risk children must include screening methods that minimize false positives and false negatives and also minimize the risk of negative labeling effects. For example, recent data on early elementary school children show high false-positive rates (up to 30 percent) in screening for risk of conduct problems (Lochman and the Conduct Problems Prevention Research Group, 1995). Fourth, recent data raise the possibility of iatrogenic effects produced by selective interventions that concentrate groups of antisocial adolescents in peer group interventions; such a strategy may inadvertently increase adolescent problem behavior (Dishion and Andrews, 1995).

Thus, despite substantial progress in the development, implementation, and evaluation of family-based prevention interventions, there are continuing needs and opportunities for research in the following areas:

  • Family interventions for high-risk groups: Research is needed on effective family interventions for groups at high risk for drug abuse, including children with conduct problems and children of drug-abusing parents. These are not necessarily independent subgroups; children of drug-abusing parents may be at risk for drug abuse partially because of their conduct problems. Because many of these programs target early elementary school children, longitudinal follow-up is necessary to assess relevant drug use outcomes, although short-term impacts on risk mediators (e.g., problem conduct) might be seen even in the elementary school years. Additionally, greater integration of preventive interventions focused on drug use or abuse, conduct disorders, and delinquency is needed (IOM, 1994b).

    Current methods for defining and assessing high-risk groups produce high rates of false positives. It is necessary, therefore, to develop reliable and valid screening and assessment instruments, as well as methods of intervention delivery, designed to minimize the likelihood of negative labeling effects for selective interventions. This might be accomplished by incorporating interventions within existing treatment programs for drug-abusing parents, within existing prevention and treatment services for children with conduct problems, or within existing treatment and prevention programs in the juvenile justice system. In these ways, children would not be further labeled.

  • Developmentally appropriate interventions: At early ages, high-quality preschool environments and social skills training may be important additions to family-based interventions. At later (adolescent) ages, interventions may also benefit from focusing on communication skills and family management of external stress (Tolan et al., 1995). However, concentrating antisocial adolescents in peer group interventions may have negative effects (Dishion and Andrews, 1995).

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement