People feel defensive about any suggestion that their screen use might be problematic. But we use a lot of screens. And we use those screens a lot.

According to our last census, 94% of New Zealand households have access to a mobile phone. And at close to six million, our number of mobile phone connections well exceeds our population. On top of that there’s also our other screens — our tablets, our smart watches, gaming consoles, televisions and laptops.

Screens have crept so intimately into our lives that it’s hard to imagine life without them. They entertain and inform. They connect us to people we love. They bring us stories about our world. They simplify life admin (and sometimes, complicate it). But that’s not all that they do.

We seem to develop relationships with our screens, especially our phones - these small dense slabs that ping and vibrate and flash. And just like in any relationship, the time we spend with screens changes us. Including, it seems, our attention.

On average, New Zealanders spend over three hours a day on their mobile phone. A 2025 survey found that 78% of New Zealand adults check their mobile phone every hour - some of us up to 50 times each of those hours. Global estimates suggest that smartphone owners interact with their phones between 85 and 204 times a day, including as soon as they wake up, just before they go to sleep, and even in the middle of the night. We are very involved with our phones.

This entanglement with mobile phones isn’t limited to adults. Phone ownership has grown amongst adolescents and children too. On Census at School, over 70% of New Zealand school children report that they own mobile phones. Similarly, in OECD countries 70% of 10 year olds own a smartphone and at least half of 15-year-olds spend 30 or more hours per week on digital devices.

Phone presence or phone use?

Clearly we are spending enormous amounts of our time using phones. Do they have a noticeable impact on us?

Research suggests the impact of phones is significant, including on our attention. For example, the mere presence of smartphones has a negative effect on our cognitive performance. In 2023 Jeanette Skowronek and her colleagues found that having a mobile phone turned off but visible on a desk resulted in poorer attention, and poorer performance on cognitive tasks. Similarly, in 2018, Jessica Mendoza and her colleagues found that university students who kept their mobile phones with them during a lecture had worse recall of the lecture content than students who did not have their mobile phones with them.

The nuances of the effects of a mobile phone’s presence are still being investigated. But still, a 2023 review concluded that overall there is a significant negative effect of smartphone presence on our cognitive performance, including attention.

What then, are the effects of using our smartphones?

Research about phone use is complicated. Phones are used for so many activities: to make video calls to family; to play games alone; to access information; to watch entertainment; to gamble; to scroll. This heterogeneity complicates research about the effects of phones, as do many other factors. Does it make a difference whether you’re doing a simple or a complex task? Should we rely on self report, parent report or objective measures? Does it make a difference how attached you are to your phone? How do you account for the use of multiple devices at the same time? Does it make a difference how old you are?

Nevertheless, as reviewed here, here and here, the results of meta analyses are converging: more screen time and more phone ‘pick-ups’ (checking our phones) are linked to multiple poor outcomes including: reduced language competence, increased anxiety and depression, reduced sleep, poorer academic performance, higher risk of obesity, and increased impulsivity and inattention.

Here at The ADHD Files, we’re especially interested in the last of these - impulsivity and inattention - considered core ADHD behaviours. What does the research tell us about any relationship between screen use and difficulties with attention and impulse control?

Subscribe now

Phones and attention

In a field where things are changing so much and so quickly it’s useful to have current summaries. This year Claire Reid and her colleagues from the University of Auckland published a review integrating recent evidence about the impact of screen use on children’s executive functions. Executive functions are higher-level thinking skills that enable people to focus attention, control impulses, plan, remember instructions, and manage emotions. You’ll recognise some of these as core criteria for a diagnosis of ADHD.

The studies included in the Auckland review used surveys, behavioural tasks and neuro-imaging. Overall, 90% of the studies found at least one negative relationship between screen use and executive function. Most (81%) of the negative studies were longitudinal, which strengthens the conclusions of the 9% that were cross-sectional. Both types of study found consistent negative associations between screen use and executive function from infancy through to 18 years. The authors highlighted the strength of the longitudinal research and the bi-directionality of the associations.

Longitudinal evidence

The Auckland review noted that earlier research focussed on cross-sectional studies, which examined how current phone use correlates with current attention difficulties. The research about the impact of a phone’s presence is an example. The Auckland review brings together results from many longitudinal studies, and found that they give consistent evidence of a longitudinal negative relationship between higher screen use and reduced executive function. What this means is that phones affect our immediate experience of attention, but also have a cumulative effect over time.

Bi-directional evidence

The other key finding of the Auckland review was bi-directionality. Some studies showed that earlier screen use impaired later executive function; some studies found that earlier executive function difficulties predicted later higher screen use; and some studies found evidence of both of these. This means that impulsivity and attention difficulties increase the risk of developing problematic screen use which in turn can lead to greater executive function difficulties over time. And high levels of screen use as a child predict poorer executive functions later on. Bi-directionality may lead to a vicious cycle of worsening outcomes. As the authors say,

given that impulsivity and attention problems both predict increased later screen use and are themselves predicted by earlier screen use, these findings [are] of particular significance for children with ADHD and ASD who already suffer from deficits in these areas. This places them at heightened risk for escalating screen engagement and its associated cognitive and behavioural consequences, potentially reinforcing a negative developmental trajectory.

Where does that leave us?

Overall, the research seems both clear and inconvenient.

Humans possess a finite capacity for sustained attention and can only maintain focus on a specific task for a limited period of time. Our capacity varies depending on the nature of the task and on individual drivers like interest, motivation and personal experience.

The presence of phones, the time we spend on phones, and how we interact with our phones can impair our attention. Experiencing attention difficulties makes it more likely that we will spend more time with our phones.

ADHD diagnoses have multiplied over the same 20 years since mobile phones have become available and more and more popular. Given the lack of identified biological causes of ADHD, and the evidence that for most people attention difficulties come and go, it’s likely that environmental factors play a role in the development of ADHD-type behaviours - which in turn suggests that environmental changes might help reduce those behaviours.

Smartphones are convenient. Their Swiss-army-knife adaptability and portability make them a desirable and useful tool that provides social, informational, and leisure benefits from just about anywhere. It’s difficult to imagine our lives without them. And more and more, it seems that phones are becoming an essential key for unlocking the administrative doors of modern life.

But this newer research does signal a potential course of action to remedy attention difficulties, and to help us prevent them in the first place: reducing the time spent with screens and smartphones, and being thoughtful about how we use and interact with them.

Organisations have developed guidelines about how to protect our executive functions in a world of screens. You can read overseas examples published by the World Health Organization and the American Academy of Pediatrics. Local examples include advice on protecting attention and impulse control from Keep It Real Online, the Australian 24-hour movement guidelines and The Paediatric Society of New Zealand.

Some recommendations are based specifically on research about attention. Some recommendations are informed by what we know about child development. Other recommendations come from research about other factors that affect attention, such as sleep. But themes emerge in the guideline advice.

The recommendations depend on how old we are, and what we’re doing on our phone. For the youngest among us, if we’re younger than two years old, we shouldn’t be using a phone on our own. Our adults should be co-viewing and helping us understand what we are seeing. Children learn by watching others, especially their parents, so our adults should also be thoughtful about what phone-using behaviour they model to us. Do we see adults prioritising the demanding pings of a phone or the in-real-life person in front of them?

As adults, if we’re concerned that our attention capacity has diminished, we can adjust how we’re using screens. We can set aside time to spend on phones, reduce notifications so that we are less interrupted, and stop using phones during the night.

For all of us, both children and adults, an important question is: What are we not doing, while we’re on our phones? For that average New Zealander, spending three hours a day on our phone, what would we choose to do with an extra 21 hours a week?

Thanks for reading. Now turn this off!

1. Skowronek J, Seifert A & Lindberg S. (2023). The mere presence of a smartphone reduces basal attentional performance. Scientific Reports, 13, 9363. doi.org/10.1038/s41598-023-36256-4

2. Mendoza JS, Pody BC, Lee S, et al. (2018). The effect of cellphones on attention and learning: The influences of time, distraction, and nomophobia. Computers in Human Behavior, 86, 52-60. doi.org/10.1016/j.chb.2018.04.027.

3. Böttger T, Poschik M & Zierer, K. (2023). Does the Brain Drain Effect Really Exist? A Meta-Analysis. Behavioural Science, 13, 751. doi.org/ 10.3390/bs13090751.

4. Reid CA, Dhamankar, Marsh MS, et al. (2026). Have we found directionality? Longitudinal associations between screen use and executive function in children. A scoping review. Developmental Review, 80, 101266. doi.org/10.1016/j.dr.2026.101266.

Despite my appreciation for the work and care put into the Four Corners programme, there’s one more issue I wish they had tackled: what is ADHD?

For starters, the programme follows many people in calling ADHD a “condition” — a word which implies a medical disorder arising from a known biological abnormality. So far we don’t have evidence that this is true.

The programme also refers to behaviours that are not part of ADHD as if they were. For example, the episode includes a person explaining his experience like this:

Having ADHD is a bunch of different things to a bunch of different people but for me emotional regulation is always something that I’ve struggled with, especially in response to feelings of rejection, which all makes sense more now, feelings of being left out or othered by people.

This person may struggle with intense emotions and feelings of exclusion. These experiences can be painful and difficult. But they are not proof of his ADHD or part of his ADHD, because they are not criteria for ADHD.

Later in the programme other behaviours such as “addiction”, and “struggling with” caring for children are presented as evidence of other people’s ADHD. These mis-inclusions of behaviours outside ADHD as being part of ADHD are common.

ADHD: Not Just Hyper screened on TVNZ in 2024. It also repeatedly quoted as proof of ADHD behaviours that are not criteria for ADHD. Over the course of the show, people stated that their ADHD included hyperfocus, hypersensitivity, sleeping problems, not being able to turn off their brain, lack of feeling normal, living in a state of fight or flight, being loud, tiring easily, being anxious, and being addicted to television. Not one of these is in the list of ADHD criteria. They can be difficult to experience, and distressing to the interviewees’ experiences, but they are still not part of or evidence of ADHD.

Newspapers

Text-based media often quote similar statements as though they were true. This Newsroom article indicates that rejection sensitivity is a characteristic of ADHD; this RNZ article links social anxiety to ADHD; and this Otago Daily Times article refers to hyperfocus as a feature of ADHD. The interviewees were struggling and were generous to share their experiences with journalists. But while their suffering and distress is real, it is misleading to imply that these experiences are evidence of the person’s ADHD — because none of these behaviours are criteria for ADHD.

TikTok

These days people get more and more information from Youtube, Instagram, and TikTok.

Videos about ADHD are very popular on Tiktok. By 2026, the #ADHD hashtag had collected over 5.2 million posts and 2.5 billion views. How reliable are these posts as sources of information about ADHD?

In 2022 Anthony Yeung and his colleagues studied the 100 most popular videos about ADHD. At that time those videos had collectively been viewed 283,459,400 times. But it turned out over half of them were misleading. Descriptions of ADHD included behaviours that aren’t part of any mental health diagnosis, such as ‘random noise making’ or ‘being competitive’. Misleading videos also described anxiety, depression, anger, relationship conflicts, dissociation, and mood swings as symptoms of ADHD. (Not one of them is).

More recently, Vasileia Karasavva and her colleagues evaluated TikTok videos about ADHD. Like Yeung, they found that over half of all claims about ADHD didn’t align with the formal diagnostic criteria for ADHD. Eighteen percent of claims better illustrated a different disorder such as depression, anxiety, or eating disorders; 42% were transdiagnostic symptoms that could reflect multiple disorders; and 69% better reflected normal human experience. The authors noted that 50% of the content creators used the videos to promote products they were selling.

Studies find that the majority of ADHD related content on Tiktok is unreliable. Given their popularity, they may be playing a key role in spreading misinformation about the definition of ADHD. But that’s just TikTok. Should we expect better information from educational material produced specifically for the public by health, mental health, or ADHD organisations? Regretfully, they too include unrelated symptoms in their lists of ADHD criteria.

Formal Information

Mytre van Langen and her colleagues analysed 41 written psychoeducational resources from the UK, Netherlands, Hungary and the USA, such as webpages, pdf-files with psychoeducational text, and downloadable flyers. They were produced by health organisations such as the National Health Service (UK); mental health organisations such as the Royal College of Psychiatrists (UK), Hersenstichting (Netherlands) and the US National Institute of Mental Health; and advocacy groups such as Children and Adults with ADHD (CHADD).

These materials’ initial definitions of ADHD mostly adhered to the key DSM-5 criteria of inattention, hyperactivity and impulsivity.

However, in subsequent descriptions of ADHD, materials included such a wide variety of symptoms, that this description broadened and blurred the definition.

Between them, the resources named a broad range of additional behaviours as indicators of ADHD: being hardworking, clumsiness, creativity, lack of social skills, learning difficulties, intelligence, tics, being determined, being good at problem-solving, anxiety, language problems, mood swings, and poor self-esteem. (Again, none of them are.)

So how do we know which behaviours are part of ADHD? Two systems are used to classify illnesses and disorders: the ICD and the DSM.

Share

ADHD Criteria

The International Classification of Diseases (ICD) is published by the World Health Organisation (WHO). Its 11th edition came into effect in 2022 and includes all illnesses and diseases. Chapter 6 is dedicated to Mental, behavioural or neurodevelopmental disorders, which includes ADHD. The Diagnostic and Statistical Manual of Mental Disorders (DSM) is published by the American Psychiatric Association. Its 5th edition came into effect in 2013. It focuses only on the experiences and difficulties we call mental illnesses, including ADHD. ICD-11 is free to download from the WHO website. DSM-5 costs NZ$286.

Although the ICD and DSM definitions aren’t exactly the same, their descriptions of the core ADHD criteria are very similar.

On their own, the presence of these behaviours doesn’t mean that a person has ADHD. A formal ADHD diagnosis also assesses when the behaviours first appeared, how long they’ve been present, where they show up, and (very importantly) whether they cause serious impairment to the person’s functioning.

None of the many other behaviours commonly attributed to ADHD are in the list of criteria. The formal criteria don’t include hyperfocus, emotion dysregulation, rejection sensitivity, hypersensitivity, sleeping problems, not being able to turn off their brain, lack of feeling normal, living in a state of fight or flight, being loud, tiring easily, being anxious, or being addicted to television.

Where does that leave us?

Publicly available information routinely misleads us about the core symptoms of ADHD. Many problematic behaviours are mis-attributed to ADHD in documentaries, in traditional journalism and on social media. These difficult experiences matter. It makes sense that people want to understand them and find out what will help. But they aren’t evidence of ADHD, and repeating that they are is confusing.

This expansion of criteria can be described as “horizontal”. It creeps sideways, pulling more people into its home through the addition of more and more behaviours to be considered part of the disorder. In her book The Age of Diagnosis, Suzanne O’Sullivan has documented another expansion also happening in ADHD: “vertical expansion”. Within each criterion of ADHD, the boundary between what is considered normal and what is considered abnormal shifts, thus gathering more people into the disorder. Vertical expansion is happening within ADHD as well as other diagnoses such as Autism, hypermobile Ehlers-Danlos Syndrome and depression. In all of them, behaviours and experiences that were once considered part of normal life have been reclassified as evidence of disorder.

In this post, I haven’t gone as far as discussing the critique by authors who point out that we still lack evidence that a “condition” called ADHD exists. Sami Timimi and Yaakov Ophir are good examples. But accepting the medical model on its own terms, people who find a medical understanding of their behaviour useful need honest information when playing by the medical model’s rules.

1. Yeung A, Ng E & Abi-Jaoude E. (2022). TikTok and ADHD: A Cross-Sectional Study of Social Media Content Quality. Canadian Journal of Psychiatry, 67, 899-906. doi: 10.1177/07067437221082854.

2. Karasavva V, Miller C, Groves N, et al. (2025). A double-edged hashtag: Evaluation of #ADHD- related TikTok content and its associations with perceptions of ADHD. PLoS ONE 20: e0319335.

3. van Langen MJM, Szőke R, Rijkelijkhuizen DNJ, et al. (2022). Lost in explanation: internal conflicts in the discourse of ADHD psychoeducation. BMC Psychiatry 22, 690. doi.org/10.1186/s12888-022-04327-x

Expansions to ADHD diagnosis in Aotearoa began on the 1st of February, so trained GPs and Nurse Practitioners can now diagnose and prescribe drugs for ADHD. The change generated intensive media coverage on television, on radio, and in newspapers. Although most of the analysis championed the change, more thoughtful views were included on Radio New Zealand’s The Panel, The Spinoff, and Stuff.

But diagnostic changes weren’t the only ADHD issue being discussed in February. Also reported was the possibility that ADHD diagnosis could itself be harmful.

Possible harm from ADHD diagnosis

Radio New Zealand (RNZ) published this 5th of February article about the potential risks to children of ADHD diagnosis. The article was picked up by the Otago Daily Times and some online ‘news’ sites, and was discussed on Reddit and LinkedIn.

The two studies referred to in the RNZ article were ones I reviewed last year, which highlight risks of harm from ADHD diagnosis. In short, a study from Australia by Luise Kazda1 and a second study from Ireland by Cliodhna O’Connor2 compared young people with ADHD behaviours who had been diagnosed with ADHD, with matched young people with ADHD behaviours who had not been diagnosed with ADHD. By ‘ADHD behaviours’ they meant inattentive, hyperactive and impulsive behaviours often described as “core ADHD symptoms”.

At the start of the studies, all the children’s core ADHD behaviours were rated the same across both groups. But when they were reviewed four to eight years later, the children who had been diagnosed did worse than the children who had not been diagnosed, even though they had all displayed the same core behaviours at baseline. Compared to adolescents without a diagnosis, diagnosed adolescents felt less accepted and included at school, said they did worse academically, displayed more negative social behaviours, and were more likely to harm themselves. Negative associations with diagnosis were largest in adolescents with mild ADHD behaviours, and in adolescents who had been diagnosed youngest (at 6-7 years old). There were no differences between the groups on general health, happiness and peer trust.

Divided opinion

I think the results signal that we shouldn’t rush to diagnose ADHD in children and adolescents. But some readers were indignant about the RNZ article. They condemned the studies as “bad science”. They complained that the article cited “two small not great quality studies and one opinion” (which I guess is me). And they argued that “to compare children with ‘similar behaviour’ and no diagnosis to children with a diagnosis (with the assumption they are all adhd) is problematic to say the least”.

Are these concerns valid enough that we should ignore the findings? Or are the researchers’ results robust? Let’s consider two issues: research quality, and treatment versus diagnosis.

Research quality

Are these studies “bad science”? In short, no.

Although the studies are constrained by the limits of observational research, this doesn’t make them ‘bad science’. Like all research studies we need to understand their limitations and how they fit in with existing knowledge. The authors also discussed the limits of retrospectively analysing longitudinal studies that relied on standardised questionnaires.

Qualitative and quantitative research complement each other: they answer different questions and have different limitations. Qualitative research about people’s subjective experiences of diagnosis is important but it doesn’t help us understand more objectively its impact on children. It can’t isolate the extent to which positive or negative effects follow from the diagnostic label itself, rather than from the behaviours that precede it. The quantitative approach used in these studies allows an empirical evaluation of the effects of diagnosis.

In both studies, the researchers were careful, cautious and transparent. Careful about how they analysed the available data. Cautious about the conclusions they reached. Transparent about their sources.

In both studies the researchers had expected the diagnosed adolescents to do better - to benefit from a diagnosis of ADHD and consequent intervention, resulting in better quality of life later. But they still published unexpected results, rare in science where null results see out their days in lonely filing cabinets and published reports skew positive.

Conflating ‘treatment’ and ‘medication’

‘Treatment’ and ‘medication’ need to be carefully distinguished in conversations about ADHD. Despite reassurances that treatment refers to more than medication, the word treatment is still routinely used to mean only medication.

The confusing melding of ‘treatment’ and ‘medication’ was prevalent in the February coverage about ADHD. For example, articles referred to people’s desperate wait for diagnosis ‘so that they could access treatment’. Although ADHD diagnosis is required for a prescription for ADHD drugs, it is not required for the non-drug interventions that can help people struggling with ADHD behaviours. In a statement about a lack of diagnosis restricting ‘treatment’, treatment only means medication, because all other ADHD interventions can be accessed without a diagnosis - behavioural strategies, classroom support, green space, parent assistance, and therapy to help people address other causes of their attention issues.

Share

Is it possible that ADHD diagnosis causes harm?

Online detractors were angry about the RNZ story. They disagreed with the studies’ findings and seemed to find the study question itself affronting. They assumed that ADHD diagnosis could not be harmful. That by definition, diagnosis is innately good, and articles investigating it can be dismissed out of hand.

However there is clear and wide-ranging evidence that diagnosis can be harmful. Suzanne O’Sullivan has documented extensively in her book, The Age of Diagnosis, the increasing diagnostic harm observed for both mental and physical illnesses. She reports negative outcomes from diagnosis in many areas including Lyme Disease, autism, cancer, and ADHD. And just as the Australian and Irish ADHD studies found, Sullivan reported greater risk of harm for people with milder symptoms.

The Australian and Irish studies appear to be the only two which have examined the impact of ADHD diagnosis itself on children. Why is the research in this area so limited, when thousands of papers are published about the efficacy of ADHD drugs? Perhaps few studies explore this topic because the benefits of diagnosis are so broadly assumed to be agreed upon. Or perhaps designing studies to answer this question is very complicated, and the research difficult to do.

It is possible that ADHD diagnoses come with risks of harm. Is it plausible?

Is it plausible that ADHD diagnosis causes harm?

Kazda and O’Connor studied the effect of ADHD diagnosis in children directly, but other related research supports their findings indirectly. Some relevant research is ADHD specific, but research about other mental disorders can also inform our understanding about the impact of ADHD diagnosis. For example,

• A 2006 cross-sectional study by Eli Lilly compared adults previously diagnosed with ADHD with peers who screened positive for ADHD but had never received a diagnosis. There were no significant differences on disability or quality of life measures between undiagnosed people and people diagnosed with ADHD. That is, adults with a diagnosis, which is presumed to be helpful, had not done better on any measures - further evidence that clinical identification of ADHD does not necessarily improve outcomes and that diagnosis may not be helpful for everyone.

• In other papers, Kazda and O’Connor both reviewed the research about people’s experiences of ADHD diagnosis. Both reported a fundamental ambivalence. While some people felt a sense of empowerment and legitimacy, understanding and acceptance, and decreased blame and anger, diagnosis also lead to disempowerment, decreased responsibility, inaction, greater feelings of isolation, shame and stigmatisation. Further research needs to disentangle how these positive and negative experiences co-exist. Still, these qualitative findings might provide clues about how diagnosis harms children’s outcomes.

• In 2023 Lexi Harari and her colleagues empirically studied the effects of mental illness self-labels on self-esteem. 427 young adolescents were followed for two years, tracking when they adopted and dropped self-labels, and the improvement and deterioration in their self esteem. Although writers argue that diagnosis and labels lead to enhanced mental health, in this study youth experienced a decrease in self-esteem after adopting a self-label, and experienced increased self-esteem after discarding a self-label. The authors conclude that labelling is harmful, even when self-selected, and that, “shedding them can yield positive effects.”

Each of these studies aligns with Kazda and O’Connor’s key finding: that ADHD diagnosis could be harmful for children and adolescents. Diagnosis is not neutral. It seems plausible that being told there is something wrong with you could detrimentally affect a child’s emotional and social wellbeing. A growing body of literature has suggested numerous ways that a young person’s life could be negatively affected by an ADHD diagnosis, or in fact any mental health diagnosis. Furthermore no research has shown that children who exhibit ADHD behaviours and are diagnosed with ADHD do better in the longterm than children who exhibit the same behaviours but are not labelled with an ADHD diagnosis.

Where does this leave us?

The Australian and Irish studies found that all else being equal, children with diagnosed ADHD fare worse than their peers with similar behaviour but without diagnosis. These studies question the long-term effectiveness and safety of diagnosing children and adolescents with ADHD, especially for younger children and those with milder ADHD behaviours. Parents considering an ADHD diagnosis for their child need to be informed about these risks.

To me, these studies were robustly executed, RNZ did well to expand awareness about them, and we need to pay attention to their findings.

Diagnosis is not the help that people need. We need to replace the prevailing medical model’s diagnosis with psychological formulation models that conceptualise distress as meaningful response to adversity or dimensional frameworks that conceptualise behaviours on continuums. To adopt these for ADHD this would mean understanding each child’s struggles carefully, in context, in order to plan how to best support them.

Ritalin is by far the most commonly dispensed drug for ADHD in Aotearoa. Also known by its chemical name methylphenidate, it accounts for over 90% of ADHD prescriptions.

The massive prescribing increase we have already seen for Ritalin and other ADHD drugs will soon escalate: from the 1^st of February 2026, GPs and Nurse Practitioners are allowed to diagnose ADHD, and to prescribe drugs to treat it.

Share

Training for change

We are assured that these new prescribers will first be trained. But no GP training has yet covered critical topics such as the risk of a diagnosis itself causing harm, or the research that shows poorer outcomes for those using ADHD drugs. Or most fundamentally, the absence of a biological basis for what-we-call-ADHD.

Good medical practice depends on careful, unbiased information. But alarmingly, the first ADHD training available for GPs after the February changes are introduced is funded by Takeda - a drug company selling ADHD drugs in Aotearoa. Industry funding notoriously affects the scope, outcomes and reporting of research. And scientific reviews consistently find that almost all published ADHD studies are at high risk of bias. GPs deserve better. We all deserve better.

Training gaps

What might good GP training about ADHD include?

New independent research reveals that children diagnosed with ADHD do worse than children with the same behaviours but not diagnosed with ADHD. On all outcomes.

Old long-term research shows that children diagnosed with ADHD and prescribed ADHD drugs do worse in the long-term than children diagnosed with ADHD and not using ADHD drugs. On all outcomes.

Persistent research finds that the youngest child in a class is up to 70% more likely to be diagnosed with ADHD than the oldest child in the class.

Research from our very own Dunedin Study found that what-we-call-ADHD rarely lasts.

And most compelling of all, fifty years of research has failed to establish a biological cause for what-we-call-ADHD.

These are just some examples of key research currently missing from GP training about ADHD.

In July 2024, I highlighted some of these omissions to trainers, but 18 months later, ADHD training still has the same gaps. As a result the parents, adults and children attending these single-diagnosis assessments are unlikely to get all the information they need to make informed decisions before being labelled with ADHD.

Where does this leave us?

The expanded access to ADHD diagnosis will channel more people into a medical model that offers only medical responses. This approach leaves no room for questions such as “Why am I struggling?” and “What might help?” And it leaves no room for non-medical responses to difficulties with attention, activity levels or impulsivity.

ADHD diagnoses can hide problems such as immaturity, depression, hunger, post-traumatic stress, learning difficulties, screen-time, excessive anxiety, mismatched teaching approaches, poor hearing, poor eyesight, substance abuse and sleep deprivation. Thorough assessments are needed to identify these other causes of inattention, hyperactivity and impulsivity so that appropriate help can be offered.

The Minister for Mental Health and his advisors are so preoccupied with increasing access to ADHD diagnosis and drugs that they are overlooking the need to consider difficulties broadly, and in context, and to provide access to a range of ways to help. Practical options include behavioural strategies, classroom support, green space, parent assistance, and therapy for other issues.

The question isn’t “Do I take a pill or not?” — It should be “What could help with my current struggles?”

Articles describe how difficult it is to find somebody to assess ADHD, and how expensive it is. They describe a huge gap between the number of people who ‘have’ ADHD and the number of people who’ve been fortunate enough to get a diagnosis. They conclude that ADHD is massively underdiagnosed.

Minister for Mental Health Matt Doocey, Green Party Co-Leader Chlöe Swarbrick, ADHD NZ and the Royal NZ College of General Practitioners all hope access to ADHD assessments will improve from the 1st of February 2026. From then, General Practitioners (GPs) and nurse practitioners will be approved to diagnose ADHD, and to prescribe funded drugs for adults with it. Nurse practitioners working in child health services will also be allowed to diagnose and prescribe for people under 18.

Matt Doocey’s changes are based on a set of assumptions: that ADHD behaviours are caused by a faulty biology, that people benefit from an ADHD diagnosis, and that the obvious ‘treatment’ for ADHD is psychiatric drugs. Like most of us, he’s been told that the medical model will help people who have trouble with concentration, impulsivity and being still.

The problem is, these assumptions are all unfounded. As some of my previous posts have explored, no biological basis for what-we-call-ADHD has been established, ADHD behaviours aren’t a lifelong condition, and ADHD diagnosis seems to do more harm than good.

The February changes will be significant and are likely to result in increased rates of ADHD diagnoses. However it’s not clear that ADHD is underdiagnosed.

Let’s take a closer look.

Share

Determining under or over

For forty years, prevalence, which is the number of people who meet the criteria for what-we-call-ADHD, has stayed steady. At any point in time, about 1.5% of children and less than 1% of adults meet the criteria for ADHD.

But rates of ADHD diagnosis - the number of people being told by a clinician that they “have ADHD” - are increasing.

The obvious question is, how can we reconcile these two observations? How can increased rates of diagnosis align with the strong evidence that the underlying prevalence of ADHD is both relatively low, and has been constant over the years? Possible explanations include that the increase in ADHD diagnosis is somehow rectifying a gap — or that we are now overdiagnosing ADHD.

Lacking a comprehensive database of all New Zealand ADHD diagnoses, we need other ways to work out whether we are overdiagnosing ADHD. We’ll consider two approaches: comparing prevalence rates with dispensing data; and applying a model for overdiagnosis of physical disease to ADHD.

Comparing prevalence and dispensing data

Academics, blogs and government documents routinely compare prescribing rates with diagnostic rates in order to measure the underdiagnosis of ADHD. Writers argue that as there are many people with what-we-call-ADHD, but only a small number of people who are prescribed ADHD drugs, there must be many people who are missing out on ADHD diagnoses and prescriptions. Therefore, they conclude, we must be underdiagnosing ADHD.

For example, this Stuff article claimed that 5% (or 260,000) of New Zealanders met the criteria for ADHD last year, compared with only 167,000 people who were dispensed ADHD drugs. It concluded that about 100,000 people (the difference) were underdiagnosed because they ‘have’ ADHD according to prevalence data, but aren’t being prescribed drugs.

There are many problems with this conclusion, including inconsistent ways of measuring prevalence and dispensing rates, and an assumption that ADHD drugs are the only, or the best, response to what-we-call-ADHD.

Lifetime prevalence rates of above 5% are routinely quoted for ADHD. However as I explained here, across the whole population it is more likely that fewer than 1% of people meet the criteria for ADHD at any one time. Point prevalence, or the number of people who meet the criteria for ADHD at any particular time, is the type of prevalence we need to compare with prescribing rates, because prescribing rates count only the people using ADHD drugs at one particular time, not over a lifetime.

New Zealand researcher Ben Beaglehole and his colleagues published an overall annual ADHD dispensing rate of 0.8%1 for people living in Aotearoa. This figure was from 2022, but for each of the last few years, Pharmac data has shown annual increases of about a quarter in the number of individuals dispensed ADHD drugs. Continuing this trend, in 2025 the proportion of New Zealanders prescribed ADHD drugs will be about 1.5%, a substantial increase on the 2022 rate of 0.8% reported by Beaglehole.

So if we compare the point prevalence of ADHD with 2025 dispensings of ADHD drugs we find less than 1% of people met the criteria for ADHD, but about 1.5% of people were dispensed ADHD drugs. Clearly, we are both overdiagnosing and overprescribing for ADHD.

Does prescribing always follow diagnosis?

Although the data clearly show we are both overdiagnosing and overprescribing for ADHD, the overdiagnosis is probably even higher than the dispensing data suggest. Not everybody diagnosed with ADHD uses psychiatric drugs, and when they do, it’s usually temporary.

ADHD advocates often rush to reassure us that diagnosis is “not just about medication”. But then they consistently use the gap between (their) prevalence data and the dispensing data as evidence of underdiagnosis, while ignoring that many people do not use ADHD drugs after they have been DSM-diagnosed. The dispensing rate should be lower than the diagnosis rate because some people diagnosed don’t use drugs. But in Aotearoa the dispensing rate is higher (at 1.5%) than the true prevalence rate (1%) — further evidence of overdiagnosis.

Another approach: applying a general overdiagnosis model to ADHD

A very different method for assessing under- and over-diagnosis has reached the same conclusion. In 2021, Australian researcher Luise Kazda applied a framework for identifying overdiagnosis in conditions2 such as heart conditions and cancer to ADHD. The framework identifies overdiagnosis when five conditions exist:

1. the potential for increased diagnosis;

2. the diagnosis has increased;

3. the newly diagnosed have mild or “subclinical” symptoms;

4. the newly diagnosed receive treatment; and

5. the harms of diagnosis and treatment may outweigh the benefits.

Although Kazda’s focus is on children, her research still informs us about what’s happening generally. Her team examined 334 studies and found convincing evidence that ADHD is overdiagnosed in children and adolescents. They observed that a large proportion of the new cases are “mild”, and encouraged clinicians, parents, and teachers to be aware of the potential for overdiagnosis. Because for people with milder symptoms in particular, the harms associated with an ADHD diagnosis are more likely to outweigh any benefits.

Does overdiagnosis matter?

Yes. Overdiagnoses are harmful in themselves, but they also deflect people from better understanding important sources of distress in their lives.

Firstly, research shows that ADHD diagnosis itself can be harmful. According to independent studies, children diagnosed with ADHD do worse than children with the same behaviours, but who are not diagnosed with ADHD. On all outcomes. In her investigations, Luise Kazda found that “a biomedical view of difficulties was shown to be associated with disempowerment”. And that ADHD diagnosis increases stigmatisation, creating “an identity that enhances prejudice and judgment, which are associated with even greater feelings of isolation, exclusion, and shame.”

Secondly, as Kazda points out, diagnosis can obscure “other underlying individual, social, or systemic problems.” An ADHD diagnosis can hide other issues which can cause the same symptoms. Many are issues which call for quite different support and responses and the range is very wide: normal changes in cognition, the effects of physical health problems, the effects of emotional problems such as anxiety and depression, substance abuse, immaturity, fear, schools whose teaching approach doesn’t suit them, lack of support after being hurt, learning disabilities, grief, family dysfunction, parenting style, sleep deprivation, hunger, poor eyesight, hearing difficulties, poverty.

Where does this leave us?

According to Keith Conners, in the midst of “an epidemic of tragic proportions”.

Concerns about overdiagnosis of ADHD have been raised by mental health leaders for over 20 years. Keith Conners, creator of widely used ADHD screening tools; Allen Frances, chair of the DSM-IV task force; Sami Timimi, psychiatrist and advocate of alternative approaches to what-we-call-ADHD; and Alan Sroufe, developmental psychologist, have all expressed dismay at the increases they have witnessed since the turn of this century.

New Zealand already overdiagnoses ADHD, and the February expansion of access to ADHD assessments is likely to make overdiagnosis even worse. Increased focus on ‘getting a diagnosis’, rather than finding out what is going on for each person, will just channel people even faster into a medical model that has only medical answers.

Turns out this is trickier to answer than you might expect, and most reported estimates are too high.

Asking how many children meet the criteria for ADHD is different from asking how many children have been diagnosed with ADHD. And it’s different from asking how many children are being ‘treated’ for ADHD. Instead it’s asking, how often does this combination of behaviours cause difficulties for children? What’s the rate? Or more technically, what’s the prevalence?

Prevalence estimates

Estimates of the number of children who ‘have’ ADHD vary a lot. A very big lot. For example, a 2013 systematic review of global data included a study that found an ADHD rate of 0.2% in Malaysia, and another that found a rate of 12.3% in Iran. Within the United States, a 2022 comparison found an average rate of 8.7% across the country as a whole, but the rates in individual states ranged from 5% to 15%. Even different editions of the DSM have published different rates: 3% in 1980, 3-5% in 1994 and 5% in 2013.

Prevalence estimates from Aotearoa vary less: the Dunedin Multidisciplinary Study reported a childhood ADHD rate of 6.1% while the 2024 New Zealand Health Study reported a rate of 5.3%. The 2001 Ministry of Health New Zealand Guidelines for the Assessment and Treatment of ADHD alluded to the complexities of estimating the true prevalence of ADHD. Their solution seemed to involve combining the DSM-IV estimate with the opinion of New Zealand psychiatrist John Werry, and settle on 5%.

Faced with such a range of estimates, how do we decide what is accurate? As for any ADHD research, we need to appraise the available research for bias, clarify which measure of prevalence the studies are reporting, and establish what types of information the studies have used to calculate the ADHD estimates.

Bias

As I’ve explained before here and here, bias is pervasive in ADHD research. It shows up in prevalence studies too. In order to work out the most likely prevalence of what-we-call-ADHD, we need to first check prevalence studies for bias. Then we need to clarify what they define as “prevalence”.

Not all prevalences are equal

Different questions about ADHD give us different answers. We can ask how many children have ever met the criteria for ADHD, at any time in their lives. Or we can ask how many children have a current diagnosis of ADHD at the time of the study, which will result in a smaller count of children.

In order to compare ADHD apples with ADHD apples, we’ll focus on the second definition: the number of children with ADHD at the time of the research. This is sometimes called incidence, or ‘point prevalence’ because it tells us how many people meet the criteria at a certain instant in time. Point prevalence is used often in global studies. It doesn’t confuse attempts to identify lifetime rates with what is happening now. It better reflects the fluctuating course of ADHD behaviours. And it makes it easier to compare prevalence with rates of ADHD treatment, which are usually reported as point prevalence. Point prevalence also establishes more accurate estimates of ADHD levels, because beyond three months people’s recall becomes less reliable, including for ADHD behaviours.

Next, we need to establish what types of data contributed to the analysis.

Share

Data sources

Different sorts of data, different combinations of data, and different methodologies, produce very different prevalence rates.

Rates of ADHD have been estimated by surveys, parent reports, teacher reports, self reports, direct observation, electronic medical records and clinical interviews with parents and children. The combination of sources used to determine ADHD eligibility makes a huge difference to the prevalence. For example, this global review and this large cross-sectional study found that not using DSM-qualifying, multi-informant methodologies inflated ADHD prevalence threefold.

Even when data from multiple informants is used, and DSM criteria are strictly applied, results may still over-estimate rates of ADHD by a factor of two. For example, a 2019 Australian study found that only 43% of children who had been screened positive as potential ADHD cases by high scores on both the parent and teacher Conners Rating Scales, or who were said to have been previously diagnosed with ADHD, met the criteria for ADHD when assessed in person by a clinician.

How does a psychologist assess whether a child meets the criteria for ADHD?

Completing a full assessment for ADHD incorporates several components. The child must display at least six inattention and/or impulsivity behaviours. Information is usually gathered from parents, teachers and the child, using rating scales and interviews, and by a school observation. The psychologist establishes whether the behaviours were present before the child was 12 years old and are evident in multiple environments - typically home and school. There must also be clear evidence that the behaviours impair the child’s so­cial, academic or occupational functioning and are not better explained by something else. Almost no ADHD estimates are based on these type of assessments.

The most common, and least accurate, way of evaluating ADHD prevalence is using screening tools. These tools are often hosted by organisations like ADHD NZ on their websites. Although they are designed to filter possible cases, with the assumption that a full assessment with clinicians will follow if warranted, people often equate screening results with diagnosis. But using only a screen or a self-report test leads to many false positives. Here’s why.

Take the Conners Parent Rating Scale - one of the most widespread scales used to estimate ADHD. It has been found to correctly identify 75% of the children who do ‘have’ ADHD (called test sensitivity) and 75% of the children who don’t (called test specificity). At first glance these percentages can appear reasonable. But for rare childhood difficulties like ADHD, sensitivity and specificity of 75% lead to very high numbers of false positives.

The Conners will correctly identify 38 of the 50 children in every 1000 (5%) who ‘have’ ADHD. (Twelve will be missed.) But it will also falsely label another 237 of every 1000 children as having ADHD when they don’t. Therefore out of 275 diagnoses suggested by the screen (the 38 children who do have ADHD combined with the 237 who don’t), 86% (237) of them are wrong. Put another way, 86 out of every 100 diagnoses suggested by the Conners Parent Rating Scale alone will be false positives. We can’t rely on screening measures alone.

To accurately estimate the level of ADHD behaviours I identified studies that account for the effects of multiple informants, impairment ratings and assessments with trained clinicians.

Best estimate

Given these issues with prevalence research, where can we find some solid ground? Especially while organisations like ADHD NZ remind us that rates are higher than we think.

Two studies stand out: a global review by Holly Erskine published in 2013 and a detailed analysis of a large US study led by Michaela Cordova in 2022.

Erskine and her colleagues analysed ADHD prevalence data as part of a broader investigation for the Global Burden of Disease Study. Using studies from 1990 to 2010 their model weighted different types of prevalence and data sources. Overall they estimated an ADHD point prevalence of 1.5% for 4-19 year olds, and found steady prevalence rates across those 20 years. But large differences were found between age groups. Uniformly across the world, ADHD rates peaked for 9-year olds at about 3-4%, and dropped off throughout adolescence.

More recently Cordova and her colleagues analysed data from the Adolescent Brain Cognitive Development Study to determine ADHD prevalence in the United States. The country’s largest longitudinal study of child mental health, it assessed 11,878 children aged 9-10 years old on multiple dimensions. Cordova’s analysis combined reports from parents, teachers and children themselves, including information about the duration and level of impairment. This analysis produced a point prevalence of 3.5% for 9-10 year olds. Because only 9-10 year old children were studied we would expect Cordova’s estimate to represent the peak prevalence for children, but to be lower overall if children if other ages were included.

Erskine and Cordova’s findings are supported by other careful but smaller studies. For example, the longitudinal Great Smoky Mountains Study found an overall point prevalence of 0.9%, with a peak rate of 2.2% for children aged 9-10 years old. A more recent Global Burden of Disease Study found an overall point prevalence of 2.2% — with a peak of 2.87% in children aged 10-14 age.

These lower ADHD estimates might be surprising when rates of over 10% are often suggested. But the best, unbiased, careful evidence finds consistently and significantly lower rates than generally reported: less than 2% across all children and adolescents, with a peak at about 9-10 years old, and no changes over the last 40 years.

What about here in Aotearoa?

Internationally, thorough studies converge on childhood ADHD point prevalence rates of less than 2%. But on its website, ADHD NZ tells us that 1 in 20, or 5% of, people in New Zealand have ADHD. For children they cite an even higher 5-8%.

Given this discrepancy, I looked for local New Zealand studies to compare with the international research, and found two, with headline ADHD rates of 5.4% and 6.1% in children. But if we look at how these two studies measured prevalence, both numbers seem to overestimate the true point prevalence in New Zealand.

The 2024 New Zealand Health Study (NZHS) has been widely cited as a source of local prevalence rates of ADHD. It reported that 5.4% of parents responded yes to the question, “Have you ever been told by a doctor that your child has ADHD?” This is more a question about whether a child has been diagnosed with, or considered to have, ADHD, rather than whether a child meets the criteria for ADHD. Secondly, it’s a question about lifetime prevalence, not point prevalence. So we would expect it to generate a rate well above the global 1.5%. Thirdly, the question is only posed to parents - a single source - and we know that single-source surveys inflate ADHD rates by up to three times. A final puzzling aspect of this question is that it asks parents whether “a doctor” told them this. In New Zealand “doctor” usually refers to general practitioners, who are not able to diagnose ADHD in children. How the question was understood by respondents is unclear, but taking these complications into consideration the Survey’s data suggest that the true point prevalence for children in the NZHS cohort is more likely to be below 1.8%.

Other local data comes from the Dunedin Health and Development Multidisciplinary Study. This study of all the children born in Dunedin, New Zealand between April 1972 and March 1973 found 6.1% were diagnosed with ADHD as children. However this whole-of-childhood estimate was not point prevalence. Similarly to international studies, the point prevalence was 4% at age 11, close to the expected peak. As seems typical, reported rates in this group of children then halved over the following two years.

Where does this leave us?

ADHD prevalence figures above 10% are proposed in media, on websites and in journal articles. But based on high quality local, international and global evidence, a more realistic estimate of current prevalence would be about 1.5% for children and lower still for adults.

However I suggest the true point prevalence is even lower.

Careful estimates of ADHD make sure that the child meets the criteria, that multiple informants are used, that the behaviours occur in more than one setting and that the behaviours clearly impair the child’s life. But I think that the prevalence literature asks the wrong question: “Does this child meet the criteria for ADHD?”

This question fails to allow for a better explanation of a child’s behaviours. It ‘rules in’ that a child meets the criteria without ‘ruling out’ other possible causes. ADHD behaviours might be better explained by other things that can mimic ADHD: anxiety, immaturity, a mismatch with their school’s teaching approach, parenting styles, learning disabilities, depression, hunger, poor eyesight, unresolved trauma responses, bullying, the cognitive effects of substance abuse, sleep deprivation or poor hearing.

A better question would be, “What is the most useful explanation for this child’s behaviours?”

This alternative question is implied in the DSM, when it directs clinicians to ensure that the behaviours under scrutiny are not better explained by something else. But in effect this component is ignored by prevalence research. Of course the DSM assumes any alternative explanation will be another what-they-call-mental-disorder. But as listed above, ADHD behaviours might be explained by many other things too.

Once we add this critical element to the assessment of ADHD, a further reduction in prevalence rates would have to follow, resulting in an overall point prevalence rate of ADHD for children and adolescents of less than 1.5%.

Subscribe now

1. Ministry of Health. (2001). New Zealand Guidelines for the Assessment and Treatment of Attention-Deficit/ Hyperactivity Disorder. Wellington: Ministry of Health.

2. Erskine HE, Ferrari AJ, Nelson P, et al. (2013). Epidemiological modelling of attention-deficit/hyperactivity disorder and conduct disorder for the Global Burden of Disease Study 2010. Journal of child psychology and psychiatry, and allied disciplines, 54, 1263–1274

3. Cordova MM, Antovich DM, Ryabinin P, et al. (2022). Attention-Deficit/Hyperactivity Disorder: Restricted Phenotypes Prevalence, Comorbidity, and Polygenic Risk Sensitivity in the ABCD Baseline Cohort. Journal of the American Academy of Child and Adolescent Psychiatry, 61, 1273–1284.

4. Ministry of Health. (2024). Annual Data Explorer 2023/24: New Zealand Health Survey. URL: minhealthnz.shinyapps.io/nz-health-survey-2023-24-annual-data-explorer/ (accessed 30 June 2025).

5. Moffitt TE, Houts R, Asherson P, et al. (2015). Is Adult ADHD a Childhood-Onset Neurodevelopmental Disorder? Evidence From a Four-Decade Longitudinal Cohort Study. American Journal of Psychiatry, 172(10): 967-77. https://psychiatryonline.org/doi/10.1176/appi.ajp.2015.14101266

Last week the Ministry of Health published its New Zealand Clinical Principles Framework for ADHD. The Framework was developed by a “Clinical Reference Group” that met between September and December 2024. It is not claiming to be a guideline, but a “clinical framework” which sets out expected standards for the assessment, diagnosis and treatment of ADHD, and who can do those tasks.

TLDR? I’m really disappointed.

The Clinical Reference Group included representatives from advocacy group ADHD New Zealand, as well as from professional bodies like the New Zealand College of Clinical Psychologists, the Royal Australian and New Zealand College of Psychiatrists, and the Royal New Zealand College of General Practitioners.

This resulting Framework represents the “consensus view” of the Group, a phrase that echoes the ADHD consensus statements published by ADHD advocates in 2002, 2010 and 2021. These consensus statements tried to state the ‘truth’ about ADHD, and also to prevent debate about it. But they also evoked critical responses.

Before describing what is disappointing about the Framework content, I’ll describe two problems with the process of creating this framework: no representation from the New Zealand Psychological Society, and complete failure to mention conflicts of interest.

The New Zealand Psychological Society is the largest professional body for psychologists in Aotearoa, and the professional home for Educational Psychologists. Psychologists come in many shapes and sizes - some focus on this and some focus on that. Educational Psychologists (Ed Psychs for short) are arguably the most involved in supporting children who exhibit ADHD behaviours, because they work most closely with schools. Excluding Ed Psychs from this Clinical Reference Group is very puzzling, and undermines my trust in the Framework development process and its outcomes. What has happened in the development of this Framework that our largest psychology organisation, and the psychologists who work most closely with children have been excluded from the start?

Confidence in the Framework is also undermined by the absence of any declarations of conflicts of interest. ADHD research is at high risk of bias, and conflicts of interest are pervasive. We know that NZ psychiatrists received drug company sponsorship to support the New Zealand launch of the Australian guideline. Confidence in guidelines is much higher when conflicts of interest are reported, like, for example, in the 2018 NICE ADHD Guidelines.

Apart from my concerns about the ways in which the Framework was developed, there are many problems with what it says - or doesn’t say. It recommends the unsound Australian ADHD Professionals Association (AADPA) guideline, ignores critical ADHD issues, and perpetuates misinformation about ADHD.

Subscribe now

Framework recommends the Australian guideline

Perhaps most disappointing is the Framework’s blessing of the flawed Australian ADHD guideline. Jeltsje Keizer and I evaluated the Australian Evidence-Based Clinical Practice Guideline for ADHD against expectations of guidelines for New Zealand clinicians. We concluded that the failures of the guideline were serious enough to bar its adoption for use here.

The Australian guideline has so many problems. It was developed by an organisation that receives drug company funding. Only 2% of its recommendations were based on certain evidence. It ignored over-diagnosis, the risks of diagnosis and Relative Age Effects. And despite including a “Special Populations” section about Aboriginal and Torres Strait Islander people, the issues raised in it were not carried through the Guideline’s recommendations. But still the New Zealand Framework specifically recommends that we use the Australian guideline.

Framework omits critical issues

Just like its Australian ancestor, the Framework is silent on two important issues: the risks from ADHD diagnoses and the well-established Relative Age Effects. Jeltsje and I talked about them in our 2023 paper, and if you want the full details, I’ve given each one its own Substack post. But if you want a refresher, read the next two sections.

Risks of diagnosis

Two independent studies have found that teenagers who were diagnosed with ADHD fared worse than teenagers exhibiting the same behaviours, but not diagnosed with ADHD. It might seem counter-intuitive, but in the long term, children are better off without an ADHD diagnosis - even if they show all the behaviours that could lead to a diagnosis. Parents, teachers and clinicians need to know that there are risks from ADHD diagnoses. And clinicians should be cautious about diagnosing ADHD in children, especially when behaviours are mild.

Parents and clinicians can address children’s behaviours without seeking diagnoses. Clinicians can support parents to help their children, and both can support children to understand themselves and what helps them thrive. This “needs-led support” should be encouraged by any Framework, as recommended by the UK’s Report of the independent ADHD Taskforce: Part 1.

Relative Age Effects

Anywhere in the world, the youngest children in class are much more likely to be diagnosed with, and medicated for, ADHD than the oldest children in class. Clinicians and parents need to know that the best explanation for this is that we are diagnosing immaturity as ADHD. This Framework should have alerted clinicians to relative age effects, and suggested other ways to support children who are being asked to behave beyond their years.

Framework perpetuates misinformation

This Framework propagates the unproven idea that ADHD behaviours are caused by biological pathology. It claims that ADHD is life-long, and implies that treatment with psychiatric drugs is the preferred primary strategy. Again, these ideas are inherited from the Australian guideline, and I’ve talked about these in earlier ADHD Files.

Biological basis

In its opening sentence, the New Zealand Framework states that ADHD is a “neurodevelopmental condition” — a declaration that ADHD is caused by pathological biology such as genetics, anatomy or neurotransmitters. In fact this has not yet been shown to be true despite many expensive studies, as I talked about in the chemical imbalance File. Even ADHD advocates agree that we don’t yet have evidence to link it to any specific biological cause.

The biological theory has been robustly disputed for at least 20 years. For example, in 2004 Sami Tamimi and his colleagues similarly concluded that:

The evidence does not support the conclusion that ADHD identifies a group of children who suffer from a common and specific neurobiological disorder. There are no cognitive, metabolic, or neurological markers for ADHD.

Repeatedly referring to biological differences between those with and without ADHD creates the illusion of a consensus about the causes of ADHD, where no consensus exists. We know that people vary, and they’re very affected by their circumstances. But the New Zealand Framework ignores warnings from local clinicians about the risks of applying a medical model to Māori mental health concerns.

“Simply describing behaviours that focus on an individual struggling to conform within a society that does not reflect a Māori worldview is no longer acceptable... this medicalised approach failed Māori” (Kopua & Skirrow, 2023).

A view echoed in the Australian guideline, when it acknowledges variation in the way that cultures view ADHD behaviours:

“Some symptoms of ADHD ... may not be considered problematic by Aboriginal and Torres Strait Islander peoples, as these may be viewed as usual and appropriate responses to the environmental context” (p157).

The New Zealand Framework reduces this important discussion to only one sentence, suggesting that clinicians should “take into account te ao Māori worldviews”. It refuses to engage with the implications of the idea that different cultures might have different views on the causes of ADHD behaviours, or to draw on New Zealand expertise about what a Māori view could add for the people who the Framework is intended to help.

Implies ADHD is life-long

The Framework states that it’s “designed for the New Zealand context”. It also claims that ADHD “persists into adulthood in half or more of cases”. Yet our very own longitudinal ‘Dunedin Study’ found the opposite. In the Study, only 5% of people diagnosed with ADHD as children later met the criteria as adults. Why does a New Zealand Framework “designed for our local context” ignore what we have learnt about ADHD in our own people?

Prioritises drugs

As I explained in the file Do Drugs Help People With ADHD, we have only uncertain, weak evidence that psychiatric drugs help people with ADHD in the short term, and no evidence that they help in the long term. But the New Zealand Framework emphasises drug interventions over other interventions, and always suggests “pharmacological treatments” before “non-pharmacological treatments”.

Where does this leave us?

This Framework is not without any merit whatsoever. It recommends that assessments take into account te ao Māori worldviews. It recommends that doctors review their patients within two to four weeks after prescribing drugs. It recommends multidisciplinary assessment for children and young people.

However, the Framework’s omissions amplify the flaws of the Australian guideline. And its single-diagnosis model will further distort how New Zealand communities think about attention and impulsivity difficulties. People need support that is uncoupled from clinician diagnosis, and they need non-pharmacological help with their struggles with attention and impulsivity.

It feels like the Ministry of Health is trying to bring the Australian guideline to Aotearoa by stealth. In 2023 I asked a Ministry of Health official whether they had plans to endorse the Australian guideline. He told me there was no Ministry mechanism in place to endorse guidelines from overseas. Sadly, it seems that they’ve found a way.

1. Ministry of Health. (2025). New Zealand Attention Deficit Hyperactivity Disorder Clinical Principles Framework. Wellington: Ministry of Health.

2. National Health Service. (2025). Report of the independent ADHD Taskforce: Part 1. London: National Health Service.

3. Timimi S, Moncrieff J, Jureidini J, et al. (2004). A critique of the international consensus statement on ADHD. Clinical child and family psychology review, 7, 59–69.

4. Kopua D & Skirrow P. (2023). Racism, Mātauranga Māori and ADHD: An Interview With Dr Diana Kopua. Journal of the New Zealand College of Clinical Psychologists, 33, 72–78.

You can’t just read the research. You certainly can’t just read the summary at the start. You need to read the whole paper, including the section at the end that lists, in very small letters, how these researchers were supported by drug companies.

Why?

Because ADHD research is at high risk of bias, and conflicts of interest are pervasive. And the research doesn’t seem to be improving over time. For example, the first comprehensive review of “virtually every investigation ever done on ADHD drugs anywhere in the world” rejected 2 107 of the 2 287 studies (92%) for being unreliable. Twenty years later, a review of Ritalin for children still found that 90% of included studies were at “high risk of bias”.

Industry funding affects all aspects of research: the study design, the outcomes, the reporting. We know that industry-funded trials report fewer adverse effects than independent trials. We know that researchers are more likely to reach pro-industry conclusions if they receive industry money. We know that industry-funded trials are published in higher impact journals, and then cited more often. All of which makes it hard to be confident in research conclusions.

One credible source is the independent Cochrane Collaboration, which produces high-quality, systematic reviews about healthcare interventions. Cochrane reviews synthesise all available evidence on a specific question, using explicit and systematic methods in order to give unbiased answers. Another credible source is the Drug Effectiveness Review Project, which publishes comprehensive reviews of published and unpublished research about the comparative effectiveness and harms of drugs.

Independent researchers consistently conclude that the evidence about ADHD drugs is uncertain, weak and of poor quality. We need to keep this in mind when considering how ADHD drugs help. But with care, we can find some solid places to stand.

In the short-term

From 2006 to 2022 the number of people dispensed ADHD prescriptions in Aotearoa New Zealand more than quadrupled — from 188 per 100,000 people to 819 per 100,000. About 90% of those prescriptions were for methylphenidate, which you might know better by its brand name: Ritalin.

The accepted wisdom is that ‘core ADHD symptoms’ - inattention, hyperactivity and impulsivity - decrease when people take drugs like Ritalin. A decrease in core symptoms means that people fidget less, wait their turn more, stay better focused and make fewer careless mistakes. The assumption about the short-term benefits is widespread and claimed for both children and adults. But independent reviews still describe that evidence as uncertain and at great risk of bias. It seems we don’t yet have good evidence to support this claim.

In the long-term

If the short term evidence about core symptoms is weak, what about in the long term? Perhaps drugs have long term benefits that make it worth people persisting. Or perhaps there are broader, real life benefits like better social skills, emotional well-being or academic performance.

For over 25 years, reviews have warned about the scarcity of studies measuring “functional or long-term outcomes”. And still there is almost nothing to provide clear long-term guidance when it comes to ADHD. Here, I’ll look through the most helpful studies and reviews.

The most useful long-term study is the Multimodal Treatment of ADHD Study (MTA). In fact, while it’s complicated and can be hard to decipher, it’s the key longitudinal ADHD study. A multi-site study, it followed 579 children aged from about eight years old until they became adults. It was designed to compare behavioural therapy with psychiatric drugs for children diagnosed with ADHD, and is still the only long-term study that compared the two directly.

Children diagnosed with ADHD were randomly allocated to one of: behaviour therapy; psychiatric drugs; a combination of behaviour therapy and psychiatric drugs; or community care. The behaviour group received parent training, a summer treatment camp and classroom management; the medication group received Ritalin; the combined group received both behaviour therapy and Ritalin; those in the community group were given information about services but otherwise left to their own devices.

When the children were evaluated 14 months later, the greatest reductions in core ADHD symptoms were found in the groups that had received medication. However 24 months after the start of the study, the benefits of medication had reduced, and by 36 months “medication use [between 24 and 36 months] was a significant marker not of beneficial outcome, but of deterioration.” MTA investigator William Pelham reflected later

We had thought that children medicated longer would have better outcomes. That didn’t happen to be the case. There were no beneficial effects, none. … In the short term, [medication] will help the child behave better, in the long run it won't. And that information should be made very clear to parents.

Long term followup later found substantial improvement for all children regardless of which group they were allocated to - including the community group. The authors concluded this probably reflected “a natural waning of symptoms” in at least some children, which aligns with the evidence that an ADHD diagnosis often reflects immaturity rather than a disorder.

The MTA children who had used Ritalin didn’t do better in the long term, but most importantly they didn’t do better on measures of emotional well-being, social skills or academic outcomes. William Pelham accepted these unexpected findings, and advocated that children diagnosed with ADHD should receive behaviour therapy and parental support first, and only then be offered low doses of stimulants if necessary. This approach is also recommended by the independent UK guideline but is not happening here, now, for our children.

Share

The MTA focused on children. ADHD in adults has received far less attention. But a similar pattern is emerging: authors report that drugs help with short-term core symptoms but not that they help in the long term. However, just like for children, adult reviews report that the evidence of these short term positive effects is of poor quality. Cochrane reviews about adult use of Ritalin by Raissa Cândido and her colleagues and Kim Boesen and his colleagues found some low certainty evidence that drugs reduce ADHD symptoms. But as no trials continued past 18 weeks, there was no evidence on which to evaluate the long term effects.

The second useful study is by Isabell Brikell and her colleagues, who measured how long people persist in taking ADHD drugs for - which is another way to find out how helpful people find them to be. In 2023, they collated information about duration of use. Surprisingly, they found that within one year, 52% of adults and 35% of children had stopped taking the ADHD drugs. Within five years, 85% of adults and 76% of children had stopped. Common reasons for stopping included the drugs being ineffective, and side-effects. It seems that people decide for themselves that the drugs don’t help.

If ADHD drugs were the cure they are assumed to be, and if ADHD were a biological disorder, wouldn’t we expect more people to persist with a drug intervention? We have only uncertain, weak evidence that they help in the short term, and no evidence that they help in the long term. What should people do?

So where does this leave us?

Clear and unbiased information from professionals about the limitations of psychiatric drugs is crucial to ensure that people have realistic expectations when deciding whether to try them. Parents need this information when considering drugs for their children. Adults need this information when considering drugs for themselves.

Our hopes for these drugs have out-stripped the evidence base for their usefulness. I’d like people to know about the weakness of the evidence for short term benefits, and about the lack of evidence for long term benefits. I also want people to know about and weigh up the risks of adverse side effects and of diagnosis itself.

We also need clear and unbiased information about what else can help with ADHD behaviours, from supporting parents and teachers to offering micronutrients. I’ll look at these other options in future posts.

1. Storebø OJ, Storm MR, Pereira Ribeiro J, et al. (2023). Methylphenidate for children and adolescents with ADHD. Cochrane Database of Systematic Reviews, 3, CD009885.

2. Boesen K, Paludan-Müller AS, Gøtzsche PC, et al. (2022).
   Extended-release methylphenidate for attention deficit hyperactivity disorder (ADHD) in adults. Cochrane Database of Systematic Reviews, 2, CD012857.

3. Cândido RCF, Menezes de Padua CA, Golder S, et al. (2021). Immediate-release methylphenidate for attention deficit hyperactivity disorder (ADHD) in adults. Cochrane Database of Systematic Reviews, 1. Art. No.: CD013011.

4. Beaglehole B, Jarman S, & Frampton C. (2024). Dispensing of attention-deficit hyperactivity disorder medications for adults in Aotearoa New Zealand. The New Zealand Medical Journal, 137, 23 – 30.

5. Brikell I, Yao H, Li L, et al. (2023). ADHD medication discontinuation and persistence across the lifespan: a retrospective observational study using population-based databases. The Lancet Psychiatry, 11, 16 - 26.

But is it?

For at least 50 years, researchers have speculated that a chemical imbalance causes ADHD. The neurotransmitter dopamine has been the focus of this hypothesis, because the stimulant drugs used to treat ADHD act on the dopamine receptors in the brain. Many different approaches have been used to study this idea, with varied results. Therefore, to determine whether low dopamine levels cause ADHD we can’t rely on any one study. We need careful reviews or meta-analyses of all relevant studies.

Low dopamine as a cause of ADHD

Last year, Hayley MacDonald and her colleagues did just that. They collated the evidence about the specific role of dopamine in ADHD and concluded

there is limited evidence that reduced levels of the neurotransmitter [dopamine] per se is a defining feature of ADHD.

And that more research is needed.

Limited evidence, after 50 years of research. For the people who’ve been working on this question over those years this must be a disappointing summary - that the evidence still isn’t there. But MacDonald’s verdict was similar to another more extensive meta-analysis published last year.

Any biological cause of ADHD

Valeria Parlatini and her colleagues looked further than dopamine. They evaluated a range of possible biological causes of ADHD, not just a chemical imbalance. Focused on biomarkers that might aid the diagnosis of ADHD, they included genetic, epigenetic, neuroimaging, neuro-physiological, neuropsychological and chemical biomarkers in their analysis and found that

despite the large number of studies and variety of techniques used, no promising biomarkers have been identified so far.

To be considered adequate evidence, Parlatini’s review required that a biomarker (a biological indicator of ADHD) be confirmed by at least two independent studies and accurately distinguish people with ADHD from people without. For each possible biomarker, Parlatini found early positive studies that had been described as “promising.” But those early studies had major methodological problems, failed to replicate in later studies, or didn’t adequately distinguish between people with ADHD and people without.

Parlatini and her colleagues are ADHD advocates. Three of the study authors disclosed extensive involvement with drug companies. They believe there are biological causes of ADHD, which haven’t been found yet. For them, the absence of evidence after 50 years work means, like MacDonald, that more research is needed.

Any biological cause, of any neurodevelopmental disorder

Broader reviews combining the research about many neurodevelopmental disorders also find comparable results. For example, in 2023, Samuele Cortese and his colleagues reviewed possible biomarkers for any neurodevelopmental disorder, including ADHD. Like MacDonald and Parlatini, Cortese concluded

we could not find any biomarker for which there was evidence.

And argued for more research.

Share

Why the interest in a chemical imbalance?

Is it reasonable to ask whether a lack of dopamine causes ADHD? Certainly similar questions have been asked about other what-we-call-mental-disorders. For example, the theory that depression is caused by a chemical imbalance was first suggested in the 1960s. In that case the proposed neurotransmitter was serotonin.

Investigators were confident an imbalance would be found, and many possible mechanisms were explored. But careful reviews, including a key analysis published in 2022 by Joanna Moncrieff and her colleagues, have concluded that the evidence does not show depression is caused by a lack of serotonin. As Moncrieff said after her analysis

I think we can safely say that after a vast amount of research conducted over several decades, there is no convincing evidence that depression is caused by serotonin abnormalities, particularly by lower levels or reduced activity of serotonin.

Just as no conclusive evidence has shown that depression is caused by a lack of serotonin, no conclusive evidence has shown that ADHD is caused by a lack of dopamine. We can’t claim that a lack of dopamine causes ADHD, we can’t claim that any biological marker identifies ADHD, and we can’t claim that any biological cause for any neurodevelopmental disorder has been identified. Do we just need more research?

Each study that fails to prove a biological cause of ADHD concludes that we need more research (to find one). How much research do we need before we decide there isn’t a biological cause of ADHD? The scientific process offers an alternative to the need-for-more-research conclusion. It could instead be concluded, that we’ve looked very hard, and it seems that ADHD isn’t caused by a lack of dopamine. Just as Moncrieff said.

The idea that ADHD is caused by a dopamine imbalance remains a theory. A suggestion. A possibility that has not been found to be true. It’s misleading, at this time, to assert that ADHD is caused by a dopamine imbalance, or by any other biological process or structure.

Does stating this continually make it true? No. But it’s easy to mistake repetition for truth. Repeatedly starting articles, guidelines and video clips with references to biological differences between those with ADHD and those without creates the illusion of a consensus about the causes of ADHD. In fact this theory has not been substantiated, and is robustly disputed. People need to know this theory is unproven, especially people being diagnosed with ADHD, and the parents of children being diagnosed with ADHD.

Although the dopamine theory has not been proven, we still know that people are varied, and they’re affected by their circumstances. Maybe it’s time to expect variation and difference, not look for deficit and disorder? And then to help people discern what works better in their life, given their differences and their circumstances? If there isn’t a biological cause of what-we-call-ADHD, if symptoms are situational or conditional, maybe what will help is based in their environment as well.

1. MacDonald HJ, Kleppe R, Szigetvari PD et al. (2024). The dopamine hypothesis for ADHD: An evaluation of evidence accumulated from human studies and animal models. Frontiers in Psychiatry, 15:1492126. doi: 10.3389/fpsyt.2024.1492126.

2. Parlatini V, Bellato A, Gabellone A, et al. (2024). A state-of-the-art overview of candidate diagnostic biomarkers for Attention-deficit/hyperactivity disorder (ADHD). Expert Review of Molecular Diagnostics, 24, 259–271.

3. Cortese S, Solmi M, Michelini G, et al. (2023). Candidate diagnostic biomarkers for neurodevelopmental disorders in children and adolescents: a systematic review. World Psychiatry, 22, 129-149. doi: 10.1002/wps.21037.

4. Moncrieff J, Cooper RE, Stockmann T et al. (2022). The serotonin theory of depression: a systematic umbrella review of the evidence. Molecular Psychiatry, 28, 3243-3256. doi: 10.1038/s41380-022-01661-0.

Chronic disorders trouble us for a long time, if not our whole lives. In articles, guidelines and consensus statements, ADHD is routinely described as a chronic disorder. But is it?

Not if you consider the results of our very own “Dunedin Study”.

The Dunedin Multidisciplinary Health and Development Study includes all 1037 children born in Dunedin, New Zealand between April 1972 to March 1973. They’ve been followed closely over the last 50 years, and had taken part in nine psychological assessments by the age of 45.

Avshalom Caspi and his colleagues (2020) analysed the mental disorders each person from the Dunedin Study had been diagnosed with by the time they were 45 years old. Eight-six percent of people had been diagnosed with at least one mental disorder by the time they’d reached midlife, and 73% had been diagnosed with more than one.

Most importantly, the kind of mental disorder people were diagnosed with changed throughout the course of their lives. For example, one person might be diagnosed with ADHD as a child, with depression as a teenager and then with a thought disorder as an adult. Another might be diagnosed with anxiety as a child, OCD as a teenager and then no diagnoses as an adult. Sometimes a person struggled in one way, and at other times that same person struggled in a different way. Any formal diagnoses shifted accordingly.

If ADHD were truly ‘chronic’, we’d expect people to present with the same diagnosis over the course of their life. Not to change diagnoses, in all directions, as was found in the Dunedin Study. Rather than ADHD enduring, this study suggests that ADHD behaviours come and go over the course of a person’s lifetime.

Child-adult overlap

Another analysis of the Dunedin Study provides even clearer evidence that ADHD doesn’t persist across a lifespan. Of the 1037 people in the Dunedin Study, 61 were diagnosed with ADHD as children (6% of all participants) and 31 were diagnosed as adults (3%). It’s common to find more children diagnosed with ADHD than adults. Figures like 6% of children and 3% of adults are within the general rates identified in ADHD research, usually explained as children growing out of their ADHD symptoms over time. What do we learn though, when we look more closely at those specific 61 people diagnosed as children and 31 diagnosed as adults? How many of the 31 adults had previously been diagnosed with ADHD as children? And how many of the 61 children went on to be diagnosed with ADHD as adults?

Terrie Moffitt and her colleagues discovered that only three of the 61 people who had ADHD as children also met the criteria for ADHD when assessed as adults. Or to put it another way, only three of the adults diagnosed with ADHD had previously met the criteria for ADHD as children. There was almost no overlap between the group of people diagnosed as children and the group diagnosed as adults. The people diagnosed with ADHD as children were different than the adults diagnosed with ADHD.

If ADHD were a chronic, neurodevelopmental disorder we would expect a big overlap between the children and adults diagnosed with ADHD, not just 3 of the 1037 individuals in the Dunedin Study.

The MTA Study

A different longitudinal study has found the same lack of persistence of ADHD behaviours as children mature into adulthood.

We’re going to have to talk more about the MTA Study. In conversations about ADHD it’s important - it’s the study. But today let’s just look at its most recent report and work backwards another time.

The Multimodal Treatment of ADHD Study (MTA) was a multisite study designed to compare behaviour therapy with psychiatric drugs for children diagnosed with ADHD. 579 children, aged 7-9, were randomly assigned to one of: behaviour therapy, psychiatric drugs, a combination of both; or routine community care (the control group). The participants have been followed for 17 years.

Last year Margaret Sibley and her colleagues reviewed how ADHD symptoms in MTA participants changed over the 17 years of the study. In contrast to the medical model of ADHD that views it as a persistent disorder, the researchers showed that for most people, their symptoms and level of impairment fluctuated over the years, often quite substantially. Only about 11 percent of the children who entered the study experienced their ADHD symptoms consistently year after year. More often, their symptoms would come and go; for a few years, they might stay above the symptom threshold, and then for a few years, their symptoms might dip below the cutoff, sometimes disappearing altogether. In other words, most people did not experience enduring, or chronic, ADHD behaviours. The behaviours came and went over the years.

Where does this leave us?

The suggestion that ADHD is a chronic disorder is not supported by pivotal studies. Research increasingly suggests that ADHD is better thought of as something that people experience, for a while or at times, rather than as a disorder that they have in some unchanging way.

Most people struggle at some time or times in their lives. And most struggle in more than one way. The ways they struggle look different at different times, because what’s happening in their lives changes, and their behaviours change in response. Sometimes their circumstances suit them. Sometimes they don’t.

1. Caspi A, Houts RM, Ambler A, et al. (2020). Longitudinal Assessment of Mental Health Disorders and Comorbidities Across 4 Decades Among Participants in the Dunedin Birth Cohort Study. JAMA Network Open, 3(4), e203221. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2764602

2. Moffitt TE, Houts R, Asherson P, et al. (2015). Is Adult ADHD a Childhood-Onset Neurodevelopmental Disorder? Evidence From a Four-Decade Longitudinal Cohort Study. American Journal of Psychiatry, 172(10): 967-77. https://psychiatryonline.org/doi/10.1176/appi.ajp.2015.14101266

3. Sibley MH, Kennedy TM, Swanson JM, et al. (2024). Characteristics and Predictors of Fluctuating Attention-Deficit/Hyperactivity Disorder in the Multimodal Treatment of ADHD (MTA) Study. Journal of Clinical Psychiatry, 85, :24m15395.

The youngest girl in class is up to 70% more likely to be diagnosed with ADHD than the oldest girl in class. She’s also 77% more likely to be prescribed medications like Ritalin. That’s in Canada.

It’s similar for boys. The youngest boy in class is 63% more likely to be diagnosed with ADHD and 73% more likely to be medicated. That’s in Taiwan. But it’s similar around the world - Australia, Brazil, Germany, Israel, Norway, Turkey, and the USA have all found the same thing.

Why would that be? How could the relative ages of children influence an ADHD diagnosis?

In ADHD research this difference between the youngest and oldest children is called the Relative Age Effect (RAE). Despite using different methods to research the RAE, studies agree that

being“relatively younger within the school year is a risk factor for an ADHD diagnosis”.

The Relative Age Effect is very consistent in ADHD research. It’s been found in large population studies involving millions of children, in clinical samples assessing the relative immaturity of children with ADHD diagnoses, and in studies comparing teacher ratings of ADHD symptoms in their pupils. The RAE is most clear in younger age groups and has become stronger over the years that it has been studied. The RAE is much less marked in countries where it is common to let younger children in a class stay another year in that class.

The Relative Age Effect is not caused by seasonal factors or birth months. It is found in both the northern and southern hemispheres, where school calendars follow their own distinct patterns. Furthermore, when the cutoff for starting school is changed, the birth month of the children most likely to receive ADHD diagnoses changes too.

Finally, the Relative Age Effect is not found in other areas of children’s mental health. It is only evident for ADHD.

So how can we explain the Relative Age Effect, and why does it only appear in the diagnosis and medication of childhood ADHD?

Share

Why is there a Relative Age Effect?

The simplest explanation is that we are diagnosing immaturity as ADHD.

Children born at the end of a school year are up to a year younger than their classmates born at the beginning of a school year. In a classroom, this age gap means the youngest child is relatively immature compared with the oldest child. For six-year olds, the oldest child in class has had almost a year longer - 15% of a lifetime - to learn and practise skills needed in a classroom.

The Relative Age Effect suggests that social factors influence which children are seen as ‘having’ ADHD. And that children who are young in their school year may be diagnosed inappropriately if teachers, parents and clinicians mistake their immaturity for ADHD.

Confusion about the cause of inattentive or active behaviours may cause harm, as ADHD diagnosis has been shown to lead to poorer outcomes. However, the Relative Age Effect is rarely acknowledged or discussed by ADHD authors.

Textbook silence

Despite extensive documentation that relative age is a significant risk factor for an ADHD diagnosis, there’s a resounding silence about it in textbooks, articles and guidelines. In a sample of 43 academic textbooks, none of the sections on ADHD referred to the Relative Age Effect. Most clinicians don’t seem to know about it, and parents certainly don’t.

Moreover, no guidelines or textbooks recommend that teachers and clinicians be aware of the potential for age-related immaturity to be misdiagnosed as ADHD, and to adjust their practices accordingly.

Where does this leave us?

The effect of relative age on ADHD diagnosis is consistent and significant. The structure of schools, and the way adults understand children’s behaviour, are influencing ADHD prevalence rates worldwide, especially for younger ages.

Parents and teachers need to know about the effect of relative age on ADHD diagnosis. And when evaluating children for possible ADHD, clinicians should consider the relative age of the child and the potential impact of their classroom environment on teacher and parent reports of ADHD symptoms.

Because of the risks of diagnosis itself, clinicians need to be especially cautious about ADHD diagnosis for mild difficulties. For these children, any harms associated with an ADHD diagnosis or psychiatric drug use, are more likely to outweigh any possible benefits.

Our schools need to provide children of all ages with environments, structures, timetables and teaching methods appropriate for their level of development and their way of being in the world — without requiring a diagnosis.

Subscribe now

So if you have ADHD, surely it’s best to get a diagnosis?

Apparently not.

In 2022 Australian Luise Kazda and her colleagues published research that found teenagers who were diagnosed with ADHD fared worse than teenagers exhibiting the same behaviours, but not diagnosed with ADHD.

The Longitudinal Study of Australian Children identified 393 adolescents who had been diagnosed with ADHD in earlier years, and matched them with 393 similar adolescents who had shown the same ADHD behaviours but had not been given a formal diagnosis of ADHD at the time. The adolescents who had been given a formal diagnosis felt less accepted and included at school, said they did worse academically, displayed more negative social behaviours, and were more likely to harm themselves. On no measure did the diagnosed teenagers rate themselves as better than the undiagnosed teenagers.

Kazda’s finding was the opposite of her expectation that Australian teenagers would benefit from a diagnosis of ADHD and the consequent treatment. In fact although the teenagers’ behaviours had been the same, those diagnosed with ADHD did worse.

This study would be intriguing, even if it were the only one signalling that worse outcomes from ADHD diagnoses were possible. But there are others: an Irish study by Cliodhna O’Connor and Fiona McNicholas found the same effect.

When O’Connor and McNicholas analysed data from Growing Up in Ireland – the National Longitudinal Study of Children, they found that by age 13, those who had been given an ADHD diagnosis at 9 years old were doing worse than those who had the same behaviours at 9, but who had not been given a diagnosis. Diagnosed children showed more emotional and peer relationship problems, displayed less prosocial behaviour, and reported poorer self concept. For example, they believed they caused trouble and couldn’t comply with adult expectations.

We need to pay attention.

Here we have similar results from two independent longitudinal studies, examining different people in different places. In both studies, ADHD diagnoses resulted in worse outcomes for the children the diagnoses were supposed to help. In contrast, no longitudinal study shows that children do better with ADHD diagnoses than without.

Can we trust these studies?

As has been documented extensively, such as here and here, ADHD research is at high risk of bias. So we need to look carefully at the quality of the studies and possible sources of bias.

Firstly, these studies did not receive industry funding, which we know affects the design, outcomes and recommendations of mental health research. The lack of industry funding makes the findings likely to be reliable.

Secondly, Kazda’s team predicted:

that children and adolescents [would] benefit from an ADHD diagnosis and that this may result in better Quality of Life.

They expected that the diagnosed children would do better - but found the opposite. And still, they published the surprising outcome.

Publishing this unexpected result is important in itself because of psychology’s "publication bias". Research findings that are predicted or show a substantial difference are more likely to be published than studies with unexpected or non-significant results. Unexpected and non-significant outcomes are often ‘filed away’ without publication. This creates a systemic bias in published research. When writers summarise what we know, they can only ‘see’ what is published. They can’t ‘see’ the studies sitting in researchers’ filing cabinets. And so summaries don’t include all studies that have been done, leading to false conclusions. Kazda’s publication of a study with unexpected outcomes strengthens the assurance we can have in its findings.

Thirdly, one author of the Irish study, Fiona McNicholas, has received industry funding and has published research in line with common ADHD views. She has been a member of a Shire Pharmaceuticals advisory board and has received funding from Ortho-McNeil-Janssen Pharmaceuticals, although not specifically for this study. Despite these industry connections, McNicholas still published a result that didn’t match the expectations of the ADHD orthodoxy. In this case, sharing results that don’t fit the normal narrative, increases the confidence we can have in this study.

So we should take the results of these studies seriously because unlike most ADHD research, in these studies there is little risk of bias.

Why might diagnosis be harmful?

Although both Kazda’s and O’Connor’s studies indicate there are risks for children diagnosed with ADHD, they can’t explain how diagnosis has this effect. Why would diagnosis make children’s lives worse rather than better? These studies don’t answer this question directly. But some possibilities are:

1. Firstly, the diagnosed group of children might be different from the not-diagnosed children in some way that isn’t yet obvious. This is unlikely because both studies looked extensively but didn’t find differences between the groups at the point of diagnosis - aside from their ADHD diagnosis status. Instead, the studies found that children became different between diagnosis and being re-evaluated as teenagers.

2. Another possibility is that labelling children’s attention and activity levels as “disorders” - that is, giving children ADHD diagnoses - leads those children to develop negative views of themselves or to understand themselves as broken. We already know that biological explanations of mental illness are associated with increased stigma or negative attitudes towards mental illness and can dissuade people from seeking help in order to avoid labels and stigma. In contrast, teenagers who drop self-labels have been found to have improved self esteem. So, diagnosis may lead to worse outcomes because of the meaning a person makes of being diagnosed, labelled or marked ‘different’ from peers.

3. Lastly, ADHD diagnoses might change how parents, teachers and others think about and treat children, contributing to poorer outcomes in diagnosed children. It’s possible that children internalise negative responses or lowered expectations from others as a result of diagnosis, contributing to a cycle of lowered self-worth and achievement.

Where does this leave us?

In the long term, children are better off without an ADHD diagnosis - even if they show all the behaviours that could lead to a diagnosis. The two longitudinal studies are reliable, but don’t explain why diagnosis makes things worse for diagnosed teenagers.

How does this information help us? Based on these findings:

• Parents, teachers and clinicians need to know that there are risks from ADHD diagnoses.

• Clinicians should be cautious about diagnosing ADHD in children, especially when behaviours are mild. Any harms associated with an ADHD diagnosis need to be carefully weighed with the potential benefits.

• Parents and clinicians can address concerns about children’s behaviours without seeking diagnoses, by supporting parents to help their children and by supporting children to understand themselves and what helps them thrive.

• Clinicians, teachers and schools can support children who exhibit ADHD behaviours to thrive in school.

Laura Batstra and Sami Timimi’s wise words are a good place to close. Perhaps many mental “disorders”, including ADHD, would be better thought of as variation and difference between people. Labelling variation as deficit and disorder can make things worse. Maybe we can support people better by talking them through the effects of this variation on themselves and others, and helping them and their communities decide whether and how they want to do accommodate or change it.

1. Kazda L, McGeechan K, Bell K, et al. (2022). Association of Attention-Deficit/Hyperactivity Disorder Diagnosis With Adolescent Quality of Life. JAMA network open, 5(10), e2236364. https://doi.org/10.1001/jamanetworkopen.2022.36364

2. O'Connor C, & McNicholas F. (2020). What Differentiates Children with ADHD Symptoms Who Do and Do Not Receive a Formal Diagnosis? Results from a Prospective Longitudinal Cohort Study. Child psychiatry and human development, 51(1), 138–150. https://doi.org/10.1007/s10578-019-00917-1

I’m a psychologist. I listen a lot, to a lot of people. They talk about being human. Their struggles. Their joys. About their thoughts, their feelings and their doings.

Attention Deficit Hyperactivity Disorder

There’s a lot of talk on the street about ADHD. But the conversation seems hollow, lopsided, full of gaps.

Like many psychologists, I’m frustrated by the gap between how the world out there is talking about ADHD, and what we know. I want you to know what we know about ADHD. I want you to have easy access to clear information.

Join me

Here at The ADHD Files you’ll find key questions about ADHD, with straightforward answers. I’m going to share important research with you so that your conversations about ADHD can be richer and more careful. Let’s talk.

Subscribe now