
These are important limitations to the analysis and available data. However, if these factors contribute to the apparent reduction in the association between deprivation and ADHD and autism PIP recipient rates, we have seen that they do not appear to have the same effect on recipient rates for conditions other than ADHD and autism.
Delving deeper: PIP claimant rates for ADHD and autism at neighbourhood level
To explore the relationship between PIP recipient rates for ADHD and autism and local area deprivation further, and to test the statistical significance of the observed faster growth of recipient rates in the least deprived local authorities, we looked at recipient rates across smaller geographical units, called Middle Super Output Areas (MSOAs). These are geographical units, or neighbourhoods, with average population sizes in the region of 8,000. In total, there are 6,856 MSOAs covering the whole of England. An advantage to using MSOAs as our unit of analysis is that they allow us to be more precise about deprivation levels than local authorities, where deprivation scores are averaged across large populations and geographical areas, meaning small pockets of deprivation (or affluence) can be obscured within larger areas.
However, the smaller populations of MSOAs create a complication: a significant proportion of PIP recipient data for 16–17-year-olds in 2019 for small areas is suppressed as counts fall below the reporting threshold of 5, to reduce the risk of identifying individual claimants. This skews the dataset because, as the 2019 deprivation gradient in PIP recipients would lead us to expect, the suppressed areas were significantly less deprived than the rest. To address this, we estimated the missing numbers using a method set out in the data notes, checking the results for sensitivity to alternative estimation methods.
The chart below groups the resulting estimates for 6,854 MSOAs 40 into five groups, or quintiles, according to their level of employment deprivation. Each group has an even number of MSOAs, but vary in the size of their 16–17 year-old populations, ranging from 266,365 (middle quintile 3) to 318,379 (the most deprived quintile, 5).
The chart allows the reader to view the estimated overall average rate at which 16–17-year-olds living within each quintile received PIP for ADHD or autism in each year, or to alternatively view the estimated rate for the median MSOA within each quintile. The charts vary a little between the two different measures because when we look at the overall average (or mean), MSOAs with larger 16–17-year-old population sizes have a greater influence in determining the rate for their quintile. By contrast, the median measure is based on the MSOA with the middle value (PIP rate) within each quintile, regardless of its population size.
Exploring PIP rates for ADHD or autism in small areas (MSOAs) shows growth everywhere but the sharpest in the least deprived areas
Chart
As the charts show, rates grew quickly across all deprivation quintiles between June 2019 and June 2024, slowing slightly in 2025. Looking at the median rate in the two most deprived quintiles (4 and 5), this growth took off from 2020, but in the least deprived quintile, the growth spurt for the median MSOA arrived later, after 2022.
In terms of overall average rates, the fastest growth was in the least deprived areas, where the estimated rate across MSOAs increased more than three-fold in quintiles 1 and 2, exceeding rates seen in the most deprived quintile in 2019 by 2024. By contrast, the slowest growth was in the most deprived quintile, where estimated rates increased 2.4 times over the six years, albeit from a significantly higher 2019 baseline.
The faster growth in the least deprived quintiles means that between 2019 and 2025 the relative impact of deprivation on PIP recipient rates for autism and ADHD reduced. Our MSOA estimates find that in 2019 the most deprived fifth of MSOAs had ADHD and autism PIP rates on average 2.5 times higher than the least deprived fifth of MSOAs. By June 2025 this deprivation multiplier – the relative gap between the least and most deprived MSOAs – had reduced to two times higher.
The chart below highlights the reduction in the deprivation multiplier between the most and least deprived MSOA quintiles, as the 2025 PIP rate in the least deprived quintile overtook the 2019 rate in the most deprived quintile.
Rates remain higher in the most deprived areas, but the least deprived areas have grown the most
ChartStatistical significance
We explored the MSOA-level data for PIP recipient counts in 2019 and 2025 using a negative binomial regression model, which allowed us to test the statistical significance of the faster change in PIP recipient rates in the least deprived area while allowing for the fact that areas vary in the size of their 16–17-year-old population. This testing confirmed that the faster increase in the least deprived quintile is statistically significant and holds up when alternative assumptions are used for estimating the suppressed figures.
While rates increased across both the most and the least deprived quintiles between June 2019 and June 2025, the increase was around 28% faster in the least deprived quintile. This means that the reduction in the deprivation multiplier shown above is statistically significant and should not be dismissed as random variation between 2019 and 2025.
To explore how other population factors might explain the variation in PIP recipient rates for ADHD and autism, we also included data on ethnicity from the 2021 Census. We found that having a higher proportion of MSOA residents recorded as being of White ethnicity was associated with higher PIP rates for ADHD and autism, in both 2019 and 2025. This effect was more pronounced in the most deprived areas, where a one-percentage-point higher proportion of White residents was associated with a 1.5% increase in the recipient rate for ADHD and autism. There was also an association in the least deprived fifth of MSOAs, but it was weaker, with an increase of around 0.6% in the PIP recipient rate for every percentage-point increase in the proportion of people with White ethnicity.
This association goes some way to explaining why, at local authority level, we saw much lower-than-expected levels of ADHD and autism PIP rates in areas such as Newham, Brent, Leicester, Bradford and Wolverhampton, as these areas have relatively low proportions of the population who identify as having a White ethnicity.
This finding may indicate that there are barriers to accessing PIP payments and/or ADHD and autism diagnoses for minoritised ethnic groups. It could also point to the influence of cultural factors in the experience and recognition of symptoms and impairments related to neurodiversity.
Spending implications
These changes in recipient rates have implications for how support for ADHD and autism through disability benefits has been distributed between areas with different levels of economic deprivation.
The average weekly personal independence payment for 16–17-year-olds with autism or ADHD in June 2025 was £169. Based on our analysis of MSOA recipient levels set out above, we estimate that in the region of £296m was spent on PIP payments for autism and ADHD to 16- and 17-year-olds over the course of 2025. Of this, around 30%, or around £87m, was paid to young people living in the 40% least deprived neighbourhoods, up from around £22m in 2019 (adjusted to 2025/26 prices) – a real terms increase of just under 300%.
These spending estimates take into account the fact that as PIP can only be claimed from the age of 16 onwards, the majority of 16–17-year-olds receiving the benefit for ADHD or autism as of June 2025 had been receiving it for less than a year.
For 18–19-year olds, the picture is very different, with over 90% of those receiving it for ADHD or autism in June 2025 having already received it for over a year, and over two thirds having received it for two years or more. Our analysis has focused on 16–17-year-olds, as this group is less geographically mobile than the 18–19-year-old group. However, if we assume the proportion of 18- and 19-year-olds receiving PIP for ADHD and autism who are linked to the least deprived 40% of neighbourhoods is similar to that seen in the 16–17 years age group, we estimate total PIP expenditure for ADHD and autism for 16–19-year-olds linked to the 40% least deprived areas to be around £222m in 2025, up from £64m in 2019 (2025/26 prices).
Addressing the prioritisation challenge
At the start of this analysis, we discussed how the concept of functional impairment has been raised as a potential tool for answering the policy challenge to distinguish higher-need cases of ADHD and autism from lower-need cases. In health and wider welfare policy, this challenge has arisen both out of a concern to appropriately prioritise resources, but also to avoid unduly medicalising people 41 on the basis of traits which, in the 16–24-year-old age group, are found in a quarter of the population. 42
In the absence of systematic data on functional impairment, we used PIP recipient data to explore how identified and supported disability compares with geographical employment and educational deprivation across England. Our analysis indicates that while PIP recipient rates for ADHD and autism – and therefore spending – remains highest in the most deprived areas, rates have grown fastest in the least deprived areas, shifting the distribution of support.
This is significant because it suggests a growing mismatch between how functional impairment (or disability) related to neurodiversity is recognised and supported in individuals and how it is reflected in population-level outcome measures of employment and educational attainment.
That mismatch has not arisen for disabilities other that autism and ADHD. There are a number of potential reasons why ADHD and autism might be different.
First, as the Timms review has stated, the PIP system itself is “not fit for purpose” with an assessment process that is inconsistent and particularly unsuitable for fluctuating and mental health-related conditions. 43 These inconsistencies might point to obstacles in claiming PIP that paradoxically make it easier for better functioning individuals – or those in a more supportive social context – to navigate the claims process, resulting in recipient rates that become increasingly less well matched to our aggregate population-level measures of educational and employment outcomes.
A second important consideration is the extent to which PIP recipient rates for ADHD and autism can be expected to at least partly reflect socio-economic inequalities in access to clinical diagnosis 44 , which is generally required to pursue a PIP claim. Research using GP records has noted higher rates of new ADHD diagnoses in the least deprived areas, which has been partly attributed to the greater use of private routes to diagnosis among more affluent groups, particularly since the pandemic. 45
These explanations are likely to be significant and should not be discounted. However, an additional explanation for the faster growth in ADHD and autism PIP recipients in the least deprived areas is that the diagnosis and recognition of these conditions may have become more detached from an objective assessment of functional impairment, with measurement involving subjective accounts and as such influenced by an individual's expectations.
This points to a further complication for the policy challenge to target support and intervention to where functional impairment is 'greatest”, as assessing impairment more objectively, without reference to outcome measures such as an individual's position in the job market, or to their achievements in education, may prove difficult.
For the disability benefit system, a shift towards such outcome measures would be in stark contrast to the original design of PIP, where the benefit is awarded regardless of an individual's economic means, work or educational achievement, and is intended to support disabled people to participate in society on a more equal footing with non-disabled people. For health care, where demand for support translates into a very substantial backlog in diagnostic waiting lists, 46 as well as pressures on prescribing and therapeutic support, a shift to prioritise patients on the basis of their employment status or qualification level would also be controversial, and would be perceived by some as a tacit or indirect means test for accessing health care.
At the heart of the challenge is a significant shift in how disability and health care need in relation to neurodiversity is understood in society, which the health care and benefit system is struggling to resource. Changes in the health and benefit system may help prioritise those in greatest need, but are unlikely to gain wide support unless accompanied by broader actions that make work, education and social environments more inclusive of those whose needs might not be deemed high enough to warrant clinical diagnosis and intervention, but who still experience challenges and forms of exclusion as a result of their neurodivergence.
Data notes
PIP data
All data on PIP recipient counts and financial awards is extracted from the Department of Work and Pensions' StatXplore database which report counts of 'cases with entitlement'- meaning successful claimants. Data is available for the 'main' disability or condition for which the entitlement is related. Additional disabilities or conditions which may be relevant to PIP awards are not identified in the data.
ADHD is referred to in the data as 'hyperkinetic disorder' and the figure we have used for autism is an aggregate of three forms of autism classified as 'autistic spectrum disorders”. In this analysis, we combine autism spectrum disorders and hyperkinetic disorder counts. This combining was in part pragmatic, as it resulted in fewer geographies falling below the DWP suppression threshold of 5. But it also reflects the lack of specificity in the DWP data regarding the presence of other diagnoses and disabilities. Autism and ADHD are increasingly seen as highly comorbid, with studies suggesting that up to 95% of adolescents with autism also have ADHD.
PIP is only payable to individuals aged 16 and over. Prior to reaching this age, eligible children and their families receive Disability Living Allowance, and are then contacted after their 16th birthday to be re-assessed for PIP. This means that at any given point in time, a number of 16- and 17- year olds receive DLA rather than PIP, as they are yet to be re-assessed for PIP. We checked the sensitivity of our LA-level results to the inclusion of DLA recipients for ADHD (as DLA data for autism is not available) and found minimal differences.
Deprivation data
Our deprivation measures are taken from the 2025 English Indices of Deprivation. This contains a single 'multiple' index of deprivation as well as separate deprivation indices (or 'domains'). To isolate domains most relevant to disability or functional impairment, we used the Employment Domain and Education Domain. A limitation in the Employment Domain indicator is that it is based on the percentage of 18 to 66 year olds in an area receiving unemployment benefit in 2022/2023, whereas our analysis focused on the 16-17 year old population. We therefore assume unemployment levels in the older age group broadly reflect those in the younger age group, as well as influence the deprivation context of the younger group.
To assist in the readability of the income deprivation score, we multiply the figures given in the IoD publication by 100. This does not affect the analysis.
The Education Domain is based on school achievement at key stages 2 and 4 (ages 11 and 16) alongside persistent absence and entry into higher education, for the period 2018/19 to 2022/23. A small proportion of the Education Domain is based on adult skills and English language proficiency in the 25 to 66 year old age group.
The education domain indicators are not available at MSOA level.
Population estimates
To calculate population rates, we used the mid year population estimates, published by the Office for National Statistics. At the time of the analysis, estimates for 2025 were not available. Each year’s rate is therefore calculated using the prior year’s estimate with no adjustments or assumptions made about growth. (For example, the rate for June 2019 is based on the mid year estimate for 2018).
Model for estimating suppressed values
To explore PIP recipient rates at MSOA level we developed a method for estimating recipient counts in MSOAs where the number receiving the benefit was below the DWP’s suppression threshold of 5, entailing the suppressed figure fell somewhere between zero and four. The relatively low level of recipients in 2019 meant that 68% of MSOAs had data suppressed that year, a figure which had reduced to 18% by 2025.
To estimate values for the suppressed MSOAs for June each year we first aggregated the reported MSOA recipient counts to their level of their host local authority and subtracted these totals from the reported local authority counts, which were not subject to data suppressions. The resulting residual (or gap) of just under 3,000 recipients in June 2019 was the equivalent of 18% of all claims for ADHD and autism in the 16-17 age group that month. Crudely divided equally between all suppressed MSOAs, this would result in a count for each MSOA of 0.6 recipients.
However our method for allocating residuals to suppressed MSOAs first gave each suppressed MSOA a weighting based on whether or not the MSOA had a reported recipient count (ie a count of 5 or above) in any of the two months either side of June in the relevant year. This was based on the logic that MSOAs which passed the suppression threshold in one or more of those months were more likely to have had a PIP recipient count in June at or closer to 4 than MSOAs that did not. We then randomly assigned each LA’s residuals to its suppressed MSOAs but allowed the weights to increase the chance of individual MSOAs being allocated a greater share of their LA’s residual (up to a maximum possible allocation of 4).
The weights meant that an MSOA with a reported figure in either May or July was given a 2.25 times greater chance of being allocated some or all of its LA’s residual than an MSOA with no reported figures between April and August, while those with a reported figure in April or August had a 1.75 times greater chance. These weights were additive, so the greater the number of months an MSOA passed the suppression threshold, the higher the chance the model gave it to receive a share of its LA’s residual. We repeated the weighted random allocation five times testing our regression results on each new set of MSOA estimates, finding minimal differences. We also ran a separate allocation method where MSOAs with a reported figure in any of the two months either side of June had twice the chance of being allocated residual recipients, which we also repeated with 5 separate randomisations. This alternative method also had minimal impact on the distribution of rates at quintile level.
A limitation in this method is that we were unable to fully adjust for inconsistencies in the data resulting from the DWP’s statistical disclosure controls, which include introducing small random errors to reduce the risk that individuals may be identified. These random errors result in a minority of LAs each year having a residual which is negative. Where this is the case, our method assumed that suppressed MSOAs within the LA had zero recipients in the relevant month but we did not adjust for the likely small level of over-reporting in the MSOAs that were not suppressed as doing so would require the introduction of additional assumptions.
We tested the sensitivity of estimation model by aggregating our estimated MSOA figures to LA level and comparing those figures to the reported LA level figures, which have no suppressions and a lower level of deliberate random errors. While this suggested our method over-estimated the average ADHD and autism PIP recipient rate for 16 to 17 year olds by 0.02 of a percentage point in 2019, and by 0.1 of a percentage point in 2025, these over-estimates were randomly distributed across our deprivation and ethnicity metrics. Overall, the method over-estimated the aggregate increase in the ADHD and autism PIP recipient rate between 2019 and 2025 by 0.07 of a percentage point, which was also randomly distributed across the deprivation and ethnicity metrics.
Suggested citation
Gainsbury S, Vassiliou S, Sherlaw-Johnson C (2026) PIP for ADHD and autism: a weakening link with deprivation? Briefing, Nuffield Trust
-
Updated 25/09/2026
-
Briefing updated to correct a sentence which implied that formal diagnosis is a pre-requisite to receiving PIP, whereas this is not the case.