Adults over 50 with ADHD had more than four times the risk of dementia or mild cognitive impairment

Those raw rates put the gap at about double: a ratio of 1.97, or 3.3 extra cases per 100 people ^[13][14]. The fourfold figure is an adjusted hazard ratio. The team calculated it after controlling for age, biological sex, cardiovascular conditions and socio-demographic measures ^[5]. Adjustment moved the estimate up. The university's account does not give the number of participants with an ADHD diagnosis or the length of follow-up, and the size of that diagnosed group sets how tightly the ratio can be estimated.

The design targets one objection in particular. Smaller registries in Scandinavia and Europe had already suggested that adults with ADHD were unusually vulnerable to neurodegenerative disease ^[1]. Clinicians were divided over whether early Alzheimer's or vascular dementia was simply being misdiagnosed as late-emerging adult ADHD ^[1]. The eight-year buffer answers that from the clinic side ^[7]. The polygenic scores answer it from the genome side. A clinician's label cannot change the variants a person inherited, so a genetic association is hard to explain as a charting error ^[8].

"This doesn't prove that ADHD causes dementia, but seeing the same pattern show up independently in medical records and in people's genes gives us greater confidence that the association is not simply an artifact of how ADHD gets diagnosed," said senior author Tharick A. Pascoal, an associate professor of psychiatry and neurology at the University of Pittsburgh School of Medicine ^[10][12]. "Two different kinds of evidence are pointing in the same direction," he said ^[11].

The university's own summary goes further than he does. It describes the genetic result as "demonstrating that the link is rooted in shared biology rather than diagnostic overlap" ^[16]. I think the evidence supports Pascoal's narrower reading. Agreement between records and genes makes a pure labelling artifact unlikely.

The thing this doesn't tell you is the route from ADHD variants to cognitive decline. The same Pittsburgh account notes that ADHD persists across adult life with cumulative neurobiological and behavioural consequences ^[2]. Shared biology could therefore mean a direct effect on the ageing brain, or an indirect one through what builds up alongside ADHD over decades. A study built on correlations between health records and genotypes cannot separate the two ^[4].

Using ADHD as a dementia-screening flag is a further step the data do not yet support. By the raw rates, 93.3% of people with an ADHD diagnosis did not develop dementia or mild cognitive impairment during the study ^[15]. A higher relative risk in a group where most people stay well is a reason for attention when symptoms appear. Routine screening would also need a trial showing that earlier detection in this group changes outcomes.