AIMS: To establish the prevalence of intoxication before sex and its association with risky sexual behavior. DESIGN: The data were from the 2006 Uganda Demographic and Health Survey which had been designed for a cross-sectional descriptive study. SETTING: The study covered the whole country-Uganda. PARTICIPANTS: The respondents were 6,253 women and 1,804 men who had ever had sex. MEASUREMENTS: The key independent variable was intoxication before last sexual intercourse while the major outcome variables were condom use and sex with non-regular partners. Weighted prevalence of intoxication was computed and multivariable logistic regression was applied to assess the independent association of intoxication with risky sexual behavior. FINDINGS: Twelve percent of men and 16% of women reported having been intoxicated before last sexual intercourse. Of the women who reported intoxication before last sexual intercourse, 78% said it was their partners who were intoxicated. Of the men who reported intoxication, 83% said it was they themselves who were intoxicated. Intoxication of the men was associated with having sex with non-regular partners (OR=1.78, 95%CI: 1.04-3.03) and having unprotected sex (OR=1.71, 95%CI: 1.07-2.73). Women who were intoxicated were less likely to have been with non-regular partners (OR=0.55, 95%CI: 0.32-0.95). The women whose partners were intoxicated before last sexual intercourse were more likely to report having had unprotected sex (OR=1.55, 95%CI: 1.12-2.15). CONCLUSION: HIV prevention mechanisms should address intoxication before sex. More effort is needed to find ways of helping women avoid unprotected sex with intoxicated partners.