Background: The link between aggression and mental disorders has been the focus of diverse studies in persons with and without intellectual disabilities (ID). Because of discrepancies in the finding of studies in persons with ID to date, and because of differences in research design, instruments used and the population studied, more research is needed. The purpose of this study was to delineate any significant association between certain psychiatric disorders and specific domains of aggressive behaviours in a large sample of persons with ID controlling for sex, age, autism and degree of ID. Method: Data from the present study were obtained from 47% of all persons with ID receiving services from New York State agencies, using the Institute for Basic Research-Modified Overt Aggression Scale (IBR-MOAS between 2006 and 2007). The IBR-MOAS was completed by the chief psychologists of 14 agencies based on information from the participants' files. Demographic information obtained included the psychiatric diagnosis made by the treating psychiatrist as well as information on age, sex and degree of ID. Data from 4069 participants were analysed. Results: Impulse control disorder and bipolar disorder were strongly associated with all five domains of aggressive behaviour in the IBR-MOAS. Psychotic disorder was highly associated with four domains except for physical aggression against self (PASLF), which was of borderline significance. Anxiety was most associated with PASLF and verbal aggression against self (VASLF); depression with VASLF; obsessive compulsive disorder with physical aggression against objects (PAOBJ); personality disorders with verbal aggression against others (VAOTH), VASLF and PASLF; and autism with physical aggression against others (PAOTH), PAOBJ and PASLF. Mild to moderate ID was associated with VAOTH and VASLF and severe to profound ID with PAOBJ and PASLF. Female sex was most associated with VASLF. Conclusions: Impulse control, mood dysregulation and perceived threat appear to underlie most of the aggressive behaviours reported. Psychosis and depression appeared to have been over-diagnosed in persons with mild to moderate ID and under-diagnosed in persons with severe and profound ID. These findings replicate and extend findings from previous studies. The pattern of associations reported can be used as helpful indicators by professionals involved in the treatment of aggressive behaviours in persons with ID.