Background: Scaling up HIV Counseling and Testing (CT) in resource-poor settings improves access to care and is an important prevention tool. Whether high risk populations access expanded CT opportunities is unknown. Methods: We assessed HIV risk and propensity to test among 895 clients accessing mobile voluntary CT (MVCT) services in 4 rural village clusters in Northern Tanzania. Clients' individual characteristics and aggregated “propensity scores” for HIV risk and HIV testing were compared to 548 randomly selected community respondents who did not access CT. Propensity scores were each expressed as Ŷ=xβ, where x describes a vector of clients' sociodemographic and behavioral risk factors for HIV infection and correlates of HIV testing, respectively; β represents a vector of parameter estimates from gender-specific logistic regression models of HIV infection (utilizing data from 3,781 first time clients presenting to a regional free-standing CT center from 2003-2007) and a proportional hazard model of community respondents' time since their last HIV test in any setting, respectively. Results: MVCT testers were more likely to be male (OR 1.47, p< 0.001), and to report ≥ 3 lifetime sexual partners (OR 1.58, p< 0.001), lower economic resources (OR 1.45, p=0.004), and “distance from test site” (OR 2.8, p< 0.001) as a reason for not having previously tested. Males in the lowest HIV risk tercile were 67.6% more likely to present for testing (p=0.007) and males with the lowest propensity to test were 33% less likely to present for MVCT (p=0.019). High risk women were 48% more likely to test than others (p=0.04); there was no association with propensity to test. Conclusions: MVCT in Kilimanjaro enhanced service uptake by high risk women and low-income persons; however, the service was also accessed by large numbers of men with relatively low risk. Future CT expansion should be further tailored to access the highest risk groups.