AIMS: In France, a non-negligible proportion of opioid-dependent individuals inject morphine sulfate. Although it has not yet been officially approved as an opioid substitution treatment (OST), some physicians can prescribe its use for people in methadone or buprenorphine treatment failure. Longitudinal data from the ANRS-AERLI study, which evaluated an educational intervention for safer injection called AERLI, provided us the opportunity to better characterize the profile, risk practices and needs of people who inject morphine sulfate (MSI), through comparison with other injectors, and to identify correlates of HIV/HCV risk practices in this group. METHODS: The national multisite ANRS-AERLI study assessed the impact of AERLI offered in volunteer harm reduction (HR) centers ("with intervention") (n = 113) through comparison with standard HR centers ("without intervention") (n = 127). All participants were scheduled to be followed up for 12 months and have 3 telephone interviews: at baseline, 6 months and 12 months. We compared MSI (n = 79) with other opioid injectors (n = 161) and then used a mixed logistic model to identify factors associated with HIV/HCV risk practices among MSI. FINDINGS: Of the 240 eligible participants, 79 were regular MSI. They were less likely to use cocaine, crack or buprenorphine and to receive OST than other participants. Conversely, MSI were more likely to inject drugs more than three times a day and to report HIV/HCV risk practices. Among MSI, multivariate analysis showed that those receiving morphine sulfate as an OST were less likely to report such practices than other participants (aOR [95%CI] = 0.11 [0.02-0.61]). CONCLUSION: Our results show that while MSI use fewer stimulants, they have more HIV/HCV risk practices than other injectors. However, when MSI are prescribed morphine sulfate as a treatment, these practices tend to decrease. Our findings suggest the importance of increasing access to morphine sulfate as a new OST in France.