1. Preventive program interventions should be directed at risk and protective factors4 rather than at categorical problem behaviors, such as school dropout, delinquency, substance abuse, and teenage pregnancy.

  2. The few studies with follow-up data suggest possible “sleeper effects” in program benefits (that is, some programs showed a greater effect at a time somewhat distant from the end of the treatment than immediately following the intervention). If this is true, programs that do not collect follow-up data may underestimate the impact of their program.

  3. Prevention programs targeting multiple domains (i.e., individual, school, and community) are more effective than programs targeting one domain.

We looked more closely at whether these conclusions held true for the 10 programs targeting 10- to 18-year-olds (see Table 6–1 for details). Do short-term programs produce time-limited benefits? This was difficult to determine for this subset of programs because half of the eight short-term program evaluations (less than one year) did not do follow-up testing. Of the three programs that did follow-up testing, two found mixed results and one found positive follow-up effects. These three programs are summarized below.

The Adolescent Transition Project randomly assigned participants to one of four groups: teen focus, parent focus, teen-parent focus, and a self-directed control group. The teen focus group received training in self-regulation development along with skill development exercises. The parent focus curriculum provided training for parents in parent management skills in areas such as setting appropriate limits and problem solving with teens. The combined parent-teen group used peer consultants to open discussions between teens and parents on issues of self-regulation, parent-child interaction, and communication patterns and life skills. In the comparison group, the teens were provided with self-directed instructional materials focused on self-regulation and life skills development. After 12 weekly 90-minute sessions, the parents and adolescents in all three treatment groups reported significantly less home-based problem behavior than at the pretest; in contrast, however, there was an in-


Risk factors, such as academic failure or poverty, are variables associated with a greater likelihood of negative or undesirable outcomes. Protective factors, such as warm, nurturing adult relationships, and the personal and social assets outlined in Chapter 3, reduce the likelihood of problem behavior in the presence of risk.

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