The relationship between drugs and homicide has been well documented for some period of time. Drugs can play many different roles in homicide events. Drug homicides are disaggregated into peripheral drug homicides and drug-motivated homicides. In the former, drugs were present at the scene or drugs were being used by the victim or offender but were not the central causal feature of the event. In the latter, the sale or use of drugs was the primary cause of the lethal interaction. Using multinomial logistic analysis, we analyze the extent to which individual, situational, and contextual factors discriminate between different drug-homicideevents. We found variables indicative of risky lifestyles were significant predictors of the different types of drug homicides. More important, findings suggest the variables considered in the multivariate model had different effects on different measures of the dependent variable. Policy implications are discussed.