Needle-exchange programs (NEPs) have been shown to be effective in reducing harm related to injection drug use and to act as an important link between the injection drug using community and preventive/treatment services. Different needle-exchange distribution methods may reach different subpopulations of injecting drug users (IDUs). We undertook this study to characterize risk behaviors by primary source of clean needles accessed by IDUs in a city with pharmacy access and fixed and mobile exchange programs. We hypothesized there would be a gradient of risk across the three types of distribution. Data were collected from within the Vancouver Injection Drug Users Study (VIDUS), a prospective cohort study. Participants who primarily obtained clean needles from pharmacies, fixed sites, or mobile exchange vans were compared using the Cochran-Armitage trend test to test for trends in increasing risk behaviors across the three types of distribution. Ordinal multivariate regression was used to adjust the associations for potential confounders. Results illustrate clear trends for increasing risk profiles from pharmacy to fixed site to mobile exchange vans. Van users were generally at higher risk than fixed-site and pharmacy users. Independent predictors of van use were fewer years injecting, difficulty finding needles, Aboriginal ethnicity, incarceration in the previous 6 months, and injecting cocaine daily. An important component of needle-exchange programs is outreach to access those who are at highest risk. Use of distribution beyond fixed sites will improve such outreach, thereby increasing program effectiveness and further preventing the transmission of blood-borne infections.