Examples of Well-Designed Programs

Much can be learned from established programs. The following are systematically evaluated multicomponent programs that focus on children at three developmental periods: preschool age, including infants and toddlers; elementary school age; and adolescence. These programs may influence later antisocial behavior.

Programs for Preschool-Age Children

Programs targeting children before they enter school can address a range of individual and family risk factors that are precursors to antisocial behavior. Unfortunately, few of these programs have been tested longitudinally because of cost and difficulty. Those that have been evaluated are discussed in more detail below.

Early intervention programs for preschool-age children may help prevent antisocial behavior, particularly because they intervene through the child's broad environment rather than through a single risk factor. Because some of these programs were instituted in the 1960's, 10-year longitudinal studies have been completed for the Syracuse University Family Development Research Project, the Yale Child Welfare Project, the Houston Parent Child Development Center, and the High/Scope Perry Preschool Project.

Although the results of these programs have been encouraging (Yoshikawa, 1995), the mechanisms of change could not be identified. The programs shared a number of features. They were intensive, included home visits, and addressed the child's early development, some before birth. They also included child education and family support, a theoretical basis with curriculums specified in treatment manuals, low staff-to-child/parent ratios, extensive training, and child- and family-focused components.

Universal interventions. The Montreal Home Visitation Study (Larson, 1980), an example of a universal intervention for infants and toddlers, tested the effects of home visits on 115 women. The visits focused on caretaking, mother-infant interaction, the mother's social support and social interactions, and child development. Children in families that received prenatal and postnatal visits sustained fewer injuries than children in the other groups and also received higher scores on the quality of the home environment provided to them.

These interventions address risk factors on a number of levels. For example, the Syracuse University Family Development Research Project (Lally, Mangione, and Honig, 1988) provided educational, nutritional, health, safety, and human services resources to 109 low-income, primarily African American families. The program, which also included home visits and quality childcare, decreased the children's involvement with the juvenile justice system. When the children were between 13 and 16 years old, 4 of the 65 treated children had probation records at followup, compared with 12 of the 54 children in the control group.

The Yale Child Welfare Project¹² targeted 17 impoverished, pregnant, African American women between 1968 and 1970, matching them with controls. The program provided a variety of services that began during pregnancy and continued until the children were 30 months old, including an average of 13 months of childcare. Ten years later, mothers in the intervention group had more education, were more likely to be employed, and had fewer children who were spaced farther apart, compared with mothers in the control group. Children in both groups remained the same in IQ and academic achievement, but children in the intervention group had better attendance records and exhibited less antisocial behavior than those in the control group.

The Houston Parent Child Development Program (Johnson and Walker, 1987) provided home visits, classes for mothers, and 4 half-days per week of preschool for children more than 2 years old in low-income Mexican American families. In a 5- to 8-year followup, treated children were found to be less obstinate, hostile, and aggressive than children in the control group. Attrition from the program, however, was very high.

The PARTNERS program supplemented Head Start programs by providing training to parents and teachers to promote consistency from home to school (Webster-Stratton, 1998). Parents receiving training were more positive, less critical, and used less physical discipline than parents not receiving training. Seventy-one percent of parents in the experimental group showed a decrease in critical statements compared with 29 percent of parents in the control group. Children in the experimental group were more compliant and prosocial and displayed less negative behavior than those in the control group. Also, most eligible parents signed up for the program, and participant satisfaction was high.

Selected interventions. The University of Rochester Nurse Home Visitation Program, an example of a selected intervention for infants and toddlers, recruited 400 low-income women who were pregnant and raising children in semirural New York (Olds et al., 1986; Olds et al., 1988). The program offered four levels of intervention, including information and support for child health and development, free transportation to prenatal and well-child medical visits, home visits by nurses during pregnancy, and followup visits until the child's second birthday.

The program found that treated mothers who smoked were more likely to have full-term, heavier babies than untreated mothers who smoked. Treated mothers were also less likely to punish their children or to be seen in emergency rooms or by physicians for infant poisonings and injuries. They were also more likely to provide their children with appropriate play materials, return to school, be employed, and delay future pregnancy (Olds, 1996).

The authors of a separate cost analysis (Olds et al., 1993) found that the program saved the Federal Government $1,772 per family (in 1980 dollars) and $3,498 per low-income family, including Aid to Families With Dependent Children, child protective services, Medicaid, and food stamps. These figures do not take into account any long-term savings resulting from delinquency prevention.

The goal of the High Scope/Perry Preschool Project, a selected intervention for preschoolers, was to prevent school failure in poor 3- to 4-year-old African Americans (Berrueta-Clement et al., 1984). The program provided home visits and monthly small group meetings for parents. Children in the intervention group performed better academically, were more likely to graduate from high school and be employed, and were less likely to be on welfare than children in the control group. Rates of juvenile delinquency also were reduced.

Programs for Elementary School Children

Universal interventions. The Seattle Social Development Project, a universal intervention for elementary school children, is one of the few programs to report significant long-term reductions in violent criminal behavior.¹³ The program offered parent management training, social competence training, and support for academic skills to increase the child's attachment to school and family, reduce involvement with antisocial peers, and reduce aggressive behavior. Six years after program completion, students reported a lower incidence of violent criminal behavior, heavy drinking, sexual intercourse, and pregnancy (Hawkins etal., 1999). They also reported higher academic achievement and less misbehavior in school. Checks of school and court records supported these reports, although not at statistically significant levels.

The 2-year Baltimore Prevention Study examined academic and behavioral interventions in three schools.¹⁴ While teacher ratings of shyness and aggression were lower and reading achievement was higher in the intervention groups after 1 year, the effects on behavior did not continue after the end of intervention, particularly for girls.

The practice of using teachers in universal programs both to implement the program and to evaluate change in participating children biases the ratings. Using multiple informants (persons supplying data to the investigator) across settings will result in more accurate and less biased evaluation (Hawkins et al., 1992).

Selected interventions. Tremblay and colleagues, in a selected intervention for elementary school children, used parent management training and social competence training in a 2-year program, which resulted in lower rates of delinquency among treated boys after 6 years.¹⁵ Another selected intervention, the FAST TRACK program, addresses rural and urban kindergarten children and ethnic groups with high rates of disruptiveness (Conduct Problems Prevention Research Group, 1992). The program intervenes at school entry and between elementary and middle school and uses parent management training, weekly home visits, social competence training, academic tutoring, and classroom contingency training, which allows for integration across settings. Results have included more appropriate parenting, greater problem solving among children, and less disruptive and aggressive behavior (Conduct Problems Prevention Research Group, 1996).

The Metropolitan Area Child Study addresses children's understanding of their context and social environmental risk factors through classroom and family intervention, social competence training, and parent management training.¹⁶ This program shares important features with the FAST program. They both use multiple-stage screening procedures to target high-risk children. They also use randomized experimental designs with multiple measures and multiple informants for assessment. Intervention broadly targets the child and family in both school and home settings. The programs also are sensitive to the communities in which they are implemented.

Children with ADHD, who are at greater risk for antisocial behavior, can benefit from selected interventions that include medication. Hinshaw and colleagues (1998) examined two types of behavioral programs: direct contingency management procedures and those involving parent management training plus teacher consultation. They found that these approaches can be effective in combination with medication, but that medication alone is more effective than behavioral treatment alone.

Firestone and colleagues (1981) looked at 7-year-olds with ADHD who were treated with parent management training alone, medication alone, or both. Only medication enhanced attention and impulse control. Medication also improved academic achievement more than parent management training alone. Gittelman and colleagues (1980) studied three groups of 6- to 12-year-olds, each of which received one type of treatment: a combination of parent management training and school consultation; stimulant medication; and both the training/consultation and medication. The latter group improved the most. The group that received only medication improved to a lesser degree, and the behavior management group improved the least.

Another study (Horn et al., 1991; Ialongo et al., 1993) assigned 96 families with ADHD children to one of six groups that received either a placebo, a high dose of medication, or a low dose of medication. Half the children in these families also received social competence training, with their parents receiving parent management training. Researchers found that combined treatment had no advantage over medication alone and that none of the groups maintained treatment gains nine months later.

Pelham and coworkers (Carlson et al., 1992; Pelham et al., 1993) examined a treatment that combined stimulants and classroom contingency training for boys with ADHD at an intensive summer camp. Medication, with or without classroom contingency training, improved all behavioral and academic outcomes. The training, without medication, improved only behavioral outcomes. Another study by this group suggested that children with ADHD who receive behavioral treatments might be able to take less medication than those who receive medication alone (Carlson et al., 1992).

The longest trials of behavioral techniques in comparative treatment studies for ADHD have lasted only a few months. These comparative studies have only begun to build on successful treatment components. One early promising study (Satterfield, Satterfield, and Schell, 1987) found that long-term treatment packages that included individual, family, and educational therapy tailored to families' needs were effective in increasing academic skills and decreasing antisocial behavior.

Long-term, multicomponent treatment packages may be best for treating children with ADHD and may result in reduced delinquent behavior, as indicated by the New York/Montreal Study (Abikoff and Hechtman, 1996). The largest study of this kind, the National Institute of Mental Health (NIMH) Multimodal Treatment Study of Children with ADHD, cosponsored by the U.S. Department of Education, Office of Special Education Programs, involves 576 children (MTA Cooperative Group, 1999a, 1999b). It will evaluate the long-term effectiveness of medication or behavioral treatment compared with a combination of the two, at the same time comparing these approaches to community care.

Psychostimulant medication is useful in treating children with ADHD, but research has not yet shown whether it can prevent antisocial behavior in the long term. It allows children to function better in the short term, giving them the chance to develop adaptive, protective skills.

Indicated interventions. Webster-Stratton and Hammond (1997) conducted an indicated intervention for elementary school children that divided families into a parent management training group, a child training group, a child and parent training group, and a waiting-list control group. At the end of intervention, all three intervention groups had made significant improvements that were maintained at a 1-year followup. Although the combined group showed the most significant improvements in child behavior at followup, all three groups showed decreased conduct problems as reported by teachers.

Kazdin, Siegel, and Bass (1992) studied 97 children who had been in a child psychiatric hospital (Kazdin et al., 1987a) or who had been referred for antisocial behavior problems to parent management training, social competence training, or both (Kazdin Siegal, and Bass, 1992). One year after treatment, both inpatient and outpatient children and families who had received combined treatment had improved their behavior, but only 50 percent of these children moved into the normal range of behavior problems, according to parents and teachers (see also Kazdin et al., 1987a).

Programs for Adolescents

Most youth who exhibit serious antisocial behavior begin to do so by age 15. At this point, it is usually easy to identify these youth, but their behaviors may be so entrenched that they are resistant to change. Also, adolescent peer groups, which exert a powerful influence, are often targeted for change or as a vehicle of change. That is, the individual is encouraged to give up his or her group, or members of the group are used as counselors or mediators. Programs for adolescents are less likely to involve families than programs for other ages, although the family still has a profound influence on the youth.

Universal interventions. Aimed at reducing violence or resolving conflict, universal interventions for adolescents are a recent development;¹⁷ many of them target African American youth.¹⁸ However, these programs have several problems, including:

• Reliance on self-report measures or nonblind teacher evaluations that are subject to bias.
  
  
• Use of anger control strategies that do not have demonstrated effectiveness (Tolan and Guerra, 1994).
  
  
• Lack of randomized assignment to intervention and control groups.
  
  
• High attrition rates (see Brewer et al., 1995, for a review).
  
  
• Paradoxical effects, which may reflect treatment unreliability or may lead to worsened outcomes for some groups of children.

Farrell and Meyer (1997) experienced this last problem in a study testing a social skills and problem-solving curriculum for African American sixth graders. Boys in both the intervention and control groups had an increased number of problems during the study, although boys in the intervention group experienced a lower rate of increase than did boys in the control group. The rate of increase was higher among girls in the intervention group than among girls in the control group. The distinction may have been caused by differences in patterns of aggression among girls or by the mixed-sex groups and male group leaders used in the study.

Orpinas and colleagues (1995) examined the Second Step curriculum (Committee for Children, 1990), a conflict resolution program using peer mediators. The study reported that the intervention had no effect on aggressive behavior and produced marginal improvements elsewhere. However, the choice of school or classroom was not random (only "good" teachers were assigned to implement the program), and only self-report measures were used.

Selected interventions. Gottfredson (1986) evaluated a selected intervention for adolescents, Positive Action Through Holistic Education (PATHE), a multicomponent school-based program for impoverished 11- to 17-year-olds. The program sought to improve student attachment to school, academic success, self-esteem, and student-faculty communication through teacher training and student counseling, tutoring, and a student leadership system. Communication with families was also enhanced. The program had no effect on antisocial behavior, and participants reported higher drug involvement at posttest.

Gottfredson and Gottfredson (1992) evaluated Project STATUS, a supplemental law-related and moral development curriculum, in which students took part in field trips and structured role-play. This curriculum positively affected academic performance, as demonstrated by higher grades and graduation rates for students in the intervention group. The same students also reported less delinquency and drug use and higher self-esteem than did students in the control group.

Gabriel (1996) evaluated Self-Enhancement, Inc., a violence prevention program for African American students in grades 8 to 10. The program sought to enhance intermediary protective factors such as self-control, self-efficacy, social competence, and social bonding through field trips, conflict resolution, and student-led antiviolence campaigns. Attrition was 28 percent, and targeted protective factors did not improve. Intervention students reported decreased fighting and weapon carrying after 1 year in the program.

Indicated interventions. Henggeler and colleagues have examined multisystemic therapy as an indicated intervention for adolescents.¹⁹ Because this approach specifically targets serious juvenile offending, the results are especially important.²⁰ Multisystemic therapy combines family therapy, parent management techniques for older children, and problem-focused interventions in peer and school settings in an intensive family preservation treatment program. It has been found to increase family cohesiveness (Henggeler, Melton, and Smith, 1992), increase the adaptability and support of families of serious juvenile offenders (Borduin et al., 1995), and decrease father-mother and father-child conflict (Henggeler and Blaske, 1990). Treated adolescents were less likely to be rearrested and spent fewer days incarcerated than adolescents in the control group (Henggeler, Melton, and Smith, 1992). Overall recidivism for those completing multisystemic therapy was 22 percent; for those completing individual therapy, the rate was 71 percent. Treatment gains were maintained, even at the 4-year followup (Borduin et al., 1995). The dropout rates for those receiving multisystemic therapy were 10 percent and 15 percent, compared with 38 percent and 25 percent for those receiving individual therapy (Borduin et al., 1995; Henggeler, Melton, and Smith, 1992), probably because multisystemic therapy targets three or more systems and individual therapy usually targets only one.²¹

13 Hawkins et al., 1999; Hawkins, Von Cleve, and Catalano, 1991; O'Donnell et al., 1995.

14 Kellam and Rebok, 1992; Kellam et al., 1994; Kellam et al., 1991.

15 McCord et al., 1994; Tremblay, M‰sse et al., 1992; Tremblay et al., 1991; Tremblay et al., 1995; Tremblay, Vitaro et al., 1992; Vitaro and Tremblay, 1994.

16 Guerra et al., 1996; Guerra et al., 1995; Tolan and McKay, 1996.

17 Brewer et al., 1995; Hawkins, Catalano, and Brewer, 1995.

18 Hammond and Yung, 1991, 1992; Ringwalt et al., 1996.

19 Borduin et al., 1995; Henggeler and Blaske, 1990; Henggeler and Borduin, 1990; Henggeler et al., 1996; Henggeler, Melton, and Smith, 1992.

20 Borduin et al., 1995; Henggeler et al., 1996; Henggeler et al., 1993.

21 OJJDP supports multisystemic therapy programs. For more information about multisystemic therapy, see Henggeler, 1997, and Muller and Mihalic, 1999.

Prevention of Serious and Violent Juvenile Offending	Juvenile Justice Bulletin • April 2000