"In this pilot pretest–posttest quasi-experimental study, we saw a lower proportion of SSP clients exclusively inject heroin and a higher proportion of SSP clients consume heroin through both injection and smoking after the implementation of a heroin pipe distribution program. The proportion of SSP clients who reported syringe reuse was also lower following the heroin pipe distribution intervention. We did not observe any difference in self-reported health outcomes associated with drug use between the pre- and post-intervention periods; however, the short follow-up period and small sample size of this pilot study may have contributed to this null finding. Our results suggest heroin pipe distribution may be a novel RTI that can be added to existing SSPs to further reduce harms associated with heroin use. This study also highlights the potential for public health service innovations to be developed by marginalized communities and the importance of placing PWUD in leadership positions in efforts to optimize harm reduction programming.

"Despite the non-randomized design of this pilot study, several findings suggest heroin pipe distribution may have prompted changes in heroin consumption behaviors among PWUH. The proportion of SSP clients who exclusively injected heroin was lower by a quarter, while the proportion who both injected and smoked heroin was higher by over a quarter after heroin pipe distribution began. Twenty-four percent of respondents who used heroin reported heroin pipe distribution had reduced their heroin injection. Higher proportions of SSP clients who received heroin pipes exclusively smoked heroin or both smoked and injected heroin compared to SSP clients who did not receive a heroin pipe. We are unaware of any prior published research investigating heroin pipes as an RTI; however, pre–post-analyses examining foil distribution at SSPs in Europe found similar changes in drug consumption behaviors, with up to 85% of SSP clients having used foil to inhale rather than inject heroin on at least one occasion [23, 28]. Our non-randomized study design cannot control for confounding and prevents firm conclusions as to whether this observed shift from injection to smoking can be attributed to the intervention. Additionally, only 14% of respondents who used heroin completed surveys during both the pre- and post-intervention periods, and thus, outcomes may have been impacted by changes in the SSP client population across time periods. Further experimental research is needed to clarify the causal relationship between heroin pipe distribution and reductions in heroin injection. Study designs that are randomized by individual may be complicated by heroin pipe sharing across intervention and control groups. Cluster randomization may better control for contamination given extensive social networks and resource exchange among PWUD [29]."


Fitzpatrick, T., McMahan, V.M., Frank, N.D. et al. Heroin pipe distribution to reduce high-risk drug consumption behaviors among people who use heroin: a pilot quasi-experimental study. Harm Reduct J 19, 103 (2022). doi.org/10.1186/s12954-022-00685-7