Homicide: drinking by the victim.

A comparison was made between homicides in which the victim had either been drinking or had not been drinking. Medical examiner records for 792 homicides in Erie County, New York from 1972 through 1984 were used. A logistic regression found that in a significantly higher proportion of certain types of homicides there was alcohol in the victim's blood: male victims, victims aged 30-49, homicides that occur in the warmer months and in the evening or at night, homicides not related to another crime, stabbings, and homicides that take place in bars and restaurants. Some circumstances associated with a very high likelihood of alcohol in the victim's blood were: men killed at night, men killed by women, black male victims and all killings on Saturday or Sunday nights. These results suggest that a high likelihood of alcohol in victims is associated with circumstances that make drinking more likely (e.g., male victims, homicide at night), but also alcohol is more likely present in homicides that arise spontaneously from personal disputes. It was also found that there is a higher likelihood of alcohol in the victim for homicides that occur during time periods when there are more homicides. It is suggested that in some cases the alcohol may be a causal factor in the homicide.