Aims Concern about crime is a significant barrier to the establishment of methadone treatment centers (MTCs). Methadone maintenance reduces crime among those treated, but the relationship between MTCs and neighborhood crime is unknown. We evaluated crime around MTCs.
Setting Baltimore City, MD, USA.
Participants We evaluated crime around 13 MTCs and three types of control locations: 13 convenience stores (stores), 13 residential points and 10 general medical hospitals.
Measures We collected reports of Part 1 crimes from 1 January 1999 to 31 December 2001 from the Baltimore City Police Department.
Design Crimes and residential point locations were mapped electronically by street address (geocoded), and MTCs, hospitals and stores were mapped by visiting the sites with a global positioning satellite (GPS) locator. Concentric circular ‘buffers’ were drawn at 25-m intervals up to 300 m around each site. We used Poisson regression to assess the relationship between crime counts (incidents per unit area) and distance from the site.
Findings There was no significant geographic relationship between crime counts and MTCs or hospitals. A significant negative relationship (parameter estimate −0.3127, P < 0.04) existed around stores in the daytime (7 am–7 pm), indicating higher crime counts closer to the stores. We found a significant positive relationship around residential points during daytime (0.5180, P < 0.0001) and at night (0.3303, P < 0.0001), indicating higher crime counts further away.
Conclusions Methadone treatment centers, in contrast to convenience stores, are not associated geographically with crime.