Youth violence has been a major cause of death for adolescents in the U.S. for over a decade. The lethal character of youth violence is evident in data on intentional deaths from weapons. From 1985 to 1991, firearm fatalities increased 127 percent among males 15-19 years of age, while declining for males over 25 years of age. Virtually all increases in homicide rates from 1985 to 1990 among people 10-34 years of age are attributable to deaths among African American males. These trends suggest that efforts to prevent and reduce adolescent violence must address the risk factors for lethal adolescent violence, and its concentration demographically and spatially among African American males in urban areas.

Research is needed to test the contributions of macrostructural factors to adolescent violence, to identify the social and economic risk factors that lead to elevated risks of injury and homicide fatality among adolescents, and to identify the neighborhood contexts that shape and influence violent transactions among adolescents.

This research has three aims. First, we specify and test theoretical models based on social isolation and concentrated poverty to determine social and economic risk factors for adolescent homicide for 1976-94 in U.S. cities.

Second, the model will be replicated for homicide and hospitalized injury cases in New York City for 1988-94, a period marked by first rising and then falling homicide rates. Geographical information systems (GIS) procedures are used to locate homicide, suicide, and hospitalized intentional injury cases involving adolescents in New York City, and establish injury-specific (e.g., victim, circumstance, method) population rates by census blocks, census tracts, and health districts. Census, health, and criminal justice data are being integrated with injury data for model testing of risk factors for specific types of adolescent violence. The analyses identify social and economic characteristics of neighborhoods with elevated risk for adolescent violence and suicide.

Third, we are conducting qualitative neighborhood-level case studies to identify and describe contextual effects that characterize mirco-areas with either increasing, stably high, declining, or stably low rates of adolescent violence over time. The results will provide information to public health practitioners to target neighborhood-specific prevention and intervention efforts to reduce social and economic risks of adolescent violence.