Aims: 1) To investigate whether genetic liability to attention-deficit/hyperactivity disorder (ADHD), indexed by polygenic risk scores for ADHD (PRS-ADHD), is associated with substance use disorders (SUD) in individuals with ADHD. 2) To investigate whether other individual- or family-related risk factors for SUD could mediate or confound this association. Design: Population-based cohort study. Setting and participants: ADHD cases in the iPSYCH sample (a Danish case-cohort sample of genotyped cases with specific mental disorders), born in Denmark between 1981 and 2003 (N = 13 116). Register-based information on hospital diagnoses of SUD was available until December 31, 2016. Measurements: We estimated odds ratios (ORs) with 95% confidence intervals (CIs) for any SUD as well as for different SUD types (alcohol, cannabis, and other illicit drugs) and severities (use, abuse, and addiction), with effect sizes corresponding to a comparison of the highest PRS-ADHD decile to the lowest. Findings: PRS-ADHD were associated with any SUD (OR = 1.30, 95% CI: 1.11–1.51). Estimates were similar across different types and severity levels of SUD. Other risk factors for SUD (male sex, age at ADHD diagnosis, comorbid conduct problems, and parental factors including SUD, mental disorders, and socio-economic status) were independently associated with increased risk of SUD. PRS-ADHD explained a minor proportion of the variance in SUD (0.2% on the liability scale) compared to the other risk factors. The association between PRS-ADHD and any SUD was slightly attenuated (OR = 1.21, 95% CI: 1.03–1.41) after adjusting for the other risk factors for SUD. Furthermore, associations were nominally higher in females than in males (OR_females = 1.59, 95% CI: 1.19–2.12, OR_males = 1.18, 95% CI: 0.98–1.42). Conclusions: A higher genetic liability to attention-deficit/hyperactivity disorder appears to be associated with higher risks of substance use disorders in individuals with attention-deficit/hyperactivity disorder.