Braithwaite wonders why we state that ‘the personality trait’ ADHD does not disappear in adulthood, while in our study it seems that ADHD does gradually lessen with age. A first comment is that ADHD is not a personality trait, but a neuropsychiatric disorder. It has an early onset and symptoms do persist into adulthood. Our study aimed to assess whether this also extends into later life and we found that this was indeed the case in 2.8–4.2% of those examined. We agree that the prevalence rates found over the lifespan decrease a little and in our study we found lower prevalence rates among the oldest old. However, the prevalence rates we found are substantial and, if replicated, this would mean that ADHD is by no means limited to children or to younger adults.

Routh & Jackson rightly point out some limitations of the paper. The first point is that we did not rule out any other DSM-IV diagnosis, and the second pertains to the limited validity of recollecting childhood memories in old age. We agree that both points are important and have discussed them in the Discussion of the paper. A third point is that according to Routh & Jackson we have found no evidence of impairment of ADHD in old age, which might be taken as evidence of limited validity of our measurement of ADHD. Only few studies have been conducted in older adults with ADHD and those studies did find impairments. ^{Reference Henry and Jones1–Reference Kooij, Buitelaar, Van den Oord, Furer, Rijnders and Hodiamont3} In our study, those diagnosed with ADHD did report lifelong impairment in four of the five areas of functioning assessed, which is substantially more than DSM-IV requires for the diagnosis. ^{Reference Michielsen, Semeijn, Comijs, Van de Ven, Beekman and Deeg4}

We agree that other psychiatric disorders may explain impairment and that the study would have been stronger if psychiatric comorbidity had been assessed. However, as the diagnosis of ADHD requires not only a current but a lifelong history of the typical symptoms, we think we have been able to discriminate from disorders with a later onset. Mild cognitive decline or dementia is indeed very impairing and an important health problem in older age. Although we did not diagnose these disorders, we did exclude respondents with a low score and/or persistent cognitive decline on the Mini Mental State Examination. Except for three excluded persons, all respondents were able to answer the questions of the interview. Therefore it is very unlikely that respondents with dementia were included in our study.

The conclusion that ‘there are greater and more relevant issues in older age that need to be tackled before we start inventing any new diagnoses’ seems ill founded. We agree that it is wise to be conservative in proposing new psychiatric diagnoses which may add to the ever increasing numbers of patients eligible for mental health treatments. However, ADHD is not a new diagnosis and it is extremely unlikely that it ceases to be active at any particular age.