Previous work has shown that adolescent hyperactivity patients are significantly more likely to smoke than controls. To determine whether this pattern persists in adults, we studied a series of 71 patients (55 males, 16 females; mean age +/- SD: 33.9 +/- 11.4 years) diagnosed with ADHD. Of the males, 23 (42%) were current smokers, 7 (13%) were ex-smokers, and 25 (45%) were never smokers. Comparable figures for males in the general population in 1991, unselected for ADHD, were 28.1%, 29.1%, and 42.1%, respectively. Of the females, 6 (38%) were current smokers, 5 (31%) were ex-smokers, and 5 (31%) had never smoked, as compared with 23.5%, 19.0%, and 57.6%, respectively, in the general population. Quit ratio (percentage of ever-smokers who were ex-smokers) was 29% for ADHD patients, compared with 48.5% in the general population. The discrepancy was accounted for by the males, whose quit ratio was 23%, compared with 51.6% in the general population; the figure for ADHD females (45%) was similar to that in the general population (44.7%). Smokers recalled experiencing a significantly higher number of ADHD symptoms (11.5 +/- 1.7) as children than never smokers (9.9 +/- 2.3; p < .01) and scored significantly higher on several indices of childhood and adult comorbidity. Our findings suggest that ADHD patients overinclude smokers, and that these smokers find it extremely difficult to quit. For ADHD smokers, smoking may have begun as an attempt to manage deficits in attention and concentration, as suggested by greater childhood symptomatology in these patients.