Background: Couples Voluntary Counseling and Testing (CVCT) is an effective HIV prevention strategy for couples who represent the largest risk group in Africa. In this region, less than 1% of couples tested for HIV. Of the estimated 100 infections per day in many African countries 70% occurs in cohabiting couples. In Ethiopia, there are only few couple based VCT studies conducted. Narrowing the gap of knowledge in these matters will help as a baseline for other studies. Furthermore it helps in forming strategies and specific interventions to deal with drawback of CVCT program. Objectives: The aim of this study was to determine the magnitude of couples coming to VCT centers and to assess the factors associated with use and nonuse of CVCT services in Addis Ababa. Methods: A cross sectional study in 6 government hospital VCT sites in Addis Ababa city was conducted in the month of April, 2007. Using quota sampling method, the 875 individuals who fulfilled the criteria who took VCT/CVCT service were included in the study consecutively until the required sample was obtained. Data was collected using standard interview questions. For supplementing the quantitative results, 4 Focus group discussions among VCT users were conducted. To explain the study population in relation to relevant variables, frequencies and summary statistics were used. Association between dependant and independent variable was assessed and presented using odds ratio and confidence interval. Logistic regression was used to control possible confounders. Result: Out of all 771 study participants, 14.5% were couples. The HIV infection rate was 21%. Based on the adjusted regression, the respondents who were in the primary education(1-8) level were four times likely to use CVCT service than those respondents who never had a formal education with AOR(95%CI): 3.88 (1.22,12.30). Adjusted for the other variable, those study participants who used condom occasionally, during sexual intercourse with their regular partners, were five times more likely to use CVCT service than those who never used condoms with AOR (95%CI): 4.50 (1.33-15.31. Respondents who claimed to never heard ix about CVCT service utilization were less likely to use CVCT service compared to those who did with AOR (95%CI) 0.29(0.13-0.65).The participants who had no history of HIV testing were less likely to use CVCT service than those who had with AOR (95%CI) 0.44(0.27-0.71).The knowledge of participants on HIV and other STIs had no significant association with CVCT service utilization. The most common reason claimed by the respondents for not using CVCT service during the study was because the partner already knew the result. Conclusion and Recommendations: The prevalence of couples VCT service utilization in Addis Ababa VCT sites is low. Frequent, clear and accurate mediabased information programs using multiple languages are needed to increase access to information on VCT, with particular attention to the importance of VCT for couples. And other Specific approaches to promote CVCT service is needed to be developed service through appropriate researches.