Introduction: The relationship between violent offences, mental disorder and substance-use disorders has been widely analyzed but has produced contradictory results. Studies examining this relationship in prison populations are scarce. Objectives: The aim of the study was to analyze the relationship between violent crime, substance-use disorders and mental disorder. Method: This is a descriptive, cross-sectional, epidemiological study of 707 male prisoners. Socio-demographic, clinical and penal data were collected by trained interviewers. Penal data were confirmed using penitentiary records. The clinical version of the Structured Clinical Interview (SCID 1) for DSM-IV Axis I Disorders was used for diagnosis of Axis I mental disorders (including substance-use disorders). Inmates who have lifetime substance-use disorders were classified by type and number of substances used. We considered violent offences: homicide and attempted homicide, aggravated assault, common assault, robbery, threatening behavior, harassment, arson and any sexual offence. Results: Violent offences in inmates who used drugs (n=370) were more prevalent than in inmates who did not (84.6% vs.15,3%, p<0.0001).The risk of committing a violent crime is double for people using a substance compared to those who do not and those who are not repeat offenders (OR= 2.03 CI 95%: 1.08-3.78). This risk increases when considering repeat offenders and those who use more than one substance . The presence of a mental disorder turns out not to be significant (OR= 1.38 CI 95%: 0.98-1.95). Conclusions: In our study, the greatest risk factors for committing a violent offence were being a re-offender and using more than one substance. Mental disorder was not found to be a risk factor for violent offences.