ADHD Linked to Higher Later-Life Cognitive Impairment Risk

Key Takeaways

• Among older adults in the U.S. All of Us cohort, cognitive impairment occurred in 6.7% with diagnosed ADHD versus 3.4% without ADHD.
• Diagnosed ADHD was associated with both mild cognitive impairment or dementia and dementia alone; the dementia-only association remained statistically significant.
• Adults in the highest ADHD polygenic-risk quintile had a statistically significant 35% higher risk of incident cognitive impairment than those in the lowest quintile.

Cognitive aging in adults with attention-deficit/hyperactivity disorder (ADHD) raises a question: whether longstanding attention difficulties relate to later cognitive decline or other factors explain the overlap. An ADHD diagnosis reflects a documented clinical history, whereas inherited liability captures a different aspect of susceptibility. For adults approaching later life, the distinction matters when interpreting cognitive symptoms because dementia can emerge gradually.

In Pascoal and colleagues' JAMA Psychiatry ADHD and cognitive impairment analysis, investigators retrospectively examined 187,341 U.S. All of Us participants aged 50 or older, with a mean follow-up of 10.26 years. All had electronic health record (EHR) and genomic data. Algorithms requiring multiple diagnoses or medication records identified ADHD and the combined endpoint of mild cognitive impairment or dementia. Investigators compared participants with and without diagnosed ADHD for that endpoint and for dementia alone. A separate comparison examined the highest and lowest quintiles of ADHD polygenic-risk scores derived from genome-wide association data. Covariates included sex at birth, self-identified race and ethnicity, cohort-entry year, and apolipoprotein E ε4 status.

For incident mild cognitive impairment or dementia, diagnosed ADHD was associated with a hazard ratio (HR) of 4.60 compared with no ADHD diagnosis (95% confidence interval [CI] 3.99–5.29; P<0.001). This estimate concerns the combined endpoint, not dementia alone.

For dementia alone, diagnosed ADHD was associated with an HR of 3.07 compared with no ADHD diagnosis (95% CI 2.12–4.44; P<0.001). This narrower endpoint excludes mild cognitive impairment.

The highest ADHD polygenic-risk quintile was associated with incident cognitive impairment relative to the lowest quintile (HR 1.35; 95% CI 1.25–1.46; P<0.001). This genetic-risk comparison differs from the analysis of recorded ADHD diagnoses and does not identify a mechanism linking ADHD liability to impairment.

Further adjustment for comorbidities, educational attainment, and healthcare use weakened but did not eliminate the diagnosis-based association. EHR-defined diagnoses may miss some clinical presentations. Residual confounding, possible reverse causation during dementia's long prodrome, and selection bias in the volunteer cohort also limit interpretation and generalizability. Pascoal and colleagues interpreted the diagnosis-based and polygenic-risk associations as pointing in the same direction, but neither establishes ADHD as a cause of cognitive impairment; whether ADHD treatment modifies risk remains unanswered.

Clinician Questions

Which racial and ethnic groups were represented in the All of Us ADHD and cognitive impairment cohort?

The U.S. All of Us cohort of older adults with ADHD and comparison participants included self-identified white, Black, and Hispanic participants; white participants were the largest reported group. Volunteer enrollment limits how confidently this composition can be generalized beyond the cohort.

What mental health and healthcare-use outcomes were associated with ADHD among adults who developed cognitive impairment?

Among All of Us participants who developed mild cognitive impairment or dementia, those with comorbid ADHD reported poorer mental health, quality of life, and social satisfaction and had greater healthcare use than those without comorbid ADHD. These associations do not demonstrate an effect of ADHD or its treatment.