Previous research has identified poverty, minority racial composition, and disrupted family structure as characteristics of neighborhoods with high levels of crime. Non-residential land use also plays a part in the prediction of crime rates for an area. One study found a correlation between the number of bars in a city block and high rates of violence and property crimes. The current study probed whether the effectiveness of police raids in reducing illegal drug activity in and around bars varied with the population and land use characteristics of the area surrounding the targeted bars. The effects of both dosage (number of raids) and duration (number of months) of the raids was also assessed. The study area included 63 nuisance bars in Pittsburgh and the circular areas around the bars within a 1/8th mile radius. Data on arrests and 911 calls relating to illegal drug use or distribution were examined for an 8-month period in 1992. Results of statistical analyses indicate that police raids are effective in reducing illegal drug activity in and around the bar during active interventions; however, the effects disappear when the intervention is removed. The effects are also further mediated by the risk characteristics of the target and surrounding areas. Bars in areas characterized by high levels of risk are more resistant to the suppression effects of police raids. This finding is further mediated by whether the risk involves land use or the population. Bars in areas characterized by high land use risk are more affected by police raids than are bars in areas characterized by high levels of population-based risk. However, even the most ambitious intervention program by police cannot sustain suppressed levels of illegal drug activity after the intervention is completed. Figures, tables, references