This research focuses on the use of the condom among injector drug users and the identification of their knowledge on HIV prevention that can inform interventions. A cross- sectional study was conducted, using respondent driven sampling (RDS); the sample was composed of 540 participants, of 18 years of age or older. The survey included counseling for HIV testing and the prevention of sexually transmitted diseases. In the multivariate analysis estimated a logistic regression model (explaining how to use the condom) multinomial with raw and adjusted OR. The proportion of drug-injector users who always used condoms during sexual intercourse with their partners was 21.1%. It was found that by adjusting the risk of not using a condom by ten inadequate assumptions on HIV prevention, factors that increase the risk are sharing the injection equipment (ORaj = 3. 19 [1. 29-7.88]), having unsafe sex practices (ORaj = 3. 74 [1. 21-11. 58]) and not increasing the use of condom despite having knowledge of HIV (ORaj = 33. 59 [9. 19-122. 86]). The low use of condoms among intravenous drug users is associated with inadequate knowledge about HIV prevention.