‘Nowadays everyone has attention-deficit hyperactivity disorder (ADHD)’

This statement has become increasingly common, not only within the general public, but also among professionals. But, is there evidence that ADHD is over-diagnosed? Answering this question seems particularly timely, given the ongoing debate in the UK and internationally on the potential over-diagnosis and over-medicalisation of individuals referred for ADHD assessment. The importance of the topic is further highlighted by the work of the National Health Service (NHS) England ADHD Taskforce (https://www.england.nhs.uk/long-read/report-of-the-independent-adhd-taskforce-part-2/), conducted in response to a sharp rise in demand for ADHD assessments, treatment and support in recent years, which is placing immense pressure on existing services.

As academics, clinicians, carers and/or people with lived experience of ADHD, we suggest that the issue of possible over-diagnosis and/or over-medicalisation of ADHD is complex and that polarised views are unhelpful. Instead, we aim to present rigorous empirical evidence to illustrate the multifaceted nature of the issue.

There is no good evidence of ADHD being over-diagnosed in the UK

‘Over-diagnosis’ is observed when the prevalence of diagnoses made in clinical services, referred to as administrative prevalence (based on healthcare databases or insurance claims) exceeds prevalence estimates based on accurate assessments in representative population-based samples. Over-diagnosis may occur when diagnostic criteria are not applied with sufficient rigour, leading to false-positive cases. Over-diagnosis may also happen when people inappropriately self-diagnose. Notably, for individuals with milder or subclinical symptoms, a diagnosis can sometimes do more harm than good, creating stigma or leading to low-benefit treatments with significant side-effects.

When ADHD standardised diagnostic criteria are consistently applied, prevalence rates are generally consistent across countries, with geographical variations primarily reflecting differences in diagnostic practices rather than actual differences in prevalence. ^{Reference Sayal, Prasad, Daley, Ford and Coghill1} The most recent analysis of the international prevalence of ADHD diagnosed via structured interviews based on DSM (from III to 5) or ICD (9-10) criteria estimated the global prevalence to be approximately 5.4% in those under 18. ^{Reference Cortese, Song, Farhat, Yon, Lee and Kim2} In the UK, data from the NHS Digital 2017 Mental Health of Children and Young People survey of a representative sample of 2- to 19-year-olds in England estimated that 3.1% had DSM-5-defined ADHD, with similar figures in 1999 and 2004. Additionally, a meta-analysis concluded that 3.3% of the general adult population globally meet criteria for ADHD. ^{Reference Popit, Serod, Locatelli and Stuhec3} There are no data on adult ADHD prevalence estimated from standardised diagnostic assessments from the UK, but equally no reason to assume that international estimates are markedly different.

In contrast, English health service records show administrative prevalence doubled between 2000 and 2018 in males, quadrupled in females among children and increased 20-fold in men and 15-fold in women among adults. Yet in 2018, the administrative prevalence was 2.5% in boys and 0.7% in girls, and 0.7% in men and 0.2% in women, respectively. ^{Reference McKechnie, O’Nions, Dunsmuir and Petersen4} Current post-pandemic administrative prevalence data are unavailable and may differ from the aforementioned data.

While the administrative prevalence of ADHD has increased over time, available pre-pandemic data suggest that it has remained substantially below the ADHD population prevalence in the UK, providing no evidence at present that ADHD is over-diagnosed at a population level.

Finally, concerns exist that a sizeable portion of parents exaggerate ADHD symptoms to General Practitioners for perceived educational benefits. If true, the diagnosis would not be confirmed after thorough assessment in a substantial part of the cases. However, unpublished data from the South London and Maudsley NHS Trust – the UK’s largest child mental health service – shows 94% of children assessed received a confirmed ADHD diagnosis.

There is now more awareness of ADHD and its misdiagnosis may occur

Increased administrative prevalence of ADHD may reflect increased awareness of the condition. While there is no good evidence of over-diagnosis at a population level, it is certainly possible that some individuals are misdiagnosed.

A heightened sensitivity to ADHD diagnoses, while beneficial for identifying and supporting those with ADHD, may sometimes lead to over-interpretation of symptoms. Moreover, failure to consider and exclude differential diagnoses may result in some individuals being incorrectly diagnosed.

The tendency to attribute a broad range of attention-related or behavioural symptoms to ADHD may particularly occur in specific circumstances, for instance, in the education system, where an ADHD diagnosis can entitle students to extra time in exams, or for adults wishing to have special occupational adjustments made.

Key challenges may contribute to ADHD misdiagnosis. First, as with other mental disorders, there are no reliable diagnostic biomarkers. Notably, the QB Test (a computerised assessment of attention and activity) is recommended by the National Institute for Health and Care Excellence (NICE) as an optional tool to support diagnostic decision-making in children and adolescents – but not adults – rather than as a stand-alone test replacing clinical assessment (https://www.nice.org.uk/guidance/DG60).

Despite the absence of biomarkers, field trials showed that when raters were properly trained to diagnose ADHD according to standardised criteria, the reliability of ADHD diagnosis was among the highest of the disorders included in the DSM-5. ^{Reference Freedman, Lewis, Michels, Pine, Schultz and Tamminga5} Nonetheless, the diagnostic validity of ADHD in adults has been challenged due to the frequent reliance on self-reported symptoms and the lack of developmental information. Hence, a proper diagnosis of ADHD requires a careful and comprehensive assessment that includes independent corroboration of developmental history, symptoms and impairment, as well as consideration of other psychiatric and neurodevelopmental conditions.

Second, there is clear evidence that ADHD symptoms are distributed along a continuum – similar to many physiological traits, such as blood pressure and weight. However, as with hypertension or obesity, there are severity thresholds that determine ‘pathology’ and interventions based on stratified risk and potential functional impairment, even if individuals may not perceive any symptoms. For ADHD diagnosis, both DSM-5 and ICD-11 require that the symptoms must cause challenges to the person’s function. In this regard, there is a need for improved training for healthcare professionals on how to assess impairment. Importantly, subthreshold ADHD symptoms can still lead to significant challenges, just as the risk from raised blood pressure or blood glucose or being overweight does not suddenly appear at a single threshold value. Affected individuals may benefit from intervention and support using a needs-led approach.

Third, it is important to reflect on the increasing popularity of the concepts of neurodiversity and neurodivergence. These terms, which originated from a social justice movement, are instrumental in highlighting the rights of people with disabilities, but are not clinical diagnoses. Some studies apply these terms to autism alone, others to both autism and ADHD, and some broaden their use to include other diagnoses. Therefore, when such terminology is used, the conflation of ADHD with other conditions under the umbrella term of neurodiversity/neurodivergence could preclude accurate estimation of the prevalence of ADHD.

In addition to misdiagnosis, there has been a tendency over time to recognise more ‘mild’ ADHD cases. A Swedish study ^{Reference Li, Coghill, Sjölander, Yao, Zhang and Kuja-Halkola6} found that the magnitude of the protective effect of ADHD medication on the risk of several serious real-world outcomes such as traffic accidents and crime has decreased over time from 2006 to 2020. As such, more research is needed on risk-stratification in ADHD to further clarify the relative benefits and risks of intervention at different levels of severity/impairment.

Clinical services in the UK are not adequately providing the right support to those individuals with ADHD who need it

While the current focus among the general public and media tends to be on the alleged over-diagnosis of ADHD, we believe that attention should also be focused on the unmet needs of individuals with ADHD. Indeed, the gap between the expected population prevalence and administrative prevalence of ADHD in England likely reflects a failure to provide appropriate treatment. Internationally, a meta-analysis ^{Reference Massuti, Moreira-Maia, Campani, Sônego, Amaro and Akutagava-Martins7} found that only one in five school-age children and adolescents with ADHD were treated with ADHD medications. Even in the USA – where prescription rates in some states exceed the expected prevalence – the overall picture across many states may still suggest undertreatment. Meta-analytic evidence showed that, for every individual prescribed medication without a formal ADHD diagnosis, there were three diagnosed individuals who might have benefited from medication but did not receive it. ^{Reference Massuti, Moreira-Maia, Campani, Sônego, Amaro and Akutagava-Martins7}

Effective treatments for ADHD are available, generally well-tolerated and supported by strong evidence. Stimulants have the highest effect sizes in psychiatry for managing the core symptoms of the condition, as demonstrated in short-term trials. ^{Reference Daley, Van Der Oord, Ferrin, Cortese, Danckaerts and Doepfner8} Although longer-term trials are often unfeasible, discontinuation trials have shown the persistence of long-term effects (references in the online supplementary material available at https://doi.org/10.1192/bjp.2026.10546). Non-pharmacological treatments, such as behavioural parent training, have demonstrated efficacy for non-core symptoms often associated with ADHD, such as oppositional behaviour and parenting challenges. ^{Reference Daley, Van Der Oord, Ferrin, Cortese, Danckaerts and Doepfner8}

We sometimes hear from colleagues that ADHD medications are not effective in adults. But evidence shows the opposite. A network meta-analysis ^{Reference Ostinelli, Schulze, Zangani, Farhat, Tomlinson and Del Giovane9} of randomised controlled trials of both pharmacological and non-pharmacological treatments for ADHD found that medications were more efficacious than a placebo, even though the effect sizes were somewhat lower compared with those in children.

However, UK ADHD services are overstretched and unable to meet the growing demand. An online survey of 7340 people, conducted by the House of Commons Petitions Committee between 2021 and 2022, highlighted lengthy wait times for NHS ADHD assessments. Among those children and young people who were eventually diagnosed with ADHD, approximately 14% reported waiting 2 to 3 years for an assessment, while 27% waited 1 to 2 years (https://committees.parliament.uk/writtenevidence/117330/default/). Notably, recognition of ADHD in adults who were previously undiagnosed in childhood has led to a new demand on adult mental health services, which typically have inadequate capacity in terms of both resources and training, leading to growing waiting lists and driving people who can pay to the private sector. Thus, additional funding and training both for specialists and General Practitioners to improve the detection and assessment of ADHD while reducing misdiagnosis rates are crucial.

When it comes to discussing treatment options with people with ADHD, clinicians often focus on the effects (adverse and desired) of medication, but the cost of untreated ADHD is often overlooked. Unfortunately, ADHD is associated with an increased risk of serious negative outcomes, including academic failure, suicidal behaviours, substance misuse, accidental injuries, transport accidents, criminality and mortality. Importantly, recent evidence using rigorous quasi-experimental designs (e.g. emulation of target trial) has shown that stimulants reduce these risks in real-world settings (references in the online supplementary material).

Consistent with the report of the recent NHS England ADHD Task Force, the failure to provide treatments with significant impact on these important outcomes represents a major ethical issue that needs to be urgently addressed, with the crucial input of people with lived experience.

From a service organisation and lived experience standpoint, the currently implemented clinical pathways in the UK are suboptimal. While they help focus resources and streamline attention on ADHD, the downside of this siloed system is that those with co-occurring conditions, which are often encountered in the real world, have to navigate multiple pathways. Furthermore, the isolated approach of ADHD-only services may increase the risk of misdiagnosis. Since some ADHD symptoms can be non-specific, clinicians who have not been well-trained in differential diagnosis may over-identify ADHD or fail to recognise important co-occurring conditions, especially if they work in generic services.

Instead of the usual approach to ADHD diagnosis and treatment as an all-or nothing categorical condition with a single treatment (i.e. medication), a stratified stepped-care approach (based on severity) may be more useful to place ADHD within a wider healthcare and preventative context. Similar approaches are successfully applied to anxiety, hypertension, hyperglycaemia and hypercholesterolaemia. This goal of stratification would be facilitated by improving operational definitions of ADHD severity levels. It will be important to coordinate these changes with educational advice for schools, which currently (in England at least) have made the diagnosis a requirement for accessing appropriate educational adjustments; schools have no legal duty to provide them outside an Education Health and Care Plan or if disability is not considered present.

In closing, there is no published evidence that ADHD is over-diagnosed in the UK, although it certainly can be misdiagnosed in some individuals. Importantly, evidence points to unmet need, as many people with ADHD are not diagnosed. Overall, we believe that the narrative of over-diagnosis of ADHD ought not be used as a pretext to deny individuals with properly-diagnosed ADHD the support that they deserve.