ADHD and memory problems after 50 — what a 187,000-person study does and doesn't show

Bottom line

Among 187,341 US adults aged 50 and over, those with an ADHD diagnosis in their health records went on to develop mild cognitive impairment or dementia more often: 6.7% versus 3.4% over a mean of about 10 years. The headline hazard ratio is 4.60. A second measure in the same people, genetic risk for ADHD, gives a much smaller estimate: 1.35 for the highest fifth versus the lowest.

The research

Bauer-Negrini and colleagues used the US All of Us Research Program, which links health records to genome data. Of 187,341 participants (mean age 56, 57% female), 3,026 (1.6%) had an ADHD diagnosis on record. 203 of them (6.7%) developed mild cognitive impairment or dementia, compared with 6,259 (3.4%) of those without ADHD. In adjusted models the hazard ratio was 4.60 (95% CI 3.99–5.29). Adjusting for other conditions, education and how often people used health care made the association smaller but did not remove it.

The authors added a genome-wide polygenic risk score for ADHD because record-based diagnoses have a known problem: people who see doctors more get diagnosed with more things, including both ADHD and memory complaints. A genetic score doesn’t depend on who saw a doctor. It still found a link, with the top fifth of genetic risk at a hazard ratio of 1.35 (1.25–1.46). Among people who did develop cognitive impairment, those with ADHD also reported worse mental health, quality of life and social satisfaction.

Limits worth carrying. Only 1.6% of this cohort had an ADHD diagnosis on record, so the people counted as having ADHD may be the more impaired or more comorbid end, diagnosed because they were already in contact with services. ADHD itself causes attention and memory complaints, which can be recorded as mild cognitive impairment. The abstract does not report whether ADHD medication changed the risk.

What this means

If you were diagnosed late and are reading this in your fifties, the 4-fold figure is not your personal risk: it is a relative number, from a group that may be sicker than adults with ADHD generally, and the genetic analysis found a much smaller link. What the study supports is practical: if memory or thinking changes at midlife, raise them rather than assume they are your ADHD as usual, and ask for an assessment that separates long-standing ADHD from a new change.

Read next on this site:

• Working memory — the everyday memory problem ADHD causes, which a cognitive assessment has to tell apart from decline.
• Late diagnosis — for people diagnosed in midlife or later.
• Is it ADHD or something else? — the conditions that overlap with ADHD symptoms.