Diagnostic Broadening Is Not the Same as “Everyone Has ADHD Now”

Over recent decades, ADHD and autism diagnoses have increased sharply. This has produced two predictable responses.

The first is concern that something new is happening in society, biology, schooling, parenting, technology, or environment.

The second is the lazy public refrain that “everyone has ADHD now.”

The second response is more emotionally satisfying than useful. It allows people to dismiss rising diagnosis rates without having to understand them. It also allows them to collapse very different explanations into one blunt accusation: overdiagnosis.

A recent JAMA Psychiatry paper by LaBianca and colleagues offers a more careful way to think about this. The authors examined whether the genetic risk profile of people diagnosed with ADHD or autism has changed over time. Their finding was that people diagnosed more recently had, on average, lower polygenic risk scores for ADHD or autism than people diagnosed earlier.

That sounds simple. It is not.

The paper does not show that ADHD or autism are becoming less real. It does not show that recent diagnoses are fake. It does not give us a genetic test for ADHD or autism. It does something more specific: it suggests that the diagnosed population has changed.

That is where the conversation should begin.

First, the genetics does not mean genetic determinism

The paper uses polygenic risk scores, or PRS. These scores summarize many genetic variants, each contributing a tiny amount of statistical risk. ADHD and autism are both polygenic. There is no single “ADHD gene” or “autism gene” hiding in the genome waiting to be subpoenaed.

This matters because polygenic risk is not destiny. A higher PRS may indicate higher inherited liability at a group level, but it does not determine whether a person will meet diagnostic criteria. A lower PRS does not mean the person is not ADHD or autistic. It means the common genetic variants currently associated with that diagnosis are less strongly represented, on average, in that person or group.

That “on average” is doing a lot of work.

Polygenic scores are useful for research because they allow comparison across groups. They are weak as individual diagnostic tools because they do not capture the whole person. They do not measure masking, school fit, trauma, sleep, poverty, parenting, culture, chronic stress, teaching style, sensory load, gendered expectations, or the quiet misery of looking functional while running on fumes.

So when this paper speaks of “genetic contribution,” it is not saying that genes determine ADHD or autism. It is saying that common genetic liability, measured statistically, appears to contribute less to the average profile of more recently diagnosed groups than it did to earlier diagnosed groups.

That is a population-level inference. It is not a clinical verdict.

Second, the paper is not offering a biomarker assessment

This is important because the biomarker fantasy has a long shelf life. Every few years, someone gets excited about the possibility that ADHD or autism might finally be diagnosed through a scan, blood test, genetic score, or some other device that can rescue clinical judgment from the inconvenience of human complexity.

This paper does not do that.

The authors used polygenic scores to compare diagnostic cohorts across time. Their logic was that genetic risk is relatively fixed across a person’s life, while diagnostic practice changes. If the diagnosed population changes over time, the average genetic risk profile of that diagnosed population might change too.

That is not the same as saying a PRS can diagnose ADHD or autism.

The study used ten genetic profiles: ADHD, autism, bipolar disorder, major depressive disorder, schizophrenia, addiction, educational attainment, IQ, neuroticism, and risk-taking. The point was not to locate a single genetic signature. The point was to see whether the pattern across these scores better matched different explanations for rising diagnosis rates.

The authors compared three broad possibilities.

The first was a lowering threshold, meaning that diagnosis has expanded to include milder or more heterogeneous presentations.

The second was shifting diagnostic boundaries, meaning that people previously diagnosed under one category may now be diagnosed under another.

The third was better detection, meaning that clinicians are catching people who always met the same diagnostic profile but were previously missed.

Their results fit the first explanation best: diagnostic broadening.

But diagnostic broadening is not the same as diagnostic failure.

What “diagnostic broadening” actually means

Diagnostic broadening means the category has expanded.

Earlier ADHD diagnosis was heavily shaped by the disruptive, externally visible, childhood-male presentation. The child who could not sit still. The child who interrupted. The child who caused classroom management problems. The child whose difficulty was inconvenient enough for adults to notice.

That was never the whole picture.

As ADHD understanding has changed, clinicians have become more aware of inattentive presentations, emotional dysregulation, executive dysfunction, internal restlessness, late diagnosis, female presentations, adult ADHD, and the impact of masking. The person who is not disruptive may still be impaired. The child who stares out of the window may be struggling as much as the child who climbs out of it.

They are just less administratively annoying.

The same applies to autism. Earlier autism diagnosis was narrower, often tied to more visible childhood presentations. Broader recognition has included people with less obvious support needs, different social camouflaging patterns, co-occurring mental health difficulties, and late-identified presentations.

So if diagnostic categories broaden, the average profile of diagnosed people changes.

That is not scandalous. It is exactly what one would expect when a field moves from a narrow stereotype to a wider clinical reality.

Why the “everyone has ADHD now” argument is lazy

The claim that “everyone has ADHD now” sounds like critique, but it is usually just impatience wearing a lab coat.

It takes a real trend — rising diagnosis — and turns it into a sneer. It treats increased recognition as proof of exaggeration. It assumes that because more people are being diagnosed, the diagnosis must be losing meaning.

That is possible in some cases. Overdiagnosis can happen. Misdiagnosis can happen. Diagnostic incentives can distort clinical practice. In some systems, access to support may depend on a formal diagnosis, which can increase diagnostic pressure.

But those are specific concerns. They require specific evidence.

The phrase “everyone has ADHD now” does not provide evidence. It provides dismissal.

The LaBianca paper does not support that dismissal. It supports a more careful conclusion: the diagnostic net has widened, and the people now receiving diagnoses may include a broader range of presentations than those diagnosed earlier.

That distinction matters.

A wider net can catch people who were previously missed.

A wider net can also catch some people who may not belong in the category.

Both can be true.

The responsible response is not cynicism. It is better assessment.

The paper’s finding cuts both ways

For advocates, this paper is useful but not comfortable.

It supports the argument that diagnostic practice has changed. That change likely reflects broader awareness of ADHD and autism across age, sex, presentation, and impairment profile. This is especially relevant for late-diagnosed adults, inattentive ADHD, girls and women, and people who were never disruptive enough to be taken seriously.

That is the good news.

The harder part is that broadening creates clinical responsibility. If the diagnosed group becomes more heterogeneous, then intervention must become more nuanced. A broader ADHD diagnosis cannot be met with a thinner understanding of ADHD.

Medication may help many people, but it cannot be the entire model. Coaching may help many people, but it cannot replace careful diagnosis. Accommodations may be essential, but they need to match the person’s functional impairment. Psychoeducation matters, but it cannot become a substitute for meaningful support.

A broader diagnostic category demands better formulation, not just more labels.

Lower genetic risk does not mean lower impairment

One of the easiest misreadings of the paper is this: if recently diagnosed people have lower average ADHD or autism PRS, then their ADHD or autism must be milder, less valid, or less real.

That is not justified.

Genetic risk scores capture one part of liability. They do not capture impairment directly. They do not capture the fit between a person and their environment. They do not capture the gap between internal effort and external performance. They do not capture the years of compensating until the compensation collapses.

Someone can have lower measured common-variant genetic liability and still experience serious impairment. A person’s functional difficulty is not determined by PRS alone.

This is especially important in adult ADHD. Many late-diagnosed adults were missed not because they were unaffected, but because they were compliant, intelligent, anxious, perfectionistic, ashamed, or able to construct fragile systems that worked until life became too complex.

Their difficulty was not absent. It was deferred.

Broadening is not automatically overdiagnosis

Overdiagnosis is a real issue. It should not be dismissed. If people are diagnosed without sufficient impairment, without developmental history, without differential diagnosis, or because ordinary distress is being medicalised, that is a problem.

But broadening is not the same as overdiagnosis.

Broadening can mean that the field has corrected for earlier blind spots.

It can mean that clinicians now recognise that ADHD is not always loud.

It can mean that autism is not always externally obvious.

It can mean that adult presentations are being taken seriously.

It can mean that internal distress is no longer ignored simply because it was not disruptive to other people.

The question is not whether diagnosis has broadened. The paper suggests that it has.

The question is whether broadening has improved clinical accuracy, diluted it, or done both at the same time.

The honest answer is probably both.

That is less satisfying than a slogan, but slogans are where nuance goes to die.

What this means for clinical practice

The implication is not that we should become more suspicious of everyone seeking diagnosis.

The implication is that we should become more precise.

A good ADHD assessment should not merely count symptoms. It should examine developmental history, impairment, context, compensatory strategies, co-occurring conditions, sleep, anxiety, trauma, substance use, learning differences, sensory issues, and environmental demands.

It should ask not only “does this person meet criteria?” but “what is driving this person’s impairment?”

That question matters because two people can meet ADHD criteria for different reasons, present differently, and require different forms of support.

A diagnostic label may open the door. It does not tell us how to build the room.

What this means for advocacy

For ADHD advocacy, this paper offers a useful line of argument.

We do not need to deny rising diagnosis rates.

We do not need to pretend that every diagnosis is perfect.

We do not need to defend careless practice simply because bad-faith critics exist.

Instead, we can say something more grounded:

ADHD diagnosis has broadened because the old model was too narrow. That broadening has helped identify people who were previously missed. It may also create risks of overdiagnosis or shallow formulation. The solution is not to retreat into scepticism. The solution is to improve diagnostic literacy, clinical assessment, and support planning.

That is the distinction the public conversation keeps missing.

Diagnostic broadening is not proof that “everyone has ADHD now.”

It is evidence that the old diagnostic lens was incomplete.