Objective: The current study examines associations between on- and off-premise alcohol retail outlets and HIV prevalence in counties across the United States during a 3-year period. Method: Health department and U.S. Census Bureau surveillance data were analyzed from 1,523 counties in 47 states, representing more than 86% of the U.S. population. Multilevel Poisson regression models were used to examine the association between the number of on- and off-premise alcohol outlets in a county and HIV prevalence in the same county, adjusting for the between-year correlation of HIV prevalence within each county. Results: When we adjusted for potential confounders, number of on-premise alcohol outlets within a county was positively associated with HIV prevalence, whereas off-premise alcohol outlets were negatively associated with HIV prevalence. Conclusions: The relations observed in this study are consistent with the niche theory of assortative drinking, which maintains that drinkers who are prone to risk taking may be attracted to alcohol outlets where they can expand their social networks to include similarly high-risk individuals who engage in both heavy drinking and sexual risk taking. This is the largest study conducted to date to examine the association between alcohol retail outlet types and HIV prevalence. Natural experiments are needed to examine specific policy changes that reduce outlet density and its association with HIV incidence.