Stimulant drug users have a greater prevalence of risky driving behaviour. This study aimed to assess how far this association remains after adjusting for aggressiveness. Cross-sectional interview study assessing associations between measures of risky driving behaviours as outcomes, measures of stimulant drug use as predictors and a measure of aggressiveness as a covariate. United States. Data were drawn from wave 3 (2012-13) of the National Epidemiological Survey on Alcohol and Related Conditions (NESARC-III) (n=36 309 aged ≥18years). Stimulant drug use, past-year DSM-5 stimulant use disorder, aggression and measures of risky driving were assessed using face-to-face interviews conducted using the Alcohol Use Disorder and Associated Disabilities Interview Schedule (AUDADIS-5) and the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5). Overall, 2714 (8.3%) respondents indicated life-time stimulant use, and 112 (0.3%) met criteria for past-year DSM-5 stimulant use disorder. More than 10% of ongoing stimulant users and one-third of respondents with DSM-5 stimulant use disorder reported stimulant-specific driving under the influence of drugs (DUID) in the past-year (both P<0.0001). Adjusted for demographics and independent of aggression, life-time stimulant users reported increased likelihood of driving [adjusted odds ratio (aOR)=3.00, 95% confidence interval (CI)=2.63-3.42] or speeding under the influence of drugs (aOR=3.39, 95% CI=3.01-3.82) and licence revocation (aOR=2.16, 95% CI=1.87-2.50) (all P<0.0001). Past-year DSM-5 stimulant use disorder was associated with all outcomes (aOR=5.48, 95% CI=2.95-10.18 and aOR=3.87, 95% CI=2.23-6.70, respectively, all P<0.0001), except licence revocation (aOR=1.72). Stimulant use appears to be positively associated with risky driving behaviours after adjusting for aggressiveness.