Introduction: Despite the great advances in serological testing for transfusion-transmitted infections, the selection of blood donors by blood bank operators remains the only way to avoid transmission within the testing window period. Part of this selection is the self-exclusion form, on which the donors can exclude their blood from donation without any explanation. This study assessed the clinical and epidemiological characteristics related to positivity for viral hepatitis and to the use of the confi dential self-exclusion (CSE) form. Methods: This transversal study analyzed the data collected from blood donors' fi les in a hospital in Southern Brazil. Univariate and multivariate analyses identifi ed the clinical and epidemiological variables related to positive serologies of viral hepatitis and to whether the donor was self-excluded. Results: Of the 3,180 donors included in this study, 0.1% tested positive for HBsAg, 2.1% for anti-HBc, and 0.9% for anti-HCV. When the 93 donors with positive serologies for viral hepatitis were compared with those who were negative, a greater proportion of the positive serology group was found to have had a history of blood transfusions (OR=4.908; 95%CI=1.628 -14.799; p<0.01), had repeatedly donated (OR=2.147; 95%CI=1.236 -3.729; p<0.01), and used the CSE form for self-exclusion (OR=7.139; 95%CI=2.045 -24.923; p<0.01). No variables were independently associated with self-exclusion. Conclusions: A history of blood transfusion, repeated donations, and self-exclusion are factors that should be considered during viral hepatitis screenings in blood banks.