Field Notes

concept

ADHD relative-age effect

StatusStableReviewdue UpdatedStands on3 sourcesSupportsnotes

The ADHD relative-age effect is the finding that a child’s position within their school cohort predicts their chance of being diagnosed with ADHD: the youngest children in a year group are diagnosed substantially more often than the oldest, on a difference of a few months of age and nothing else. The effect is well replicated across countries and study designs, strongest for diagnoses made before age 10, and increasing in recent cohorts. Sayal and colleagues put the excess at 26 to 31 percent comparing the youngest with the oldest in a year.1

What the effect represents

For years the gradient was ambiguous between two mechanisms. The relatively young might be overdiagnosed, their ordinary immaturity read as pathology; or the relatively old might be underdiagnosed, their relative maturity masking genuine ADHD. Both predict the same gradient. Two lines of evidence separate them.

The additional cases are mild. Kazda and colleagues’ scoping review of 334 studies applies a five-question framework for establishing overdiagnosis and concludes that ADHD is overdiagnosed and overtreated in children and adolescents, locating the problem at a specific margin: “for individuals with milder symptoms in particular, the harms associated with an ADHD diagnosis may often outweigh the benefits.”2 Eleven of the twelve relative-age studies it reviewed found the youngest more likely to be diagnosed. An excess concentrated among mild presentations is the signature of overdiagnosis; better case-finding among the relatively old would instead add cases across the severity range.

The gradient is not specific to ADHD. Root and colleagues followed 1,042,106 UK children and found the youngest quarter of the school year more likely to be diagnosed not only with ADHD (1.36 times) but also with intellectual disability (1.30 times) and depression (1.31 times) — three unrelated diagnoses, three near-identical gradients.3 An underdiagnosis mechanism specific to ADHD cannot produce a parallel gradient in intellectual disability. A general mechanism can.

The third mechanism the binary misses

Framing this as overdiagnosis or underdiagnosis omits a possibility the same evidence supports: that being youngest in the year causes genuine difficulty rather than only the appearance of it. Root and colleagues raise it for depression, noting that relative youth is associated with poorer academic performance and poorer peer relationships, either of which could produce real symptoms rather than misread ones.

On that reading the youngest children in a cohort are both overdiagnosed and genuinely worse off, and the two are not competing explanations. It changes the remedy. If the gradient were pure misattribution, age-norming the diagnostic criteria would fix it. If part of it is caused difficulty, age-norming would correct the label while leaving the disadvantage in place, and the intervention would have to change what schools demand of the youngest children in a room.

Bearing on the sex gap

Boys are diagnosed with ADHD at roughly twice the rate of girls, and it is tempting to read the relative-age effect as part of that explanation. It is not, at least not directly. Sayal and colleagues found the effect similar in magnitude for both sexes: an incidence ratio of 1.26 for boys and 1.31 for girls. What it establishes is that the diagnostic process is sensitive to developmental position within a cohort — the mechanism a developmental account of the sex gap would need — without showing that the mechanism produces that gap. Male variability, education, and life outcomes uses the effect as one link in the chain from behaviour to grades to diagnosis, and it is the clearest documented case in that chain of a school-structural feature converting into a clinical label.

  • Find whether any jurisdiction has implemented age-normed ADHD diagnostic criteria, and what happened to the gradient and to treatment rates.
  • Establish whether the increasing trend across cohorts tracks rising academic demands in early years, which would test the caused-difficulty channel directly.

  1. Sayal, Chudal, Hinkka-Yli-Salomäki, Joelsson, and Sourander, “Relative age within the school year and diagnosis of attention-deficit hyperactivity disorder: a nationwide population-based study”, The Lancet Psychiatry 4, no. 11 (2017). ↩

  2. Overdiagnosis of ADHD in children and adolescents (Luise Kazda, Katy Bell, Rae Thomas, Kevin McGeechan, Rebecca Sims, and Alexandra Barratt, “Overdiagnosis of Attention-Deficit/Hyperactivity Disorder in Children and Adolescents”, JAMA Network Open 4, no. 4 (2021): e215335). Synthesises 334 studies, 273 of them primary research. Cited from the open-access full text; the publisher blocks automated retrieval, so the wiki holds no local snapshot. ↩

  3. Root, Brown, Forbes, Bhaskaran, Hayes, and colleagues, “Association of Relative Age in the School Year With Diagnosis of Intellectual Disability, Attention-Deficit/Hyperactivity Disorder, and Depression”, JAMA Pediatrics 173, no. 11 (2019), a population cohort of 1,042,106 children aged 4 to 15. ↩

Built on 3 sources (3 external).

Working out connections…