Identification of Drug Abuse Prevention Programs

Literature Review

Sections
- Introduction
- Purpose
- Types of Prevention Interventions
- Part A - Universal School-Based Prevention Programs

Part B - Selective School-Based Prevention Programs:
   - High-Risk Youth Programs
   - Alternative Programs
   - Tutoring or Mentoring Programs for High-Risk Youth
   - Cultural Competency or Pride Programs
   - Family-Focused Selective Prevention Approaches
   - Summary of Selective Approaches

- Part C - Indicated School-Based Prevention Programs
- Comparison of Effectiveness of Different Prevention Approaches
- Summary of Results of Effectiveness Studies
- References

Selective Prevention Programs

Selective prevention programs are offered only to at-risk groups of youth and families. They include programs for children of alcoholics or drug abusers, children in poverty that frequently include minority youth and families, immigrant acculturating families, student transfers from other schools, families with high mobility, steroid prevention programs for athletes, or other special prevention programs targeting high-risk youth and families. To be effective, selective prevention programs should be longer (generally more than 45 hours of services) and more intensive than universal programs, and should include activities that are directly targeted to reducing identified risk factors and increasing protective factors found in the group. To maintain effectiveness over time, booster sessions are needed to review prior learnings and skills and to introduce developmentally appropriate new material. Table 3 lists a variety of selective prevention programs.

Programs are targeted only to those who are likely to need prevention; therefore, selective programs can lead to more efficient use of limited resources for prevention. The content is more tailored to the specific risk or protective factors of the group, thus increasing effectiveness. Additionally, it is easier to measure improvements from a prevention program if the participants have more serious risk factors at intake. Typically, the effectiveness of universal prevention programs is smaller than that for selective programs for high-risk participants because fewer participants have room for improvements (e.g., a ceiling effect). Consequently, a number of universal prevention researchers (Dielman et al., 1989) are conducting post-hoc statistical analyses separating low- and high-risk subjects. They find larger program effect sizes for high-risk youth and families.

Identifying, recruiting, and attracting high-risk youth can be more difficult than providing universal programs to all students in schools. Criteria for designating risk status can be broad and sometimes not well substantiated, thereby raising the potential for negative labels and stereotyping. For example, "minority" youth are portrayed as a homogeneous group "at risk" for substance use/abuse when in fact (a) they are quite heterogeneous; (b) preliminary research on risk and protective factors suggests similar precursors for substance use across ethnic groups, but possible differences in exposure and reaction to these risk factors (Newcomb, 1993, 1995); and (c) considerable epidemiological evidence suggests that most ethnic youth use some substances (e.g., tobacco and alcohol) at rates lower than European American youth. Although universal prevention programs may be overinclusive, selective programs may overlook or fail to voluntarily recruit some youth and families at risk.

Many of the selective prevention programs being developed for high-risk youth are also using the popular social skills training approaches developed for general populations of students. For instance, an analysis of the first 75 high-risk youth grants funded by the Office for Substance Abuse Prevention (OSAP, now CSAP) revealed that 84% offered life skills training, 81% offered communication skills training, and 69% offered social influence and peer resistance skills training (OSAP, 1992). However, in line with the general philosophy of the high-risk youth grantee programs, the demonstration/evaluation programs were also to test many new approaches and let the "thousand flowers bloom." Hence, whereas many of the universal prevention programs are based solidly on the well-researched social skills training approach to learning new behaviors and social competencies, a wider variety of intervention approaches are being used in selective prevention programs for high-risk youth.

Many of these new approaches are being tested in demonstration/evaluation or services grants with DOE, CSAP, and state or county funds. These approaches (see Table 3) are tailored to address different hypothesized theoretical precursors of drug use:

• Increased genetic or environmental vulnerability to drug use in high-risk students (e.g., athletes, children of substance abusers [COSAs], inner-city poor children, or high-risk youth experiencing many stressors) is addressed by providing special education, support, and resiliency and social skills training for COSAs.
• Poor use of leisure time and boredom are addressed by providing alternative activities such as experiential challenge initiative programs, youth clubs, and sports and recreation (CSAP, Technical Report 13, 1997).
• Lack of involvement or bonding to traditional institutions is addressed by creating peer-led programs.
• Academic failure is addressed by providing tutoring and mentoring services.
• Lack of ethnic pride and low self-esteem is addressed by providing cultural pride activities.
• Lack of multicultural competencies is addressed by programming to increase multicultural skills and ability to be successful in the white culture as well as their own.
• Lack of responsibility and internal control is addressed by providing rites of passage programs, peer-led programs, and youth councils.
• Lack of parent involvement and supervision is addressed by providing family-strengthening approaches (Kumpfer, 1993, 1997c).

Because of the popularity of delivering drug prevention programs universally in schools, historically fewer selective prevention programs have been tested in true randomized clinical trials. In the past 10 years, selective or targeted prevention programs for high-risk youth and families have become more popular with researchers and practitioners. CSAP has funded demonstrations/evaluations of many of these selective prevention programs, including a high percentage of those for minority youth and families. Unfortunately, until recently, most of these research efforts involved nonexperimental designs, rather than higher quality quasi-experimental and true randomized designs capable of determining whether the interventions caused the changes in the participants. Hence, many of these innovative and promising programs, which were developed by nonresearch institutions, such as CSAP, the DOE Drug-Free Schools Program, the CDC, and the OJJDP, do not have strong outcome effectiveness results. In fact, it has been very difficult to locate any published evaluation results for most of these programs. Some of the original 1987 CSAP High Risk Youth and Families programs have been published in the special 1992 issue of the Journal of Community Psychology (Ross, Saavedra, Shur, Winter, & Felner, 1992) and the CSAP Prevention Monograph 12: Working With Youth in High-Risk Environments: Experiences in Prevention (Marcus & Swisher, 1992).

Examples of selective prevention programs (see Table 3) include many of the culturally tailored school-based, after-school children's skills and cultural competency training programs funded by CSAP, such as the Hispanic Youth Leadership Institute (HYLI) program (Kumpfer, Alvarado, & Kendall, 1996). This program, like many other CSAP school-based programs (Stoil & Hill, 1996) targeted at ethnic minorities, provided after-school youth clubs that combined youth mentoring, tutoring, social skills training, guided recreation, and cultural competency and cultural pride activities. Although many of these programs were not well evaluated, those with control or comparison groups have found significant improvements in many risk factors for drug use. These major selective prevention approaches are described in more detail in Table 3.

Table 3: Types of Selective School-Based Prevention Programs

Sections

High-Risk Youth Programs

Despite this dearth of research on selective programs, NIDA (1997) cites four examples of research-based selective prevention approaches with effectiveness in reducing substance use in youth. Three are family-focused programs (the Strengthening Families Program, the Iowa Strengthening Families Program, and Families in Focus), which are discussed in the final part of this section on selective programs. Only one school-based selective program - the ATLAS Program - is discussed in the NIDA "red book." It targets high-risk male high school athletes and is discussed below.

This 14-session multicomponent program includes seven sessions of classroom training in physical consequences, drug resistance, normative education, drug vulnerability, media education, goal setting, and nontolerance of drug use and seven sessions of weight-lifting at school and sports nutrition (Goldberg et al., 1996; Wolf, 1996). Four booster sessions are provided each year of high school. Parents are involved in parent-student homework employing the booklet "Family Guide to Sports Nutrition." Results suggest changes in all of the areas targeted in the curriculum as well as continued resistance after 1 year to anabolic steroid use and a commitment to better exercise and nutrition.

Other examples of selective social influence or social skills training selective prevention programs include Sussman's program delivered to high-risk youth in alternative schools (Sussman, in press) and Schinke's comprehensive social influence program delivered in community youth clubs (Schinke & Cole, 1995).

This targeted, social competency training program is typically implemented in communities with groups of high-risk 10- to 11-year-old youth through Boys and Girls Clubs in poor inner-city areas and housing projects (Schinke & Cole, 1995). Research suggests that 44% of youth in public housing are latchkey children and are known substance abusers. This program is based on four primary theories of drug use: (a) social cognitive theory, (b) problem behavior theory, (c) peer cluster theory, and (d) family network theory. The 40 sessions of youth skills training include experimental role plays and recreation biweekly led by peer leaders to increase youth interest and effectiveness (Tobler & Stratton, 1997). Falco (1993) reports: "Smart Moves is one of the few prevention curricula designed specifically for very high-risk children. . . . Smart Moves tailors its program for Boys and Girls Clubs in inner-city neighborhoods. Children participate in after-school classes as well as recreational, educational, and vocational activities. An evaluation in 1987 reported that the boys who participated in Smart Moves were more able to refuse marijuana and cocaine and less willing to try drugs than other boys" (p. 59). The effect size of this program has been found to be larger than the other similar universal social skills training programs (Tobler & Stratton, 1997), possibly because a higher percentage of the participants are in need of behavioral skills training. Additionally, this program makes more of an effort to include culturally sensitive recruitment, retention, and training methods.

A cross-site evaluation of Weed and Seed Programs (Office of Justice Programs (OJP), 1995), including 15 Boys and Girls Programs, reports overall reductions in substance abuse, drug trafficking, and drug-related criminal activity in clubs having Smart Moves and fewer damaged units and less criminal activity in those housing developments with or without Smart Moves but with these youth clubs. Schinke, Orlandi, and Cole (1992) report that residents in housing developments with Smart Moves also have a greater sense of hope and belonging compared to those that do not have the program.

This social competency curriculum has been found to reduce antisocial behavior in kindergarten children who are deaf (Greenberg & Kusche, 1993). The lessons are developmentally sequenced and include dialoguing, role playing, modeling by teachers and peers, social and self-reinforcement, attribution training, and verbal mediation.

Other examples of selective prevention programs targeted at high-risk youth, such as children of alcoholics and substance abusers, include The Strengthening Families Program (Kumpfer et al., 1989) and Focus on Families (Catalano, in press), which are described in more detail under family strengthening approaches below, and the children of alcoholics programs developed by the National Association of Children of Alcoholics and the CASPAR Program. In addition, some new programs are being developed that focus on resiliency skills training for youth experiencing many stressors (e.g., death, divorce, losses) (Richardson, 1996; Wolin & Wolin, 1993).

Sections

Alternative Programs

This approach to substance abuse prevention is based on the assumption that involving high-risk youth in activities that are free of alcohol, tobacco, and other drugs will occupy their leisure time with prosocial activities and allow them to make friends with more social peers. Aguilar and Munson (1992) argue that alternative interventions "should introduce youth to more positive, acceptable expressions of leisure" (p. 23) and should be incorporated into adolescent substance abuse prevention and treatment programs. Additionally, research on mediators of drug use supports a link between difficult temperament in childhood and sensation-seeking or thrill-seeking personality and drug use in adolescence (Cloninger, Sigvardsson, & Bohman, 1988). For this reason, many of the currently popular alternative programs are experiential education programs or challenge initiatives involving wilderness experiences, ropes courses, mountain climbing, rappelling, and rafting. These challenging activities are discussed below in a separate section.

Bukoski (1991) states: "Alternative programs provide opportunities to youth who may be at risk for drug use. . . . Positive alternatives to drug use have included Outward Bound or wilderness experiences, cooperative community services or restoration projects, sky diving, and volunteering one's time or talent to help less fortunate persons." (p. 9) (Cook, 1985; Cook, Lawrence, Morse, & Roehl, 1984; Tobler, 1986). Because of the increased cost of alternative activities as compared to classroom curricula or activities, this strategy is generally used primarily with high-risk individuals. As such, it represents either selective or indicated prevention approaches. This indirect approach to drug use avoids labeling and stigmatizing youth as high risk for drug use, because the programs often never mention their underlying purpose.

According to the CSAP Technical Report 13: A Review of Alternative Activities and Alternative Programs in Youth-Oriented Prevention (CSAP, 1996), these activities are likely to include one or more of the following elements:

• Promotion of skills, knowledge, and/or attitudes that support youth in refraining from future substance abuse
• Occupation of free time that might otherwise be idle or unstructured
• Community service that provides meaningful involvement in socially responsible activities
• Opportunities to interact in positive ways with peers
• Adult supervision or development of positive relationships with adults

These programs do not include drug education content directly.

This approach is very popular despite meta-analyses and reviews of the literature suggesting weak effects (Tobler, 1986) and even iatrogenic effects (Swisher & Hu, 1983) from some types of alternative programs (i.e., alternative activities such as rock bands, housing construction, and sports that involve nonusing youth with using older peers or adults). For example, over 75% of all CSAP's Community Partnership programs involve alternatives. According to a recent annual report (CSAP, 1996), grantees reported that the largest percentage of these coalition funds was spent on alternative programming. Some of the most popular alternative activities include weekend and afterschool activities, such as experiential wilderness programs and ropes challenge courses; vocational skills training programs; youth clubs; sports and recreation programs; cultural competency including ethnic music, theater, and arts; alcohol- and drug-free dances or graduation events; entrepreneurial ventures; community service; and community drop-in centers or teen crisis hotlines. Often a number of these activities are combined into a single program or with other prevention approaches - school climate change programs, such as Project HI PATHE (Kumpfer et al., 1991), or indicated prevention programs for school dropouts (Eggert, Thompson, Herting, & Nicholas, 1994; Eggert, Thompson, Herting, Nicholas, & Dicker, 1994).

One of the few well-researched alternative programs is Amazing Alternatives (Murray & Perry, 1985). This program helps youth to identify health-enhancing alternative activities for each function served for them by substance use. Some empirical support for involving youth in engaging alternative activities is found in a study reporting that adolescent substance abusers were more likely than non-substance abusers to experience leisure as boredom, suggesting their current leisure activities fail to satisfy "their need for optimal arousal."

The existing research evidence of alternative programs is limited, with few replications of similar models, the major reviews being those already mentioned (CSAP, 1996; Swisher & Hu, 1983; Tobler, 1986). These reviews suggest a small positive effect from alternative programs when universally applied; however, these programs may be more effective with high-risk or indicated youth capable of more positive change and lacking traditional opportunities for prosocial activities. A meta-analysis of the CSAP high-risk youth and families programs (CSAP, 1996) suggests that alternative programs were second in effectiveness after family-oriented approaches or youth-led drug education approaches. Schaps and his associates (1981) found that the effectiveness of this approach was equivocal. Research by Swisher and Hu (1983) helped to explain the contradictory findings for this approach. They found that some alternative activities promoted decreased drug use and others promoted increased use. Activities such as those involving entertainment, sports, social, extracurricular, and vocational activities are associated with increased use of alcohol or drugs, whereas activities such as academic and religious activities and active hobbies are associated with decreased use. One key consideration when selecting the type of alternative activity to be promoted or the staff to hire should be the alcohol or drug use patterns of the role models running the alternative program or the other youth to be involved.

Through the CSAP high-risk youth grantee programs, enough research exists to suggest that several of these alternative approaches are promising for later Phase III Controlled Intervention Trials, namely those that promote attachment to social institutions (e.g., family, school, and church), academic achievement, educational aspirations, social competency, and unfavorable attitudes toward substance use. The outcome research suggests that community service-oriented programs produce mixed results. The cultural competence programs are popular with participants and effective in several respects. The most effective alternative programs involve youth in the development of the activities, include social and life skill building made more attractive by including creative or highly valued recreational outings, are more intensive with activities several days a week with booster sessions, and provide opportunities for positive relationships with adult role models or potential mentors. Alternative programs can enhance their effectiveness if they also focus on environmental change by having youth work on legislative advocacy to reduce the availability of tobacco, alcohol, and drugs.

According to Carmona and Stewart (1997): "Despite inconclusive research (for alternative programs), it's a good bet that youth who are surrounded by caring adults providing supervision, and who have age-appropriate opportunities to learn and grow, are likely to develop fewer problems - including substance abuse problems - than youth left to fend for themselves and whose contacts with adults are mostly negative" (p. 19). Because these activities provide additional positive adult contact and opportunities for high-risk youth lacking such developmentally appropriate, resilience-building activities, additional research to determine critical ingredients and effectiveness principles is warranted. The effectiveness of some of the major components of alternative programs is examined below.

Challenge initiative programs are novel, noncompetitive games and activities typically employing a supportive facilitator and team. Originally challenge initiatives were used as part of larger experiential education programs conducted by Outward Bound (Drebing, Willis, & Genet, 1987) and Project Adventure (Schoel, Prouty, & Radcliffe, 1988). The Association of Experiential Education defines experiential education as "a process through which a learner constructs knowledge, skill, and value from direct experiences" (Wurdinger, 1990, p. 60). These experiential exercises are now used as a tool in many different types of programs to provide direct learning experiences to effect lasting changes in participants (i.e., increase a youth's sense of personal confidence, increase competencies, and boost self-esteem). Therefore, they are becoming very popular in community coalitions and in youth-involvement programming in schools and communities.

Applications of these recreation programs to substance abuse prevention are presented by Dodd (1996). The underlying assumptions are that challenge initiative programs are excellent ways to increase self-esteem, improve self-concept, cultivate prosocial behaviors, enhance decision making, improve problem solving, and enhance team work and leadership skills (Johnson, 1992). According to the Social Ecology Model of Adolescent Substance Use, self-esteem and self-concept are mediators leading to association with drug-using peers and later increased substance use in high school students (Kumpfer & Turner, 1990/1991). Self-esteem is strongly influenced by school involvement and bonding in white students, but not so much in minority students. Hence, providing other community opportunities for enhanced self-esteem may be very important for ethnic students.

Parchem (1975) observed that "experiential learning is like a black box, where we put people in one end and they come out better at the other end, but we don't really know what happened inside the box, but it works" (p. 2). For this reason, Wurdinger (1990) and White (1997) believe it is critical for research purposes that experiential programs be based on logical theoretical assumptions capable of measurement. The following theories explain why experiential activities should be effective in altering the developmental trajectories of youth:

• They enhance personal self-confidence by allowing the individual to view himself/herself as increasingly capable and competent (Rohnke, 1989) and increase understanding of personal capabilities, frequently demonstrating to youth that they are more capable than they thought (Zook, 1987).
• The challenge activities can produce a profound sense of "I exist" or "I can affect the world," increasing a sense of internal locus of control and human agency, leading to increased personal goals and perseverance (Kimball & Bacon, 1993).
• Facilitators' questions can help youth to clarify goals, problem-solve differently, make stronger commitments, and accept personal responsibility for success or failures (Warren, Sakofs, & Hunt, 1995).
• Group activities increase trust in others or awareness that interdependence produces greater levels of success for all (Nadler & Luckner, 1992; Warren & Tippet, 1988).
• Higher levels of self-efficacy are developed by completion of more difficult tasks (Tang & Baldwin, 1990).

Several different types of experiential challenge programs currently are being implemented for prevention purposes: (a) experiential wilderness programs, (b) ropes courses, or (c) recreation and sports programs. Some of these approaches have been used for a number of years as experiential therapy programs for mentally ill or disadvantaged youth (Kumpfer, 1994). These programs are popular with youth because of their attractiveness in providing new fun and challenging activities. Experiential wilderness programs include backpacking, river running, mountain climbing, survival courses, and ropes courses.

Ropes course programs include a series of experiential education and skill-building activities conducted in an artificially constructed environment allowing participants to climb and traverse trees, telephone poles, concrete walls, or wooden constructions using ropes, steel cable, and webbing. Typically there are a series of more challenging activities, including ground activities, low-altitude initiatives, and high-altitude initiatives conducted under strict supervision. According to Carver (1996), participation in this series of sequenced initiatives improves social development by activating four mechanisms: active learning, use of prior experience, authenticity, and cognitive processing of the experience for implications about future opportunities. Metaphorical applications of the students' current experience are encouraged to change the ways youth think of themselves (Coombs & Freedman, 1990). Students also experience successes that enhance personal agency (Bandura, 1989) or sense of self-efficacy, belonging, and competence through active involvement in the experience.

Although challenge initiative programs have increased in number significantly since they became popular in the 1960s (Schoel et al., 1988), there are few published empirical studies of their effectiveness (Gass, 1993). Very few have been applied to the prevention of drug use, and even then they are used more in drug treatment programs (Gillis & Simpson, 1991). Most of the research findings can be found in unpublished theses and dissertations. A meta-analysis by Carson and Gillis (1994) found that only 25 of 235 were published and only 9 of these were reasonably well designed. White (1997) reviewed the impact of these programs on hypothesized self-esteem and external locus of control mediators of substance abuse and concluded that many of these studies were too poorly designed to make use of the results. They involved nonexperimental designs with nonrandom assignment of subjects, very small sample sizes leading to misinterpretations of a lack of effectiveness when the power was too low, low-intensity programming or too small a dosage over a very short period of time to affect self-esteem or locus of control, too short a follow-up period for evaluation, failure to measure changes appropriately, and inaccurate interpretations of the data.

The few studies that found significant impact on self-esteem or interpersonal behavior in participating youth (a) were conducted with youth with significant problems such as inpatients in a psychiatric hospital (Blanchard, 1994) or abused children in residential care (McDonald, 1988); (b) involved random assignment to a treatment as usual condition or traditional recreation programming (McDonald & Howe, 1989); (c) had longer and more intensive programming of at least 27 hours over at least 3 weeks; and (d) used well-known standardized tests for measurement of hypothesized change constructs such as the Achenbach Child Behavior Checklist (CBCL)-Teacher Report form, California Psychological Inventory, Coopersmith Self-Esteem Scale, and Piers Harris Children's Self Concept Scale. Significant improvements were found in the children's self-report of their attitudes about their own behavior, anxiety level, popularity, and happiness on the Piers Harris (McDonald & Howe, 1989).

Long positive effects recently have been found for a 9-week ropes course using 14- to 18- year-old adolescents (White, 1997). This program specifically included positive verbal persuasion ("You are a pro!" "You are really skillful!" "I am impressed at how well you can do this!"), which has been found to contribute to self-esteem and to affect participants' judgments on their ability to conduct a specific task (Ellis, Maughan-Prichett, & Ruddell, 1993).

Little empirical published research could be found for this approach. One research study (Hackler, 1966) with a 4-year follow-up (Hackler & Hagen, 1975) found a slight negative effect of the Opportunities for Youth Project on prevalence of recorded delinquent offenses. This program provided Saturday jobs with adult supervision or remedial education using a teaching machine that results in a small positive effect on delinquency. The Swisher and Hu (1983) review of the literature suggests that vocational skills programs can have negative effects on drug use as an alternative strategy. The author has some unpublished data on possible reasons for this effect. An evaluation of an urban housing development project to take dropout youth and enroll them half days in school and half days in building homes for the poor revealed that those youth trained in certain trades locally associated with a number of alcohol- or drug-using workers (e.g., roofers and painters) increased their drug use on a 6-month follow-up test compared to youth trained in construction trades locally associated with non-using workers (e.g., electricians, sheet rockers, and framers).

A cross-site evaluation involving 2,500 14- to 15-year-olds from five cities in the Summer Training and Education Program (STEP) was conducted by Public/Private Ventures (Walker & Vilella-Velez, 1992). When half-time summer jobs are combined with half-time remedial reading, math instruction, and life skills training, immediate and short-term positive outcomes are found for employment, earnings, reading and math scores, and academic skills. Unfortunately, the 3.5- to 4.5- year follow-up study revealed no significant differences between the experimental and control groups for high school dropout, college entrance, teen pregnancy, and employment rates. However, replications of STEP in 100 sites in 15 states with over 20,000 youth have produced improvements in academic skills.

Many school and community-based substance use prevention programs involve after-school programming to address the higher risk for drug use in latchkey youth (Richardson, Dwyer, & McGuigan, 1989). These programs generally involve a number of activities, such as tutoring, mentoring, social or life skills training, and recreation. The effectiveness of these multicomponent programs is frequently not tested, and only a few evaluations could be found in the research literature. The results are mixed, probably depending on the level of risk in the youth, the quality of the programming, the implementation fidelity, and the quality of the evaluation. Generally, the evaluation designs are weak, with no control or comparison groups or long-term effectiveness.

Ross and associates (1992) evaluated a 144-day, 2-hour after-school program for 540 latchkey elementary students against 296 comparison students and found no effects on personality variables or in-class behavior and even a possible negative effect on some variables. There was a significant interaction effect, however, between participation in self-esteem-building exercises and improvements on standardized achievement tests. The multicomponent program included supervised homework, self-esteem-building exercises, free play, and creative dramatics. The major flaw in this study was that lack of random assignment to groups resulted in baseline differences, with the comparison students having more risks and needing the program more. The authors recommend that such after-school programs make a "greater effort to recruit latchkey students who are truly in need of what the program offers" (p. 37) with greater outreach to the parents of the neediest youth. Bellamy and associates (1997) conclude that after-school recreation programs "that aggressively recruit youths and maintain high rates of attendance may be promising interventions for prevention delinquency and violence" (p. 63).

The Substance Abuse and Mental Health Services Administration (SAMHSA) has launched its Girl Power initiative to get 9- to 14-year-old girls more involved in sports based on the findings that young female athletes have higher self-esteem, have increased self-confidence, and are happier and healthier than their less active counterparts (Health and Human Services [HHS], 1996). Information about this initiative can be found on their Web site at http//ncadi.samhsa.gov/gpower/index.htm. HHS Secretary Donna Shalala, professional women's basketball star Dawn Staley, and cosponsor NIKE recently launched a five-site pilot program, called P.L.A.Y. CORPS, to involve 1,000 trained student coaches mentoring male and female students. The student coaches receive $500 toward their college tuition. The President's Youth Council on Physical Fitness and Sports and corporate backers have begun a million-dollar media campaign through over 1,000 television and radio stations to encourage America's youth to get more active. CADCA and the U.S. Youth Soccer Association are promoting the KICK IT! Campaign to provide antidrug information to over 250,000 soccer coaches and 500,000 volunteers. HHS, CDC, and U.S. Soccer also are sponsoring the Smoke Free Kids and Soccer program to encourage teen girls to participate in soccer and resist pressures to smoke. The program's approach includes distributing television and radio public service announcements and having members of the women's national soccer team speak to local schools on the dangers of smoking.

Although little research exists on the effectiveness of this strategy for the prevention of drug abuse, Join Together (1997) reports: "We know that when youth are active in sports and other recreational activities, they are less likely to engage in unprotected sex, get into trouble or commit acts of violence." Youth Sports: Part of a Strategy to Reduce Substance Abuse, published by Join Together (1997), provides examples of promising programs using sports and recreation activities and encourages communities to implement them because youth can learn peer pressure skills, gain confidence, and learn goal setting and the value of hard work from sports. Coaches, if well trained, can serve as mentors and help guide young people through difficult choices. The following activities were implemented by the Join Together sites:

• Local youth sports teams networked to share ideas and equipment.
• Coaches were trained in health promotion and drug prevention.
• High school athletes were paired with high-risk elementary school youth by the Glouster Prevention Network.
• The Detroit Recreation Department initiated a literacy program by establishing reading centers and requiring Little League teams to gain points for book reports turned in.
• In the Athletes Reaching Out program, 47 professional and Olympic athletes visited schools to talk about the importance of a healthy, drug-free lifestyle.
• The Dedham, Massachusetts, coaches hosted a Family Unity Day that taught parenting skills (e.g., limit setting, monitoring, role modeling) while the kids attended a movie.

The effectiveness of this approach probably depends on three major factors: (a) Is the sport or recreational activity associated with role models that promote drug nonuse or use? (b) Is the sport a supervised activity that will involve the youth with nonusing, supportive coaches who will promote social competencies and skills? and (c) Is the sport or recreational activity one that teaches anger management or team work and cooperation under stress?

Sections

Tutoring or Mentoring Programs for High-Risk Youth

This selective or indicated approach can be implemented in schools or community settings with high-risk youth or disadvantaged youth, but because of the extra cost and need for staff time to coordinate, it is more often found in community agencies employing volunteers. Mentoring programs attempt to convey positive values, attitudes, and life skills through a one-to-one relationship with a positive role model, who may be a culturally matched community volunteer, college student, parent, senior citizen, or business professional. Cross-generational tutoring and mentoring using senior citizens is becoming very popular.

Possibly because of the difficulty in recruiting, adequately training, supervising, and matching mentors to youth, the mentoring approach is not used as frequently as other approaches. It was found in only 14% of the first 75 CSAP high-risk youth grants. Some intergenerational mentoring programs using retired or older persons have been effectively implemented and evaluated with CSAP funds, such as Full Circle implemented in four urban and rural sites in Colorado and Across Ages (LoScuito, Rajala, Townsend, & Taylor, 1996). To facilitate recruitment, Partners in Denver combines Big Brother/Big Sister programs with wilderness outings.

Mentoring programs, such as the OJJDP Juvenile Mentoring Program (JUMP) and Big Brothers/Big Sisters, are described as promising programs for the prevention of delinquency (Grossman & Garry, 1997). However, according to Bellamy and associates (1997), "Evidence available from evaluations on mentoring programs consistently indicates that non-contingent, supportive mentoring relationships do not have desired effects on outcomes such as academic achievement, school attendance, drop-out, various aspects of child behavior (e.g., misconduct), or employment" (p. 50). The type of volunteer mentor or whether mentors were paid or not made no difference on effectiveness outcomes.

One study of the Buddy System mentoring program in Hawaii found significant positive effects on truancy (Fo & O'Donnell, 1974). This study used a true experimental design, randomly assigning 11- to 17-year-olds with behavior problems to three experimental groups (relationship-only, contingent social approval, and contingent social and material reinforcement) or a control group. The mentoring groups using contingent reinforcement were more effective. Programs that support sustained, long-term relationships are more likely to be successful, as suggested by Project RAISE implemented through community organizations (e.g., churches, universities, fraternities, and businesses) in Baltimore with 6th graders identified as reading below grade level and at risk for school dropout (McPartland & Nettles, 1991).

Because academic achievement is a protective factor for drug use, and academic tutoring has been found to lead to decreased peer rejection, fewer disruptive behaviors, and significant gains in both math and reading skills (Coie & Krehbiel, 1984), tutoring programs are also used to reduce future drug use in high-risk youth with academic problems. Some of these programs implement a cross-age approach that uses high-risk students to tutor younger students. Other programs use mentors, college students, or teachers to tutor students in school or after school. Tutoring in phonics and math is a promising strategy to be implemented in the CSAP Developmental Predictor Variable cross-site studies, particularly by Tolan and associates in Chicago and Kumpfer and associates in Salt Lake City.

Sections

Cultural Competency or Pride Programs

Many targeted or selective prevention programs developed for ethnic youth and families have been based more on practitioners' responses to ethnic community needs than on theoretical advances involving ethnic youth (Kim, Coletti, Williams, & Helper, 1995). Kim and associates (1995) state: "Prevention practitioners frequently place more emphasis on the personality styles of the fieldworkers and their level of commitment than on a particular theoretical orientation" (p. 312). Promising approaches being tested with different ethnic groups are described below and in Drug Abuse Prevention with Multiethnic Youth (Botvin, Schinke, & Orlandi, 1995). The theoretical models that could inform these programs include acculturation theory, differential family acculturation, and orthogonal cultural identification (OCI) theory (Oetting & Beauvais, 1991).

Although etiological studies have not found a consistent relationship between cultural pride and substance use, the research of Oetting (1992) suggests that youth with more multicultural competencies are less vulnerable to substance use. Many of the high-risk youth programs implemented with ethnic youth include some sessions on cultural pride or history and teaching of cultural dances, art, and theater. They seek to employ attractive role models to increase cultural pride. Intensive experiential wilderness programs and ongoing youth clubs organized around activity programs have been evaluated in CSAP high-risk youth programs with ethnic students, primarily Native American youth, and found to be culturally relevant and effective in reducing risk factors (CSAP, 1993, Signs of Effectiveness).

Particularly popular with African American youth and families, these programs focus on skills training, but through a theoretical approach involving building resiliency in a number of live domains. Some rites of passage programs include only parent training, such as the Harambee or Mulicultural Violence Prevention Program developed by Marina and Steele, or youth only, such as the Rites of Passage Program developed by Ron Johnson. These programs stress the development of responsibility in youth and acceptance of youth as adult members of a community. The importance of the youth's link with the community and spiritual values is stressed. Some of these prevention programs targeting selective or indicated youth populations include group discussions, and others use more skills training and competency development. The Comprehensive Afro-American Adolescent Services Project operated by the University of Cincinnati included twenty 2-hour rites of passage discussion groups with community action projects and entrepreneurial training.

One promising activity for Native American youth is to include storytelling as an effective tool for cultural competence in prevention activities. At a substance abuse prevention conference sponsored by the Arizona Department of Health and Wholistic Health Education and Empowerment for Life (WHEEL). In more traditional cultures, storytelling is used to teach cultural values, provide role models, and pass on cultural identity and character traits needed for youth to succeed in life. The values taught by stories include finding inner strength to face adversity and building life in harmony within society and the natural world. Storytelling is used in all cultures as a means of teaching children crucial values and providing examples of what happens if one does not follow the traditional rules. This promising approach warrants evaluation because of its traditional wisdom. Some school curricula or family programs with young children, such as the Strengthening Families program (Kumpfer et al., 1989) incorporate some story telling.

Sections

Family-Focused Selective Prevention Approaches

Parent or family skills training programs have been effective in addressing youth problems such as conduct disorders and aggression (McMahon, 1987) or delinquency (Lipsey, 1992) and in reducing individual and family risk factors for drug use in children of drug-abusing parents in treatment (Catalano, Haggerty, & Gainey, in press; DeMarsh & Kumpfer, 1985; Kumpfer, 1993) or not in treatment (Kumpfer, 1990). Most parenting and family skills training programs can demonstrate immediate reductions in precursors of negative child behaviors, such as depression, aggression, conduct disorders, poor family management, intentions to use tobacco and alcohol (Bry, 1983; DeMarsh & Kumpfer, 1985), and school achievement and delinquency in preadolescents (Fraser, Hawkins, & Howard, 1986; Patterson, Chamberlain, & Reid, 1982). In a follow-up study, Tremblay and associates (1995) reported that parent training combined with child social skills training for kindergarten boys reduced school adjustment problems and delayed the onset of delinquency.

A number of selective family-focused interventions have been effective with at-risk youth and families. For an overview of programs effective for multiethnic youth and families, see Kumpfer and Alvarado, 1995, or Kumpfer, 1993, and 1-page descriptions of effective family programs on OJJDP's Strengthening America's Families Web site (http://www-medlib.med.utah.edu/healthed/ojjdp.htm). The NIDA (1997) publication discusses two family-focused selective prevention programs: Strengthening Families Program (Kumpfer et al., 1989) and Focus on Families (Catalano, et al., in press), which are described below, as well as three promising CSAP family-focused programs.

SFP is a multicomponent selective prevention program designed primarily for 6- to 10-year-old children of substance abusers; it also has been used with high-risk youth and families not abusing alcohol or drugs (Kumpfer et al., 1989). This 14-session program includes three separate curricula, the behavioral parent training program, a children's social skills training program, and a family skills training program designed to cover similar content at the same time. For instance, parents and children both learn about communication and problem solving, and in the last hour they practice together in two multifamily groups. This program is one of the most independently evaluated and culturally modified substance abuse prevention programs for high-risk ethnic youth and families.

In seven true randomized intervention trials or demonstration/evaluation research projects, SFP has been found to be very robust with similar positive results on youth's drug use and mediators for drug use (i.e., parenting efficacy; parenting skills of monitoring, supervision, and discipline; parent depression and drug use; parent/child relationships; children's problem behaviors; depression; aggression) and tobacco and alcohol use in the older children of drug abusers (for overviews see Kumpfer, Williams, & Baxley, in press, or Kumpfer, in press). This program has been effective with urban African American families (Aktan, Kumpfer, & Turner, 1997), rural African American mothers, Asian and Pacific Islander families, Hispanic families, and rural high-risk preteens (Kumpfer, Molgaard, & Spoth, 1996). In a 5-year follow-up study of 78 families interviewed by an independent Harvard psychiatrist, over 90% of the families reported that the positive changes they had made in communication, discipline, and family relations were still continuing. A large percentage (64%) were still having family meetings at least monthly, and 38% were holding weekly family meetings to continue the program intervention. Because program effects were found to decrease gradually with time, booster sessions are recommended and are being developed.

In this new selective prevention program, parents on a family retreat receive methadone maintenance treatment of 5 hours as well as 32 parent training sessions of 1.5 hours each with case-management services and booster sessions (Catalano et al., in press). Children attend 12 of the sessions to practice skills with their parents. The sessions address risk and protective factors that mediate drug use such as family goalsetting, relapse prevention, family communication, family management, creation of family expectations about alcohol and other drugs, and teaching of social skills (i.e., problem solving, decision making, peer resistance skills, and help in succeeding in school). The intensive program has a primary goal of reducing relapse in the parents. Because this program is still being evaluated, only the early results are available, which indicate that the parents' drug use and relapse reduced significantly in the first year, but not by the second year as compared to a group without the program. The effects on the children by the first year were nonsignificant but may increase with time.

The Substance Use Program and Early Resource II program (SUPER II) is a selective family skills training program for high-risk urban youth, ages 11 to 17, conducted in Atlanta (Bruce & Emshoff, 1992). It involves parents and youth in a 2- to 3-week program of eight 2-hour sessions including educational, experiential, and skills-building activities focusing on drug education and the law, communication skills, and parenting. All sessions but three were family sessions. Sessions three and four were separate parent sessions on "Being a Strong, Caring Parent" and "Building Your Child's Self-Esteem" and for youth, "Feeling Good About Being You/Thinking for Yourself" and "How to Say 'No' and Mean It." Session eight was a graduation and posttest attended only by the youth. The program was delivered by adults, peers, and police officers (session two on the law only). Retention mechanisms included meals or snacks, transportation, and child care. Prizes for attendance and participation also were given. Results were quite positive with pre- and posttest improvements in parent and youth knowledge of drugs and good communication. Parents (N = 132) reported significant improvements in family functioning and increase in positive esteem of youth. The 214 youth who took the posttest and the 98 who took the 3-month follow-up test reported significant decreases in the frequency and amount of drug use and the physical effects of use. The primary limitation of this program is its nonexperimental design; it has no control group. Because drug use generally escalates during the junior high and high school period, the results are probably valid if students were reporting accurately.

A very different approach to improving family risk and protective factors for substance abuse involves improving the parent and child relationship through play and arts programs, some including art therapy. A search of the research literature on play or art therapy and its application to substance abuse prevention produced very few citations. This mirrors Phillips (1985) observation that "a comparable body of methodologically adequate play therapy research has not emerged" (p. 752). Likewise, McNiff (1987) reported that art therapy did not have a substantial literature of empirical research results. Art therapy is used with COSAs for diagnostic purposes and to help them demonstrate their feelings (Robinson, 1989).

The 12-week CODA creative art therapy program in Downey, California, found highly significant positive effects on parent-reported pre-post Achenbach CBCL scores in 23 girls and 30 boys in areas of competencies, depression, and hyperactivity, and also in delinquency and aggression in the boys only. The program included parent-only sessions, child-only sessions after school, and evening family art therapy sessions to improve the dosage. Given the large effect size, the evaluators (Springer, Phillips, Phillips, Cannady, & Kerst-Harris, 1992) conclude: "CODA represents a program model that demonstrates strong promise for intervention with children of chemically dependent families" (p. 73).

This model represents a less rigorous approach to family interaction and works with possibly less risky children - African American kindergartners living in the Robert Taylor Homes in Chicago. Although environmentally these children are at great risk, they do not have the same level of emotional and behavioral risks and problems that the COSAs did in the prior CODA program. The author was the cross-site evaluator for both of these programs, and although this program was a "little piece of heaven" for these children who had no safe place to play outside, the program was less intensive than the CODA program and less focused on risk and protective factors. As a consequence, this program, which provided after-school parent/child art activities, parent education classes, and monthly family outings, produced no statistically significant findings among 32 program participants and 22 comparison group members. However, the model warrants additional research with larger numbers, random assignment, and a longer evaluation period because there was suggestive evidence of positive impacts on the mothers returning to school and seeking alcohol and drug treatment (Ruch-Ross, 1992).

These programs attempt to improve child development early, even prenatally, by providing nurse, social worker, or paraprofessional in-home services to high-risk or indicated parents. The outcome results are promising (Yoshikawa, 1994), but there are very few randomized clinical trials.

One of the programs with the strongest outcome results, the Nurse Home Visitation Program or the Prenatal/Early Infancy Project (Kumpfer, Alexander, McDonald, & Olds, in press; Olds & Pettit, 1996), has been funded by OJJDP to be replicated in a number of sites. This program (Olds, Henderson, Kitzman, & Cole, 1995) reports significant reductions in child abuse (4% in experimentals compared to 10-19% in control families), greater positive maternal involvement with children, improved discipline, increased completion of high school by mothers, increased employment (84% higher), and delayed subsequent pregnancies with single, poor mothers (rates 43% lower). Moreover, the 2-year follow-up found a 35% decrease in the toddlers' emergency room visits and fewer physician visits for injuries, ingestions, and social problems. Maternal prenatal smoking and alcohol consumption were reduced, resulting in decreased intellectual impairment in 3- and 4-year-olds and probably later decreases in conduct disorders and substance abuse (Olds, Henderson, & Tatelbaum, 1994). Based on these positive results, this program has been incorporated into some Housing and Urban Development (HUD) Public Housing Drug Elimination Programs.

The Success By Six program involving in-home paraprofessionals providing case management services to poor, high-risk families is being implemented through the United Way in over 100 communities. The results of this program suggest significant positive improvements in access to community services, decreases in basic needs, and self-reported improvements in family strengths (Kumpfer, Alvarado, & Kendall, 1997). Home visitation programs such as the Houston Parent-Child Development Center program (Johnson & Walker, 1987) have resulted in long-term positive outcomes such as improved behaviors in participating children when they were 8 to 11 years old compared to controls and reduced juvenile delinquency by age 15 as found for the Family Development Research Project (Lally, Mangione, & Honig, 1988). Other exemplary and promising family-strengthening approaches to prevention can be found in Kumpfer (1993, 1997) and Ashery and Kumpfer (NIDA, in press).

Sections

Summary of Selective Approaches

It has been much more difficult to locate outcome effectiveness results for selective prevention approaches because they are newer or have not been evaluated in randomized Phase III Controlled Intervention Trials, but in demonstration/evaluation grants frequently using nonexperimental designs. Because of the popularity of these selective programs, the most promising warrant further examination of their effectiveness.

Sections