Objectives In June 2011, closed-circuit television (CCTV) was introduced in Footscray (a suburb of Melbourne, Australia) to help deter street-based drug trading. We investigate whether there were subsequent shifts in the settings (e.g., street, house) in which heroin was purchased or injected by people who inject drugs (PWID).

Methods Using heroin purchase data from the Melbourne Injecting Drug User Cohort Study, multinomial logistic models with fixed effects for CCTV introduction were used to estimate the percentage of: (1) heroin purchased on the street, from mobile dealers and in house settings; and (2) heroin injections occurring in street, car, public toilet, and house settings. Displacement effects were investigated with a logistic model capturing the likelihood of traveling to Footscray to purchase heroin.

Results Following CCTV introduction, the percentage of heroin injections occurring in public toilet settings decreased by 13 % (95 % CI −27 %, −0 %). This was accompanied by a non-significant increase in the percentage of heroin injections in street settings of 23 % (95 % CI −1 %, +41 %). Changes in other settings were small and nonsignificant. No suburb displacement effects were found.

Conclusions The introduction of CCTV in Footscray may have displaced PWID who previously injected heroin in public toilets to street settings. Apart from this, Footscray’s street-based heroin market operates much as it did before CCTV.

This work was supported by the National Drug Law Enforcement Research Fund [NDLERF 2012/13-13]. The Melbourne Injecting Drug User Cohort Study is funded through National Health and Medical Research Council grants and untied philanthropic contributions. PD is the recipient of an Australian Research Council Future Fellowship. PH is supported by a Curtin University Research Fellowship. The National Drug Research Institute at Curtin University is supported by funding from the Australian Government under the Substance Misuse Prevention and Service Improvements Grants Fund. The authors gratefully acknowledge the contribution to this work of the Victorian Operational Infrastructure Support Program.

This work was supported by theNational Drug Law Enforcement Research Fund [NDLERF 2012/13-13]. The Melbourne Injecting Drug User Cohort Study is funded through National Health and Medical Research Council grants and untied philanthropic contributions. PD is the recipient of an Australian Research Council Future Fellowship. PH is supported by a Curtin University Research Fellowship. The National Drug Research Institute at Curtin University is supported by funding from the Australian Government under the Substance Misuse Prevention and Service Improvements Grants Fund. The authors gratefully acknowledge the contribution to this work of the Victorian Operational Infrastructure Support Program.