Epidemiology of juvenile violence.

It is difficult to review the epidemiology of juvenile violence because few studies focus specifically on this topic as opposed to childhood aggression or delinquency in general. More research is needed specifically on juvenile violence, which is generally measured using official records or self-reports. Self-report research shows that a substantial fraction of the male juvenile population commits violence, and that very few violent acts are followed by arrests or convictions. Racial differences in violence may be explainable by reference to racial differences in community contexts. There is a great deal of versatility in juvenile violence. Juveniles who commit one type of violent offense also tend to commit other types and nonviolent offenses. Violent offenders tend to be persistent or frequent offenders, and there is little difference between violent offenders and nonviolent but equally frequent offenders. Nevertheless, there is some degree of specialization in violence. More research is needed to investigate whether risk factors exist for violence that are not risk factors for serious nonviolent delinquency (e.g., biologic factors). Violent juveniles tend to have co-occurring problems such as victimization, substance abuse, and school failure. Often, they might be described as multiple-problem youth. There is considerable continuity from childhood aggression to juvenile violence. An early age of onset of violence predicts a large number of violent offenses. The major long-term risk factors for juvenile violence are individual (high impulsiveness and low intelligence, possibly linked to the executive functions of the brain), family (poor supervision, harsh discipline, child physical abuse, a violent parent, large family size, poverty, a broken family), peer delinquency, gang membership, urban residence, and living in a high-crime neighborhood (characterized by gangs, guns, and drugs in the United States). More research is needed on interactions among risk factors, and especially on interactions between biologic and psychosocial risk factors. Important short-term situational factors include motives of potential offenders (e.g., anger, a desire to hurt), alcohol consumption, and actions leading to violent events (e.g., the escalation of a trivial altercation). More specific research is needed on protective factors against youth violence, for example, by investigating why aggressive children do not become violent juveniles. More research is also needed on the development and validation of risk assessment instruments. To investigate developmental and risk factors for juvenile violence, longitudinal studies are needed. Such studies should include multiple cohorts, to draw conclusions about different age groups, and should include both boys and girls and the major racial and ethnic groups. They should measure a wide range of risk and especially protective factors. They should be based on large, high-risk samples, especially in inner-city areas, incorporating screening methods to maximize the yield of violent offenders while simultaneously making it possible to draw conclusions about the total population. They should include long-term follow-up studies to permit conclusions about developmental pathways. They should make a special effort to study careers of violence and to link developmental and situational data. It will not be easy to mount new longitudinal studies focusing specifically on juvenile violence, but such studies are needed to advance knowledge about the epidemiology of juvenile violence, including risk factors and developmental pathways.