Abstract Background Substance dependence is considered an international health issue and studies that access the characteristic of this population are required to develop public health programs for harm reduction. To this aim, we purpose to to identify, in a population undergo addiction treatment, if the use of substances leads to sexual risky behavior and also check if another variables influences in this behavior. Method Observational study of clinical sample collected of adult patients seeking treatment to dependence of alcohol and cocaine. The data collected were: socio-demographic, substance use, sexual behavior and childhood abuse. Risky sexual behaviors were considered: inconsistent use of condoms and presence of multiple sexual partners in the past six months. An exploratory analysis of the association between the variable “risky sexual behavior” and the other variables was performed using Pearson’s chi-square, followed by a multivariate logistic regression analysis. Results After analyzing the variables with the presence or absence of risky sexual behavior, were included in the logistic regression the data that presented association with sexual risk behavior, and age appears as an independent factor for risky sexual behaviors. Other factors, such as schooling and unemployment, influence as much as the use of substances in risky sexual behaviors. Conclusion Attributing just to the substance use this risky sexual behavior seems too simplistic. Other structural factors such as schooling, work, age and sexual abuse in childhood can influence decision making for safe sex.