Risk and needs assessment of children with conduct-disordered behaviour has been hampered by a lack of validated and easy-to-use decision-support tools. In this study, we investigated reliability and concurrent validity of the Swedish translation of the Early Assessment Risk List for Boys (EARL-20B). Parents to 76 boys aged 6-12 years old, referred to child psychiatric clinics, completed questionnaires about child symptoms and were thereafter administered the EARL-20B. Good to excellent inter-rater agreement and moderate concurrent validity was found. We also investigated possible subgroups of children based on the assessment with the EARL-20B. Cluster analysis yielded a tentative four-cluster solution, where subgroups of boys differed with respect to external measures. Although further study of predictive validity and usefulness for treatment planning is needed, the findings suggest that the EARL-20B is a promising tool for structured assessment of boys presenting with antisocial behaviour in clinical child psychiatry.