A case-control study involving 255 small ruminants herds randomly selected was carried out in Portugal between January and December 2004, to identify risk factors associated with brucellosis seropositivity. To achieve this objective, two groups of herds selected according their prevalence status were compared: "cases" (farms with seroprevalence higher than 5%, n=123) and "controls" (farms seronegatives, n=132). A carefully structured questionnaire was used to collect data from each herd. A statistical analysis to compare "case" versus "control" herds was performed with the variables obtained from the questionnaire and the seroprevalence results. The effects on seroprevalence of several variables such as: individual characteristics; farm management practices; farm characteristics; animal health; knowledge and characteristics of farmers were evaluated. Data were analysed using logistic regression. Univariable analysis was used to screen the variables used in the logistic regression model. Nine variables were associated with brucellosis seropositivity in univariable analysis p<0.10. These variables were retained for multivariable logistic regression model. Regression model identified five variables as risk factors for seropositivity. The odds of brucellosis were increased: herds with more than 116 animals (OR=2.99); in herds with no cleaned-watering places (OR=3.05); in herds with insufficient manure removal and insufficient cleaning of premises (OR=2.87); in introduction of animals from non-free brucellosis herds or from herds of unknown status (OR=12.11). In the other hand, farmers' age (the eldest) was related to decreased odds (OR=0.4). Potential risk factors identified in this study were consistent factors associated with brucellosis seropositivity and support current recommendations for the control of brucellosis. Considering the paucity of epidemiological reports on brucellosis in the Northeast of Portugal and the absence of any data concerning factors related to either the prevention or the spread of the disease, our results could make a useful contribution towards the prevention of small ruminants brucellosis in the area.