In 2023, the Centers for Disease Control and Prevention put a number on something that clinicians had been describing anecdotally for two years. Roughly 15.5 million American adults had an ADHD diagnosis. More than half of them, 55.9 percent, had been diagnosed in adulthood. The figure did not represent a population that had only recently developed a neurodevelopmental condition. It represented a population whose ADHD had, for various reasons, not been identified until adulthood. The question of who those people are, and what the diagnostic surge reflects, is more complicated than any single narrative can capture.

The numbers themselves are not contested. A CDC report by Staley and colleagues, published in Morbidity and Mortality Weekly Report in October 2024, established the 6 percent adult prevalence figure and the 55.9 percent adult-diagnosed share. A retrospective cohort study by Erick Messias and colleagues at St. Louis University, published in Psychiatric Research and Clinical Practice in March 2025, showed that adult ADHD diagnoses rose 15 percent between 2020 and 2023, reversing a decline of nearly 11 percent in the previous four years. Komodo Health data analyzed in 2024 showed that stimulant prescriptions for adults aged 20 to 39 rose 30 percent between 2020 and 2022, and non-stimulant ADHD medication prescriptions in the same group rose 81 percent. Adult women’s diagnoses doubled between 2020 and 2022. Between 2007 and 2016, before the pandemic-era surge, diagnoses in adult women had already risen 344 percent, compared to 264 percent for adult men.

The interpretation is where the discourse fractures. Three competing narratives have settled into the public conversation about what these numbers mean. The first holds that the surge represents a real and overdue correction, the long-delayed identification of populations whose ADHD was missed by a diagnostic system calibrated for childhood hyperactivity in boys. The second holds that the surge represents overdiagnosis, a medicalization of ordinary distractibility driven by social media, telehealth-only diagnostic mills, and a culture that increasingly seeks psychiatric labels for everyday struggle. The third, which appears less often in coverage, holds that the diagnostic category itself is doing too much work, that the behavioral criteria for ADHD cannot reliably distinguish between several distinct populations whose presentations look the same but whose underlying conditions are different.

All three are at least partly right. The article that follows is an attempt to take each seriously.

What the Headlines Get Right

The case that the surge represents long-overdue identification of missed populations is supported by the strongest data and by the most coherent mechanism.

Women have been systematically underdiagnosed with ADHD for decades. The childhood diagnostic apparatus, calibrated for externalizing behaviors like classroom disruption, fidgeting, and impulsive outbursts, catches boys more reliably than girls. Girls with ADHD are more likely to present with the inattentive subtype, which produces internal struggles, daydreaming, organizational chaos, slow processing, rather than the behaviors that prompt teacher referrals. They learn to mask. They develop compensatory strategies that work passably until the cognitive demands of adulthood overwhelm them. When they finally arrive at evaluation, often in their thirties or forties, the diagnosis is correct and the ADHD has been present all along.

The 344 percent rise in women’s adult diagnoses between 2007 and 2016 is not a surge produced by the pandemic. It is a surge that was already underway when the pandemic arrived and accelerated it. The pandemic-era doubling of women’s adult diagnoses between 2020 and 2022 is the continuation of a longer correction, not the beginning of a new phenomenon. Decades of research documents that the population of adults newly diagnosed with ADHD as adults is disproportionately female and disproportionately presenting with the inattentive subtype that pediatric screening misses.

The pandemic mechanism, where it operates, is not that the pandemic caused ADHD. It is that the pandemic stripped away the structural supports that many adults with undiagnosed ADHD had been quietly relying on. Office routines that imposed external accountability. Commutes that forced specific transitions. Coworkers whose presence shaped attention. Children’s school schedules that organized the household day. When these structures collapsed in 2020 and 2021, the adults who had been compensating for executive function challenges through environmental scaffolding lost their scaffolding. The struggle that had been manageable while the structures held became unmanageable when they fell away. Many of those adults sought evaluation for the first time.

Braun Tueller, a physician’s assistant at the Huntsman Mental Health Institute, described the dynamic in a 2024 interview with University of Utah Health. “In the past two weeks alone, more than half of the outpatient patients I have treated have had ADHD as a chief complaint or as a co-occurring diagnosis,” he said. “That’s an enormous number of people who are struggling with focus and concentration.” The clinical experience is consistent with the population data: pandemic-era conditions exposed undiagnosed ADHD in adults who had been managing it without knowing they had it.

Telehealth changed the access picture. According to the CDC, about one in five adults with ADHD received their diagnosis through telehealth, and about half have used telehealth for ADHD care. Before 2020, adult ADHD evaluation typically required an in-person appointment with a psychiatrist or neuropsychologist, often after long waitlists, in office hours, in geographic locations many adults could not easily access. After 2020, the same evaluation could happen from a kitchen table during a lunch break. The access barrier dropped. The diagnostic rate rose. These are causally connected.

The most authoritative response to the “overdiagnosis surge” framing came in a 2025 systematic review from researchers at King’s College London, which examined whether ADHD prevalence had actually risen between 2020 and 2024. It had not. The underlying rate of adults meeting diagnostic criteria appeared stable. What had changed was the rate at which adults meeting those criteria were being identified. That is the technical signature of an identification gap closing, not of disease inflation.

What the Headlines Get Wrong

The skeptical counter-narrative, in its most common form, blames TikTok. The argument runs that the surge in adult ADHD diagnoses is driven by young adults watching short-form social media content describing ADHD symptoms in vague, relatable terms, recognizing themselves in those descriptions, seeking evaluation, and receiving diagnoses they would not have received under stricter clinical practice. The narrative typically points to the explosion of ADHD-related content on TikTok during the pandemic, the documented poor quality of much of that content, and the rise of telehealth services accused of providing rapid diagnoses without rigorous evaluation.

The narrative has elements of truth and significant problems.

It is true that ADHD-related content on TikTok exploded during the pandemic. A frequently-cited analysis found that fewer than half of the symptom claims in the top 100 TikTok videos about ADHD aligned with DSM-5 diagnostic criteria. ADHD content on the platform has been reported to have generated more than 36 billion views. Some clinicians have reported that a substantial share of new evaluation requests are arriving from patients who explicitly cite social media as the prompt.

It is also true that telehealth-only diagnostic services have, in some cases, operated under business models that prioritized throughput over diagnostic rigor. Several high-profile telehealth ADHD services have faced regulatory scrutiny for diagnostic practices that fell below clinical standards. The concern that some diagnoses are being made without adequate evaluation is real.

What the skeptical narrative gets wrong is the causal direction. The claim that TikTok caused the surge is undermined by the data showing that the surge in women’s diagnoses was already in progress before TikTok existed, that it accelerated during a pandemic that was characterized by exactly the kind of structural collapse that exposes undiagnosed ADHD, and that underlying prevalence is not rising. TikTok did not invent ADHD. What it appears to have done is make adults who already had ADHD aware that what they were experiencing had a name. That awareness then drove evaluation requests. Evaluation, when conducted by competent clinicians, often confirmed what the adult had suspected.

The concern about diagnostic mills is more substantive but more narrow. It applies to a specific subset of telehealth providers operating under particular business models, not to telehealth ADHD care in general. Most adults receiving ADHD diagnoses through telehealth are seeing the same kind of clinicians they would see in person, in evaluations of comparable rigor, with the difference being only the modality of delivery. The blanket “telehealth caused overdiagnosis” framing collapses a real but narrow problem into a sweeping critique that the data does not support.

The overdiagnosis claim also runs into a basic arithmetic problem. The current adult diagnostic rate of 6 percent is below the best estimate of underlying adult prevalence, which the most recent global meta-analyses place at around 6.76 percent. If anything, the population is still underdiagnosed in aggregate, not overdiagnosed. The headline numbers on the surge are catching the system up to where it should already be, not pushing it past where it should land.

What the Headlines Miss Entirely

The third population, the one the public discourse rarely names, is the most clinically interesting and the hardest to talk about without being misread.

The DSM-5 criteria for ADHD are behavioral. They describe what a person does and what they struggle to do. They do not describe what is happening in the brain. They do not, and cannot, distinguish between behaviors that reflect underlying neurological differences and behaviors that reflect a person who never learned the skills the behaviors require.

The distinction matters. Executive function is not purely a property of the brain. It is also a set of skills that develop through practice over the first two and a half decades of life. Children who navigate age-appropriate challenges, manage their own logistics, tolerate frustration, plan multi-step tasks, recover from failure, and absorb the consequences of their own decisions develop the executive function skills that those experiences are designed to build. Children who do not, do not. This is not a marginal claim in developmental psychology. It is the rationale behind structured early childhood curricula like Tools of the Mind. It is the foundation of the research showing that executive function predicts adult outcomes more reliably than IQ. The skills emerge through challenge. Without challenge, the skills do not emerge.

A young adult who arrives at age twenty-four having never managed their own schedule, never absorbed the cost of forgetting an obligation, never had to plan a multi-week project independently, never sat with a frustrating task long enough to find the breakthrough on the other side, will exhibit a recognizable set of behaviors. They will have difficulty initiating tasks. They will struggle with time estimation. They will get overwhelmed by multi-step projects. They will procrastinate, then panic, then produce work below their capacity. They will lose things. They will forget appointments. They will feel scattered. These are the behaviors that anchor the DSM-5 ADHD criteria. A clinician interviewing this young adult will record them honestly. The young adult will recognize them in themselves. The diagnosis is, by behavioral criteria, correct.

Whether the underlying condition is ADHD in the neurological sense is a different question, and one the diagnostic interview is not equipped to answer.

The phenomenon has several names in the broader discourse. Lawnmower parenting describes parents who systematically remove obstacles from their child’s path. Snowplow parenting is a similar metaphor. Helicopter parenting refers more specifically to constant monitoring and intervention. The common thread is the suggestion that a generation of parents, motivated by genuine care, has been managing the executive demands that their children were supposed to be developing the skills to manage themselves. The children, in this account, arrive at adulthood with the executive function skills of much younger people, simply because the developmental conditions that build those skills were not present.

The argument extends beyond parenting. Sociological research has documented that young adults today reach traditional adult milestones, financial independence, household management, sustained employment, significantly later than prior generations. Some of this is economic. Some of it is cultural. Some of it may be developmental. A 26-year-old who has never lived independently, never managed a budget without parental backstop, never absorbed the full consequences of a missed deadline, is in a different developmental position than a 26-year-old of forty years ago. Their executive function profile, measured behaviorally, will look different. Whether it reflects neurological difference, skill underdevelopment, or some combination is a question current diagnostic tools cannot easily answer.

This is not an argument that ADHD is not real. The condition is well-established, the biology is increasingly understood, the medication response is consistent with what neurological accounts of the condition would predict. The argument is more specific. It is that the population of adults who meet behavioral criteria for ADHD almost certainly includes several distinct subpopulations whose underlying conditions are different. Some have neurological ADHD that was missed in childhood. Some have ADHD that emerged or intensified due to specific adult stressors. And some, plausibly, have skill deficits that developed under conditions that did not allow the skills to emerge, whose presentations look like ADHD because the behaviors are the same even when the cause is not.

The clinical implication is uncomfortable. A diagnosis cannot reliably distinguish between these subpopulations. Stimulant medication may produce short-term improvement in all three groups, because the mechanism of stimulants, as recent research has shown, is to elevate arousal and reward predictiveness in any brain. The improvement may not mean the same thing in each population. For adults with neurological ADHD, medication offsets a real underlying difference that will persist throughout the lifespan. For adults whose presentation reflects skill underdevelopment, medication may produce performance improvement while the underlying skill gap remains, and the same person might benefit more durably from skill-building interventions that medication does not provide.

The honest framing is that the diagnostic surge captures a heterogeneous population, and that flattening that heterogeneity into either a celebration or a panic obscures the more interesting question of who the surge contains and what each population actually needs.

What the Discourse Should Not Do

Several misreadings of the skill-deficit argument are common and worth heading off.

The argument is not that ADHD does not exist. It is well-established, biologically grounded, and consistent with decades of research. The argument is that the behavioral diagnostic category captures more than one underlying condition, and that the field’s diagnostic tools cannot easily distinguish between them.

The argument is not that women’s adult diagnoses are skill deficits in disguise. The female adult diagnosis cohort is dominated by people whose ADHD was systematically missed in childhood by a screening apparatus calibrated for male presentations. That cohort needs the diagnosis it has been denied. The skill-deficit argument is about a different and more recent population, primarily younger adults whose developmental conditions were different from prior generations, not about women whose ADHD has been masked for decades.

The argument is not that medication is being overprescribed. The medication research is consistent and the clinical efficacy is robust. The argument is that medication may be doing different things in different populations, and that adults whose presentation reflects skill underdevelopment may also benefit from interventions, coaching, structured skill-building, environmental scaffolding, that medication does not substitute for.

The argument is also not an argument against neurodiversity. Neurodiverse adults whose ADHD has been correctly identified deserve the supports that come with that identification, and the cultural shift toward affirming rather than pathologizing neurodivergence has been substantively positive. The skill-deficit argument is compatible with that shift. It says only that the category of “behaviors that look like ADHD” probably contains more than one population, and that good clinical practice will eventually need to distinguish them.

The Telehealth Question

The role of telehealth in the diagnostic surge is worth examining separately, because it is the area where the most legitimate clinical concerns and the most overheated public criticism collide.

The case for telehealth ADHD care is strong. Access to evaluation has historically been a major barrier for adults who suspect they have ADHD. Specialists are concentrated in urban areas. Waitlists for in-person evaluations have routinely been six months to two years. Time off work for daytime appointments excludes adults whose employment does not accommodate it. The geographic and scheduling barriers have systematically excluded the populations whose ADHD was already most likely to be missed: lower-income adults, adults in rural areas, adults working jobs without flexibility, adults whose family responsibilities precluded daytime evaluation.

Telehealth flattened these barriers. An evaluation could happen on a phone, on a laptop, during a lunch break, on a weekend. The population of adults who could realistically pursue evaluation expanded substantially. That expansion is part of why diagnoses rose. It is also part of why the rise was disproportionately concentrated in populations that had previously been underdiagnosed.

The case against certain telehealth practices is also real. Some telehealth services have been documented operating under business models that incentivized rapid diagnosis and stimulant prescription with limited evaluation rigor. Several high-profile cases of regulatory action against telehealth ADHD services have established that the concern is not hypothetical. Patients seeking evaluation should distinguish between telehealth providers who are delivering the same quality of evaluation they would in person, with the difference being only modality, and providers whose business model is built on rapid throughput.

The practical distinction worth drawing is between telehealth as a delivery mechanism and telehealth as a diagnostic mill. Telehealth as a delivery mechanism is good clinical practice in modern form. Telehealth as a diagnostic mill is a specific business-model failure that bears on a subset of providers. Conflating the two, which much of the skeptical coverage has done, mischaracterizes both.

The Drug Shortage Backdrop

The diagnostic surge in adults has happened alongside a sustained shortage of stimulant medications. Adderall and its generics have been on the FDA shortage list for nearly four years. Vyvanse, Ritalin, and other stimulant formulations have faced intermittent shortages of varying severity. According to CDC data, more than two-thirds of adults taking stimulant medication for ADHD reported difficulty filling their prescription in 2023 because of supply issues.

The shortage is sometimes cited as evidence of overdiagnosis. The argument runs that the demand for stimulants has exceeded the supply because too many people are receiving prescriptions, and that the shortage demonstrates a system that has lost its calibration. The argument is partly true and partly misleading.

It is true that demand has risen sharply. It is also true that the rise has produced strain on a manufacturing system that was not designed to scale rapidly. What the argument misses is that the demand could be rising for legitimate reasons, the long-overdue identification of adults whose ADHD had been missed, while still producing supply pressure that the manufacturing system cannot easily absorb. Legitimate demand can exceed supply. The shortage does not, by itself, demonstrate that the demand is illegitimate.

The shortage has also fallen hardest on populations that the diagnostic surge has most plausibly helped. Women newly diagnosed in adulthood. Adults in rural areas who finally accessed evaluation through telehealth. Adults whose ADHD was identified through workplace difficulties that the pandemic exposed. The shortage has, for these populations, been a second barrier laid on top of the first one they had just gotten past. The supply problem is a manufacturing and distribution problem, not an evidence-of-overdiagnosis problem.

What Adults Considering Evaluation Should Know

For adults reading this article who have been wondering whether to pursue evaluation, the practical implications are several.

The diagnostic surge is real, the underlying prevalence appears stable, and the surge is concentrated in populations that the previous diagnostic system was failing. If you are an adult whose ADHD was likely missed, particularly if you are a woman, if you are a person of color, if your childhood ADHD looked more like internal struggle than external disruption, the case for seeking evaluation is strong. The supports that an accurate diagnosis can unlock, medication if appropriate, accommodations, behavioral interventions, are meaningful. The cost of not pursuing evaluation, of continuing to attribute decades of struggle to character flaws rather than to a treatable cognitive difference, is substantial.

The evaluation itself should be rigorous regardless of modality. A competent ADHD evaluation, in person or over telehealth, includes a structured clinical interview, a review of developmental history, validated rating scales completed by the patient and ideally by someone who knew the patient in childhood, and screening for conditions that can mimic ADHD or co-occur with it. Evaluations that consist of a fifteen-minute phone call followed by a stimulant prescription are not what good clinical practice looks like, and the prescription that follows such evaluations does not represent a reliable diagnosis. Adults seeking evaluation should ask about the structure of the evaluation before committing.

The medication question, if a diagnosis is confirmed, is a clinical conversation with a prescribing clinician, not a settled outcome of the diagnosis itself. Medication helps many adults with ADHD substantially. Medication is not the only intervention, and for some adults, particularly those whose presentation may reflect skill underdevelopment rather than neurological ADHD, behavioral coaching, structured skill-building, and environmental design may produce more durable benefit than medication does. A good clinician will discuss all of the available options rather than assuming the diagnosis automatically leads to a prescription.

For adults whose ADHD-pattern struggles have emerged primarily in adulthood, particularly in their twenties and early thirties, the diagnostic question is more complicated. The clinical interview cannot easily distinguish between adult-onset cases of underlying ADHD that had been compensated in earlier life, and adult-emerging patterns that reflect skill underdevelopment under different developmental conditions. Both can be real. Both can benefit from support. The kind of support that helps most may differ. The honest answer is that the field’s diagnostic tools are not yet able to make this distinction well, and that good clinical practice involves staying curious about what is actually going on rather than treating the diagnosis as a final answer.

What Comes Next

The discourse about the adult ADHD diagnostic surge has not yet settled into accurate description. The two dominant framings, celebration of overdue identification and panic about overdiagnosis, both contain real elements but neither captures the full picture. The verified data points clearly toward identification catching up to prevalence, with women and historically underserved populations driving most of the rise. The skeptical concerns about diagnostic rigor and the specific role of certain telehealth practices are also real, but narrower than the coverage typically suggests. And the third population, the adults whose behavioral presentation may reflect skill underdevelopment rather than neurological ADHD, is part of what the surge contains but is rarely named in either framing.

The field’s path forward, if it wants to do better than the current discourse, requires several distinct conversations. A conversation about how to ensure that telehealth-delivered ADHD evaluations meet clinical standards regardless of business model. A conversation about how to develop diagnostic tools that can distinguish between neurological ADHD and behaviorally similar presentations with different underlying causes. A conversation about what kinds of interventions, beyond medication, are most appropriate for which subpopulations. And a conversation about how to keep the cultural shift toward affirming neurodivergence from collapsing into a one-size-fits-all framing that obscures real clinical heterogeneity.

None of these conversations is happening at scale. The public discourse is mostly two competing narratives shouting past each other. The substantive work, when it happens, will happen in clinical practice, in research that the headlines do not cover, and in conversations between adults seeking to understand themselves and clinicians equipped to help them figure out which population they are part of.

The surge is real. The underlying prevalence is stable. The populations being identified are heterogeneous. The system catching up to them is uneven. None of this fits comfortably into a headline. All of it is part of what is happening.