Attention‐deficit/hyperactivity disorder in old age and the risk of dementia: A national cohort study in Israel

Abstract

AbstractBackgroundEvidence of the association between Attention‐Deficit/Hyperactivity Disorder (ADHD) and the risk of dementia is scarce, inconsistent, and subject to potential sources of bias that are as yet unconsidered. Our objective was to examine the association between ADHD and dementia risk.MethodA prospective national cohort study born 1933‐1952, without a dementia diagnosis for two years, were followed up for incident dementia from 2002‐2020. Information was from electronic health records in a healthcare maintenance organization in Israel. ADHD diagnosis was classified as present from the age of the first diagnosis, otherwise absent. Competing risk Cox regression models were fit to quantify the risk of incident dementia with the Hazard Ratio (HR) and their associated 95% Confidence Intervals (CI) with inverse probability weights, age as the underlying time‐scale, unadjusted and, in the primary analysis, adjusted by 18 covariates.ResultOf 110,196 participants aged 50‐70 years in 2002, 56,914 (51.6%) were female. During follow‐up there were a total of 746 (0.7%) cases of ADHD, and 8,339 (7.6%) of incident dementia. The percentage of incident dementia was 13.1% (N = 98) among those with ADHD present and 7.5% (N = 8,241) among those with ADHD absent. In the primary analysis, the group with ADHD was statistically significantly (P<.05) associated with a higher risk of dementia (unadjusted: HR = 3.57, CI = 2.89, 4.42; adjusted HR = 2.73, CI = 2.04, 3.65) compared to the group with ADHD absent. Eight of ten sensitivity analyses replicated the statistically significant association observed in the primary analysis. Psychostimulant medication offset the association between ADHD and dementia risk to null, and evidence of reverse causation was mild.ConclusionADHD is associated with increased dementia risk. Policymakers, caregivers, patients, and clinicians may wish to monitor ADHD in old age reliably.