The revision of the attention-deficit hyperactivity disorder (ADHD) age-of-onset criterion (criterion B) in the DSM-5 has been widely discussed, particularly the extension of the onset threshold from age 7 to age 12. ^{Reference Riglin, Blakey, Langley, Thapar, Agha and Smith1,Reference Sanders, Thomas, Glasziou and Doust2} Less attention, however, has been paid to the implications of removing functional impairment from the age-of-onset criterion in adult ADHD assessments. For instance, Sanders et al ^{Reference Riglin, Blakey, Langley, Thapar, Agha and Smith1} concluded that the change lacked robust empirical justification and that its impact on prevalence and diagnostic validity remains insufficiently studied, and Riglin and colleagues ^{Reference Sanders, Thomas, Glasziou and Doust2} have even questioned the validity of the age-of-onset criterion per se.

However, the specific issue of the effects of removing the functional impairment component from the age-of-onset criterion in adults has not been directly addressed. To clarify its nature, it is useful to examine the change in wording in the DSM-5. The DSM-IV-TR required that: ‘Some hyperactive-impulsive or inattentive symptoms that caused impairment were present before age 7 years’. ³ DSM-5 revised this wording to: ‘Several inattentive or hyperactive-impulsive symptoms were present prior to age 12 years’, ⁴ without linking those symptoms to impairment. Meanwhile, the functional impairment criterion (criterion D) still requires that symptoms interfere with or reduce the quality of social, academic or occupational functioning. ⁴ Importantly, impairment is required at the time of assessment.

This revision may create a potential bias. If I am assessed at age 11 or younger, the functional impairment criterion falls within the developmental period specified by the age-of-onset criterion. In this case the onset of the presence of clinically significant impairment overlaps with the age-of-onset criterion. In other words, if my symptoms emerge according to the age-of-onset criterion, before age 12, but are not associated with functional impairment, then a diagnosis cannot be issued. However, if I am assessed later in life, as an adult, for instance at age 30, I need only demonstrate current impairment (criterion D), alongside retrospective evidence of onset of symptoms before age 12 (criterion B), without showing that those childhood symptoms were causing impairment during the age range specified in the age-of-onset criterion. Strictly following the DSM-5 criteria, I could therefore be diagnosed with ADHD as an adult but fail to meet the criteria if assessed in childhood. And this is the concern raised here.

ADHD is defined as a neurodevelopmental disorder and neurodevelopmental disorders are described as conditions with onset in the developmental period, characterised by developmental deficits that produce impairment of personal, social, academic or occupational functioning. ⁵ This implies that symptoms must be developmentally inappropriate and impairing within that developmental window. If, when assessing adults, the age-of-onset criterion requires only symptom presence before age 12, without evidence that those symptoms were associated with impairment at that time, then what is being established retrospectively may be better understood as ADHD-like behaviours, potentially reflecting an aetiology and neurobiology that are distinct from those underlying ADHD.

The distinction is not semantic. It concerns whether the diagnosis is anchored in dysfunction at the onset or in the later emergence of a symptom pattern that is interpreted, retrospectively, as continuous with childhood. In practice, this approach risks operationalising adult ADHD as a present-tense symptom syndrome, rather than as a condition associated with impairment within the developmental window specified by the age-of-onset criterion. If the former interpretation is adopted, the condition being operationalised may diverge from the original neurodevelopmental construct of ADHD. Even describing adult ADHD as the persistence of a childhood-onset disorder becomes difficult to justify, since developmental continuity would no longer be necessary.

This concern, however, may be theoretical, as competent mental health clinicians generally evaluate childhood functional impact during adult assessments. Clinicians recognise that ADHD is classified as a neurodevelopmental disorder, and within DSM-5, such disorders are defined by developmentally inappropriate symptoms that produce clinically meaningful effects at onset. In addition, criterion A specifies that inattention and/or hyperactivity-impulsivity must interfere with functioning or development. It follows that within the DSM framework, the term ‘symptom’ implicitly entails such interference. Hence, when applying the age-of-onset criterion in adults, the competent clinician should consider whether early symptoms were associated with interference during that developmental period.

Thus, removing impairment wording does not make childhood impairment irrelevant but assumes it is assessed within a comprehensive adult ADHD assessment. If impairment at the age of onset is not clearly present, the competent clinician will also assess how internal and external factors modulate symptom expression and impact, including scaffolding, compensatory mechanisms and strategies that can minimise and even mask impairment. However, this assumption warrants thorough examination in the context of contemporary adult ADHD assessment practices. In current healthcare systems there has been a rapid expansion of adult ADHD diagnostic services. Assessment is increasingly delivered across diverse professional backgrounds, including clinicians who may not have specialist training in mental health, let alone in neurodevelopmental psychopathology. Where the diagnostic text does not explicitly anchor impairment to the age-of-onset criterion, there is a risk that childhood symptom presence alone may be interpreted as evidence of neurodevelopmental onset.

The concern raised here is not that DSM-5 permits diagnosis of ADHD without impairment, as criterion D clearly requires clinically significant current interference. Rather, the issue is whether the current structure adequately safeguards the developmental integrity of the diagnosis when applied in adult populations. Since ADHD is considered a neurodevelopmental disorder, evidence of impairing symptoms at onset should remain central even in adult diagnostic assessments. If impairment is not apparent, then the competent clinician must clearly document the evaluation and presence of compensatory and masking strategies and behaviours that may have contributed to obscuring it.

Further research is needed to determine whether removing impairment from criterion B has altered severity, diagnostic thresholds, prevalence and persistence in adult ADHD populations. The key issue is that impairment, or, alternatively, compensatory and masking strategies, must be evaluated within the developmental age range specified by the onset criterion. Whether this is consistently done in practice remains uncertain.