The preceding chapter focused on preventive interventions that target change in the systems that most influence the cognitive, emotional, and behavioral development of young people: the family, schools, and the community. This chapter explores available preventive interventions that are targeted at specific mental, emotional, and behavioral (MEB) disorders. Many of these are designed to address the specific risk and protective factors associated with those disorders, although some also target risk factors that are common to multiple disorders.
The disorders targeted by preventive interventions tend to emerge at different development stages; for example, anxiety begins to emerge at a relatively young age, whereas schizophrenia tends to emerge closer to adolescence and young adulthood. Depression, eating disorders, and substance use and abuse tend to become a significant problem in the middle and high school years. The chapter organizes discussion of disorder-specific interventions in terms of the order in which they tend to appear in the developmental course of young people’s lives. Many of the interventions discussed in the previous chapter include among their outcomes improvements in one or more disorders, particularly externalizing disorders (e.g., substance abuse, conduct disorder, attention deficit hyperactivity disorder [ADHD]) (see Box II-1). Those results are not repeated here. Similarly, other low-frequency disorders for which little preventive literature is available, such as bipolar disorder, autism spectrum disorder, and pervasive developmental disorders, are not discussed.
The chapter also includes interventions targeted at mental health promotion, including strategies related to fostering positive development among children and adolescents and to modifying lifestyle factors that have been associated with a range of MEB disorders. The programs described here are delivered across mental health, physical health, and school settings and have involved intervention directly with children, with parents, and with the whole family. The chapter closes with conclusions and recommendations based on the evidence presented in both Chapter 6 and Chapter 7.
PREVENTION OF SPECIFIC DISORDERS
Prevention of Anxiety
Anxiety symptoms and disorders typically emerge in childhood (see Chapter 2); lifetime rates of anxiety disorders by adolescence may be as high as 27 percent (Costello, Egger, and Angold, 2005). Anxiety disorders typically precede depression and may contribute to its development (Wittchen, Beesdo, et al., 2004). Although a number of studies have shown the effectiveness of cognitive-behavioral therapy (CBT) in treating anxiety disorders in children and adolescents (Barrett, 1998; Kendall, 1994; Kendall, Safford, et al., 2004; Manassis, Mendlowitz, et al., 2002; Mendlowitz, Manassis, et al., 1999), and there is some evidence of the benefits of anxiety prevention for college-age individuals with anxiety symptoms (Schmidt, Eggleston, et al., 2007; Seligman, Schulman, et al., 1999), relatively little research has been done on the prevention of these disorders. However, Bienvenu and Ginsburg (2007) recently reviewed evaluations of anxiety preventive interventions, most of which were conducted in Australia. All of the interventions are variants of CBT applied to prevention, and most involve parents in some way.
Rapee (2002) and Rapee, Kennedy, et al. (2005) report a selective intervention for 3- to 5-year-olds whose behavior was inhibited according to parent and child reports and a behavioral assessment. Parents were randomly assigned to a no-intervention control condition or to an intervention involving six 9-minute group sessions that taught them how to practice gradual exposure and techniques for dealing with different situations, such as entering school. At 12-month follow-up, the intervention group children had a significantly lower prevalence of anxiety disorders, although there was no effect on parental or maternal ratings of inhibition or inhibition as assessed through behavioral testing.
Barrett and colleagues conducted several studies of universal interventions to prevent anxiety problems among children and adolescents (Barrett, Lock, and Farrell, 2005; Barrett and Turner, 2001). The interventions consist of 10–12 classroom sessions and 4 parent sessions guided by a framework called FRIENDS: Feeling worried; Relax and feel good; Inner helpful thoughts; Explore plans; Nice work, reward yourself; Don’t forget to practice; and Stay calm for life (Barrett, Lowry-Webster, and Turner, 2000). Barrett and Turner (2001) randomized 489 children ages 10–12 to one of three conditions: (1) usual care, (2) the program led by a teacher, or (3) the program led by a psychologist. Those assigned to the active interventions had significantly fewer anxiety symptoms at the end of the intervention. In other studies, the program reduced the proportion of 10- to 13-year-olds who were at risk for anxiety problems (Lowry-Webster, Barrett, and Dadds, 2001) and at 12-month follow-up had significantly lowered anxiety among sixth and ninth grade students (Barrett, Lock, and Farrell, 2005). There was some evidence that the intervention produced greater reductions than the control condition for the high- and moderate-risk groups (Barrett, Lock, and Farrell, 2005).
Dadds, Spence, and colleagues (1997) evaluated an indicated intervention for 7- to 14-year-olds who had anxiety symptoms or who met criteria for an anxiety disorder but did not have severe problems. The intervention followed Kendall’s FEAR strategy: Feeling good by learning to relax, Expecting good things to happen, Actions to take in facing up to fear stimuli, and Rewarding oneself for efforts to overcome fear or worry (Kendall, 1994; Bienvenu and Ginsburg, 2007). The intervention was provided to young people in 10 weekly group sessions; three sessions were provided to help parents learn to manage their own anxiety and to model and encourage their children’s use of the strategies. Six months after the intervention, young people in the intervention group had significantly fewer anxiety disorders than controls (16 compared with 54 percent). The difference was not significant at one-year follow-up, but it was at two-year follow-up (20 compared with 39 percent).
Schmidt, Eggleston, and colleagues (2007) report on a randomized trial of a selective intervention predicated on evidence that sensitivity to anxiety—the fear people have of having anxiety symptoms—is a predictor of the development of anxiety problems. Participants who were high in anxiety sensitivity were randomized to a brief intervention that taught about the symptoms of anxiety and the fact that they are not harmful. Participants were recruited from a university, the community, and local schools, with an average age of 19.3 years. Compared with the no-intervention group, participants had reduced concerns about the physical and social consequences of anxiety by the end of the program, although the effect was not maintained at follow-up. Intervention participants were also significantly more comfortable than control participants when exposed to a CO2 challenge that elicits anxiety, and significantly fewer had developed anxiety disorders one to two years after the intervention.
Seligman, Schulman, and colleagues (1999) used a randomized design to test an intervention consisting of 10 two-hour group sessions with 231 university students selected on the basis of their pessimistic views compared with controls. The sessions focused on changing cognitions, for example, replacing automatic negative thoughts with more constructive ones. At three-year follow-up, participants had experienced significantly fewer episodes of generalized anxiety disorder and fewer moderate (but not severe) depressive episodes than controls.
Although the preventive interventions for anxiety disorders evaluated to date are all based on CBT approaches, recent research suggests that these approaches may not be optimal (Biglan, Hayes, and Pistorello, 2008). Growing evidence suggests greater effectiveness for acceptance-based interventions (Hayes, 2004; Hayes, Luoma, et al., 2006), which teach people to accept anxiety as a normal part of living a value-focused life. Support for this approach also comes from evidence that efforts to control unwanted thoughts and feelings may exacerbate them (e.g., Wegner, 1992, 1994). Additional research is needed to develop and evaluate preventive interventions based on acceptance-based approaches and to determine the effectiveness of these approaches relative to traditional CBT.
Prevention of Posttraumatic Stress Disorder (PTSD)
Although it appears plausible that providing some sort of counseling to all trauma victims could prevent PTSD, empirical research has not shown this to be the case. Critical incident stress debriefing (CISD) is a technique widely used to prevent adverse reactions to trauma. As soon as possible after the traumatic event, victims are encouraged to discuss the details of their experience, their emotional reactions, any actions they have taken, and any symptoms they have experienced. They are reassured that their reactions are normal, told of adverse reactions that are typical, and encouraged to resume usual activities. The intervener tries to assess whether any adverse reactions have occurred and, if so, refers the person for further assistance. Typically there is a follow-up contact with the victim. Recent research found that CISD is ineffective and possibly harmful (American Psychiatric Association, 2004). A meta-analysis found no benefit from its use and suggested a detrimental effect compared with no intervention or minimal help (van Emmerik, Kamphuis, et al., 2002).
In contrast, randomized controlled trials of CBT for individuals who are symptomatic in the weeks after a trauma reveal significant efficacy (Boris, Ou, and Singh, 2005). Some evidence suggests that this includes children (Chemtob, Nakashima, and Hamada, 2002).
In a quasi-randomized controlled trial, Berger, Pat-Horenczyk, and Gelkopf (2007) evaluated a school-based intervention consisting of an eight-session structured program designed to prevent and reduce children’s stress-related symptoms, including PTSD. Compared with the wait-list controls, the study group reported significant improvement on all measures. Finally, there is some evidence that adolescents who maintain their routines have less posttraumatic stress (Pat-Horenczyk, Schiff, and Doppelt, 2006), a finding consistent with other findings that catastrophizing puts individuals at risk for developing PTSD (Bryant and Guthrie, 2005).
Prevention of Depression
In 1994, when the Institute of Medicine (IOM) report Reducing Risks for Mental Disorders: Frontiers for Preventive Intervention Research was released, available trials of interventions targeting depression were able to demonstrate only a reduction in symptoms (Muñoz and Ying, 1993). Since that time, methods have been developed for consistently identifying individuals at significant risk of experiencing depression within the next year, and some trials have demonstrated a reduction in the incidence of major depressive episodes, particularly among those at high risk (Muñoz, Le, et al., 2008). Of the trials that have shown a significant reduction in new episodes, all have focused either on high-risk adolescents (Clarke, Hawkins, et al., 1995; Clarke, Hornbrook, et al., 2001; Young, Mufson, and Davies, 2006) or pregnant women (Elliott, Leverton, et al., 2000; Zlotnick, Johnson, et al., 2001; Zlotnick, Miller, et al., 2006), and at least one intervention prevented episodes among those who had prior episodes (Clarke, Hornbrook, et al., 2001). On the basis of these advances, Barrera, Torres, and Muñoz (2007) assert that prevention of depression is a feasible goal for the 21st century, with the promise of being able to reduce incidence by as much as half.
Preventive Interventions for Children and Adolescents
Recent meta-analyses have concluded that interventions to prevent depression can reduce both the number of new cases in adolescents (Cuijpers, van Straten, et al., 2008) and depressive symptomatology among children and youth (Horowitz and Garber, 2006). In a review that included seven trials targeting adolescents, Cuijpers and colleagues (2008) report that preventive interventions for adolescents can reduce the incidence of depressive disorders by 23 percent. They caution, however, that since the follow-up period in most studies did not exceed two years, the projects may have delayed onset rather than incidence. Both meta-analyses showed slightly higher effect sizes for selective and indicated interventions, although the number of universal interventions was very small.
Significant benefit has been reported for preventive interventions for reducing depressive symptoms in children and adolescents, with small to modest effect sizes (Horowitz and Garber, 2006; Jané-Llopis, Hosman, et al., 2003). In a systematic review of preventive interventions with children and adolescents, Merry and Spence (2007) highlight several promising approaches. However, they also describe failed attempts to repeat results in real-world school and primary care settings, limited follow-up periods, and methodological flaws, and they conclude that there is not yet sufficient evidence of effectiveness for preventive interventions for depression. In an analysis of the high-quality studies reviewed by Horowitz and Garber (2006), Gladstone and Beardslee (in press) demonstrate that although symptom reduction, a powerful goal in itself, is possible, very few studies of adolescents have examined actual reduction in new episodes of major depression, the work of Clarke and colleagues cited above being the notable exception. They emphasize that future studies should examine prevention of episodes as well as reductions in symptomatology.
In the committee’s judgment, the balance of evidence suggests that some interventions can significantly reduce the symptomatology and incidence of depression. The potential to increase the sample sizes and reach of interventions has been highlighted by work done to adapt behavioral interventions to a range of settings and cultural groups, including conducting worldwide randomized controlled trials via the Internet (Muñoz, Lenart, et al., 2006).
The Clarke Cognitive-Behavioral Prevention Intervention (see Box 7-1), an indicated program targeting adolescents at risk for future depression, has successfully prevented episodes of major depression in several randomized trials. A recent replication indicated that it is not as effective for adolescents with a depressed parent (Garber, Clarke, et al., 2007). The Penn Resiliency Program (PRP) (see Box 7-2), a school-based group intervention that teaches cognitive-behavioral and social problem-solving skills to prevent the onset of clinical depression, has also had promising results.
Preventive Interventions for Families with Depressed Parents
Children of parents with depression and related difficulties have a substantially higher rate of depression than their counterparts in homes with no mental illness (Beardslee and Podorefsky, 1988; Hammen and Brennan, 2003; Lewinsohn and Esau, 2002; Beardslee, Versage, and Gladstone, 1998; Weissman, Wickramaratne, et al., 2006). They are also at risk for a variety of other difficulties in such areas as school performance and interpersonal relationships (Goodman and Gotlib, 1999). Beardslee and colleagues developed two public health preventive interventions (see Box 7-3) specifically aimed at providing information and assistance in parenting to children of depressed parents, both of which have shown positive results in multiple randomized trials.
PREVENTION OF SUBSTANCE USE AND ABUSE
Many of the interventions discussed in Chapter 6 have had effects on outcomes related to substance abuse. Additional intervention strategies specifically targeting prevention of substance abuse are discussed here. School-based programs with this focus emerge primarily in the middle school years, when initial risk for use is greatest.
Cuijpers (2002) reviewed three meta-analyses of classroom-based substance abuse prevention programs (Rooney and Murray, 1996; Tobler, Roona, et al., 2000; White and Pitts, 1998) and a set of studies that analyzed mediators of the effects of these programs. Their synthesis led to six conclusions about effective programs. First, programs that involve interactions among participants and encourage them to learn drug refusal skills are more effective than noninteractive programs. Second, interventions that focus on direct and indirect (e.g., media) influences on use of drugs appear to be more effective than those that do not focus on social influences. Third, programs that emphasize norms for and a social commitment to not using drugs are superior to those without this emphasis. Fourth, adding community components to school-based programs appears to add to their effectiveness (see also Biglan, Ary, et al., 2000). Fifth, use of peer leaders may enhance short-term effectiveness (see also Gottfredson and Wilson, 2003). Sixth, adding training in life skills to that in social resistance skills may increase program effectiveness (see also Faggiano, Vigna-Taglianti, et al., 2005).
A meta-analysis to assess potential moderators of program effectiveness by Gottfredson and Wilson (2003) determined that programs that can be delivered primarily by peer leaders have increased effectiveness. An analysis by Faggiano, Vigna-Taglianti, et al. (2005) found that the most effective programs are those focused on life and social skills. Skills-based programs increased drug knowledge, decision-making skills, self-esteem, and peer pressure resistance and were effective in deterring early-stage drug use.
Derzon, Sale, and colleagues (2005) report on an analysis of a 46-site, five-year evaluation of school- and community-based substance abuse prevention programs that included behavioral skills programs, information-focused programs, recreation-focused programs, and affective programs. Using a meta-analytic technique to project potential impact by accounting for methodological and procedural differences, they calculated a mean adjusted effect size of 0.24 for decreasing 30-day substance use (tobacco, alcohol, and marijuana).
Life Skills Training (see Box 7-4) is one of the most prevalent substance use prevention curricula in the nation’s public schools and has been endorsed as a model program by both the Blueprints for Violence Prevention and the Surgeon’s General’s Youth Violence Report. Another successful alcohol, tobacco, and marijuana preventive intervention for middle school students is Project ALERT (see Box 7-5). The Drug Abuse Resistance Education (DARE) Program, based primarily on scare tactics, has been found in multiple trials to be ineffective in its original form; a modified version is currently being tested.
College Interventions Targeting Prevention of Alcohol and Drug Use and Abuse
The evidence on alcohol and drug abuse prevention in colleges is limited and inconclusive because, although many colleges have such programs, very few studies have evaluated them (Larimer, Kilmer, and Lee, 2005). More robust evaluation has been done of interventions focused on reducing drinking among college students. Carey, Scott-Sheldon, et al. (2007) report on a meta-analysis of 62 interventions. They found that, although on average the interventions reduced alcohol consumption both immediately and at follow-up, the majority of studies failed to produce a significant effect. Variables associated with positive outcomes include motivational interviewing (MI, a nonconfrontational approach to asking students to describe their drinking behavior and its consequences), feedback about expectancies and motives for drinking, and decision-making procedures that prompt the individual to weigh the benefits and negative aspects of drinking. Skills training approaches were less effective, as were interventions for men and for those who were already drinking heavily.
An intervention reported by Carey, Carey, and colleagues (2006) did produce significant benefit. They evaluated MI as a means of reducing problematic drinking among 509 heavy-drinking undergraduates who were randomly assigned to one of six conditions. The students received one of two versions of MI or no interviews. The standard version of MI stressed the students’ autonomy in deciding what they wanted to do, discussed norms about drinking, provided tips for reducing drinking, and reinforced talk about change. The second, “enhanced” version included a worksheet containing a decisional grid to help students clarify the pros and cons of changing their behavior. Students were also assigned to receive or not receive a Timeline Follow Back (TLFB) interview that took the students back through the previous 90 days, starting with the most recent period, and helped them reconstruct their drinking behavior during this time. Assessment of the students’ drinking behavior and alcohol-related problems occurred at baseline and 1, 6, and 12 months postintervention. They found that the TLFB by itself reduced alcohol consumption compared with the no-intervention control. The standard MI produced significantly greater reductions in alcohol use and alcohol problems than did the TLFB; those who received the enhanced MI did not improve as much. On the basis of this evidence, motivational interviewing coupled with the TLFB appears to have the greatest potential to reduce drinking significantly among undergraduates.
In addition to school-based and college interventions, efforts to prevent substance use and abuse among young people often include other community, media, regulatory, or policy approaches. These more broadly based strategies tend to target norms and policies rather than trying to reach individuals with behavior change strategies, although in many cases they are combined with components that target individuals more directly through schools and families. Many of these interventions, particularly those targeting alcohol, also focus on reducing the consequences of substance use as much as use itself.
The Centers for Disease Control and Prevention’s Guide to Community Preventive Services (n.d.) recommends restrictions on outlet density and zoning to reduce excessive alcohol consumption and enhanced enforcement of laws prohibiting the sale of alcohol to minors. Nationally oriented recommendations related to reducing and preventing underage drinking call for these and other approaches, such as limiting the marketing of alcohol and specifically youth-oriented alcohol products, use of media campaigns targeted at parents, and creation of community coalitions; two policy reports also call for continued research on developmental considerations and early alcohol use (National Research Council and Institute of Medicine, 2002; U.S. Public Health Service, 2007). The Task Force on College Drinking concluded that evidence was strongest for indicated interventions that included cognitive skills training, norms or values clarification, motivational enhancements, or challenging of expectancies, but it recommended comprehensive integrated community coalitions targeting individuals, the student population as a whole, and the college and surrounding community (National Institute on Alcohol Abuse and Alcoholism, 2001).
A review of interventions in nonschool settings designed to prevent substance abuse among those under age 25 found insufficient evidence to draw conclusions about the effectiveness of these programs (Gates, McCambridge, et al., 2006). The authors were able to identify only 17 randomized controlled trials, which varied greatly in their program components and included four types of interventions: MI or brief interventions, education or skills training, family interventions, and multicomponent community interventions. Some interventions, including three family interventions, MI, and two interventions with both community and school components, showed potential benefit in reducing marijuana use. Compared with the more robust data on school-based substance abuse prevention programs, existing research is insufficient to determine the effectiveness of efforts to prevent substance abuse through interventions in other settings.
A review of the impact of universal prevention programs on alcohol use (Foxcroft, Ireland, et al., 2002) found a lack of clear evidence for effectiveness in the short or medium term. This analysis, which included school-based, family, and community interventions, found the most promising effects for long-term outcomes of a culturally focused school and community skills-based intervention with American Indians, which reduced the likelihood of weekly drinking over 3.5 years, and the Strengthening Families Program (described in Chapter 6), which reduced alcohol initiation behaviors over four years.
Almost none of the community interventions aimed at preventing adolescent tobacco, alcohol, or other drug use have been in the subject of more than one experimental evaluation. However, the emphasis on these more broad-based approaches in national recommendations and the progress that has been made since 1994 in this area warrant some discussion of a few example programs that include a significant community and policy component.
The Midwestern Prevention Program (MPP), a multimodal community-wide drug prevention program, evaluated effects on high-risk and general youth populations (Chou, Bentler, and Pentz, 1998;Johnson, Pentz, et al., 1990; Pentz, MacKinnon, et al., 1989a, 1989b; Pentz, Trebow, et al., 1990). The intervention included the following: (1) classroom curriculum targeting students in sixth and seventh grades, (2) parent training addressing prevention policy and parent–child communication skills, (3) training of community leaders in development of a drug abuse prevention task force, and (4) media promotion of prevention policies and norms (Pentz, MacKinnon, et al., 1989b). The intervention was evaluated in a quasi-experimental trial and a subsequent experimental trial. In the formal trial, the intervention was equally effective for both high- and low-risk youth (Johnson, Pentz, et al., 1990). In the latter trial, there was significantly less tobacco and marijuana (but not alcohol) use in the MPP schools than in control schools, with effects found primarily in private and parochial schools (Pentz, Trebow, et al., 1990); through 3.5 years postbaseline, the percentage of students with reports of substance abuse during the past month declined from one assessment to the next (Chou, Bentler, and Pentz, 1998). MPP produced significant declines in cigarette, alcohol, and marijuana use across all follow-ups. There were limited effects for baseline marijuana users and diminishing effects for early alcohol and cigarette users over time.
Project Northland was a multimodal intervention aimed at delaying the onset of and reducing underage drinking (Perry, Williams, et al., 1996; Perry, Williams, and Komro, 2000, 2002). It was initially evaluated in a randomized trial of 24 small Minnesota communities and subsequently in a randomized trial in Chicago inner-city schools. The intervention included social-environmental approaches and individual behavior change strategies along with community organizing, youth action teams, print media regarding healthy norms about underage drinking, parent education and involvement, and classroom-based social-behavioral curricula. In the Minnesota trial, alcohol use was prevented among 8th grade students, and those who were not using alcohol at the beginning of the project reported significantly less alcohol, marijuana, and cigarette use at the end of 8th grade. The effects were not maintained by the time students were in 10th grade. The results were not replicated in the Chicago trial (Komro, Perry, and Veblen-Mortenson, 2008).
Other programs have focused primarily on changing community policies and norms. Communities Mobilizing for Change on Alcohol developed a social-environmental intervention to reduce underage alcohol access through changes in policies and practices of major community institutions (Wagenaar, Murray, et al., 2000). Strategy teams comprised community groups and organizations focused on decreasing the number of alcohol outlets selling to youth, reducing access to alcohol from noncommercial sources (e.g., parents, siblings, peers), and changing cultural norms that tolerate underage access to and consumption of alcoholic beverages. Fifteen communities in Minnesota and Wisconsin were randomized into intervention or control groups. The intervention reduced youths’ commercial access to alcohol and arrests for driving under the influence of alcohol among 18- to 20-year-olds (Wagenaar, Murray, et al., 2000).
Two quasi-experimental studies have also shown benefits in reducing alcohol-related problems. The Community Trials project reduced alcohol-related injuries and deaths among all age groups through community-wide environmental prevention activities and policy change (Holder, Saltz, et al., 1997). The study matched but did not randomize communities in California and South Carolina. In the intervention communities, the following were targeted: (1) community mobilization, (2) responsible beverage service, (3) increased enforcement of drunk driving laws and perceived risk of drunk driving detection, (4) reduced underage access, and (5) reduced availability of alcohol through the use of local zoning and other municipal controls on outlet quantity and density. The intervention produced significant reductions in nighttime injury crashes, alcohol-related crashes, assault injuries, and hospitalizations. Adults reported lower rates of drinking and driving, and sales of alcohol to minors were reduced. Adolescent alcohol use was not assessed.
Saving Lives (Hingson, McGovern, et al., 1996) aimed to reduce alcohol-impaired driving and related risks. The study compared six Massachusetts intervention communities and five control communities using a quasi-experimental design. The intervention involved a task force that designed specific activities for its community, including business information programs, media campaigns, speeding and drunk driving awareness days, high school peer-led education, speed-watch telephone hotlines, and police training. During the five years of program activity there was a 25 percent decline in fatal crashes and a 25 percent decrease in fatal crashes involving alcohol compared with the prior five years.
In contrast with the positive results of media messages related to smoking, however, evaluations of the National Anti-Drug Media Campaign have yielded mixed results. While there is some evidence consistent with a favorable effect of the campaign on parent outcomes, there is no evidence that the effect on parents translates into improved outcomes for their children (Orwin, Cadell, and Chu, 2006).
Derzon and Lipsey (2002) reviewed 72 studies evaluating the effects of a broad range of media interventions on substance use behavior, attitudes, or knowledge. Using pre-post gain effect size statistics, they found positive effects for those receiving media interventions compared with controls, including smaller increases in substance use, greater improvement in substance use attitudes, and larger gains in substance use knowledge. Intervention characteristics consistently associated with greater gains include communications directed at parents and other adults with influence over young people; messages communicated by video (compared with television, radio, or print); and the use of supplementary components, such as group discussion, role play, or supportive services. The authors acknowledge significant methodological challenges for both the research evaluating media interventions and the meta-analysis, and the effect sizes they found were small. However, they conclude that media interventions can be effective, and that the wide reach of such interventions can potentially translate a small effect into significant cumulative changes for large numbers of young people.
Prevention of Eating Disorders
The lifetime prevalence of eating disorders, including anorexia nervosa, bulimia nervosa, and binge eating disorder, is relatively small, more common among females, and most likely to occur during the teen years (Stice and Peterson, 2007). In a meta-analysis of 53 randomized and quasi-experimental trials focused on prevention of eating disorders, Stice and Shaw (2004) found, on average, significant effects (generally small to modest) for each of the included dependent variables: body mass, thin-ideal internalization, body dissatisfaction, dieting, negative affect, and eating pathology. Some effects were detectable as much as two years after the intervention. The effect sizes were smaller for universal interventions, which included many participants not at risk for eating disorders. Didactic programs were less effective than those that engaged participants in interactions. Single-session programs were less effective than longer ones, and programs were more effective if they targeted those over age 15. Interventions that simply provided education about eating disorders were significantly less effective than other interventions for most outcomes. The effective interventions varied in content and included ones that focused on resistance to cultural pressure for thinness, addressed body dissatisfaction, and taught healthy weight management. A meta-analysis of five studies of Internet-based interventions to prevent eating disorders found no statistical significance for pooled outcome data but recommended additional research given the small number of studies (Newton and Ciliska, 2006). Stice and Shaw (2004) similarly point to the need for improved methodological rigor and theoretical rationale in order to progress from promising to conclusive interventions.
PREVENTION OF SCHIZOPHRENIA DURING A PRODROMAL STAGE
There has been limited work on early prevention of psychotic disorders. Given the severity of such disorders as schizophrenia and bipolar disorder (McFarlane, 2007) and their extraordinarily high associated lifetime risk for suicide (Palmer, Pankratz, and Bostwick, 2005) and early mortality (Fenton, 2000), it is essential to investigate opportunities for prevention before onset or when symptoms are in the prodromal stage (a period of nonpsychotic symptoms that precedes onset). Findings from a number of treatment studies of early detection and intervention indicate that both atypical antipsychotic drugs and psychosocial interventions are good candidates for testing in youth who are at high risk for a psychotic episode (Haas, Garrett, and Sweeney, 1998; Leucht, Pitschel-Walz, et al., 1999; Lieberman, Perkins, et al., 2001; Loebel, Lieberman, et al., 1992; Marshall, Lewis, et al., 2005; McFarlane, 2007; Pilling, Bebbington, et al., 2002a, 2002b).
To be effective, however, these preventive and early intervention strategies need to overcome some important challenges. First, epidemiological and developmental factors make it challenging to conduct universal, selective, or indicated preventive intervention trials aimed at those who have not yet had an episode (Faraone, Brown, et al., 2002; Brown and Faraone, 2004). Second, preventive intervention strategies are limited by incomplete understanding of the genetic, neurological, and environmental factors leading to these disorders. Third, ethical challenges are posed by the testing of interventions that may do harm and the stigma regarding labeling someone as being at high risk for psychosis. None of these challenges appears insurmountable, however. Moreover, the very high costs of these illnesses when they occur and the fact that experiencing the illness itself predisposes to more episodes make effort in this area warranted.
A number of prodromal clinics worldwide identify subjects from the community at high risk for a psychotic episode. These clinics provide training to mental health professionals, school and community professionals, and the public regarding early warning signs and opportunities for referral. Several are testing an active intervention, including early pharmacological intervention, against a control condition. The prodromal phase is characterized by schizoid characteristics or familial risk, brief or attenuated psychotic symptoms, and social deterioration or negative symptoms (McFarlane, 2007). The criteria used by these clinics to distinguish those in the prodromal phase from those who are not at elevated risk or have already had a psychotic episode are not identical across clinics. However, there is compelling evidence that those identified in such prodromal stages have a very elevated risk for experiencing a psychotic episode in the near future (Yung and McGory, 1996a, 1996b; Yung, McGorry, et al., 1996; McGlashan, Addington, et al., 2007).
The published studies of these preventive interventions indicate a substantial reduction in rates of development of frank psychosis and in prodromal and psychotic symptoms, although one study did not show statistical significance. Using a simple meta-analysis, McFarlane (2007) estimated that the mean conversion rate across studies is about 11 percent of treated cases and 36 percent of untreated or treatment-as-usual control cases.
Given the limitations of many of these studies and the risk of serious adverse events, the positive results found are not sufficient to recommend such interventions as a standard for practice. However, the interventions show considerable promise, and several studies are under way. Continued research in this area should be a high priority. The existence of standard criteria across multiple sites, such as in the North American Prodrome Longitudinal Study (a collaborative, multisite investigation into the earliest phase of psychotic illness), would be invaluable in conducting such research.
MENTAL HEALTH PROMOTION
Mental health promotion programs aim to improve positive outcomes among young people. Some programs share elements with universal prevention programs when they attempt to reduce negative emotional and behavioral outcomes as well as to improve positive mental health outcomes. As a natural consequence of shared risk and protective factors, mental health promotion and prevention strategies also have shared outcomes. As mentioned in Chapter 3, meta-analytic and qualitative reviews demonstrate significant overlap between the strategies, although promotion programs are distinguished by their primary emphasis on positive aspects of development, including developmentally appropriate competencies. This section first reviews interventions aimed at fostering positive development. It then examines lifestyle factors that promote mental health.
Interventions Aimed at Fostering Positive Development
A common feature of most validated programs aimed at fostering positive development and preventing the development of problems is the emphasis on supportive environments or “nurturance.” From the prenatal period through emerging adulthood, such interventions are supportive of individuals and their caretakers and provide positive reinforcement for prosocial behavior. Home visitors encourage young mothers to develop new skills, including how to comfort and interact warmly with their infant. Preschool teachers attend to, praise, and reward the developing skills of their children. The Good Behavior Game reinforces cooperative behavior among teams of children. Trainers praise parents for trying new skills in nurturing their children.
The creation of supportive environments also involves acceptance. Parents who are aggressive toward their children are not confronted; they are simply prompted to try more positive methods of being with their children (Webster-Stratton, 1990). College students who are drinking too much are gently questioned about their drinking and its consequences and are given tips for changing their behavior if they choose to do so. People who have been exposed to traumatic events are helped to accept that these events have happened and to move forward in their lives. Families struggling with parental depression are helped to understand and accept and to develop a shared approach to coping with it. Adolescents and young adults experiencing psychotic symptoms for the first time receive assistance in dealing with them.
In contrast to many punitive societal reactions to young people’s problem behavior, none of these interventions emphasizes punishment. The Good Behavior Game helps teachers reinforce desirable behavior and thereby reduce the behaviors that commonly draw punitive responses. Parenting programs help families replace harsh and inconsistent discipline practices with time-outs and brief removal of privileges, while parents are prompted to greatly increase positive reinforcement for desirable behavior. Several studies with families that have experienced major disruptions, such as marital separation and bereavement, have provided consistent evidence that the ability of such parenting programs to increase nurturance (warmth) and improve effective discipline accounts for their effectiveness in reducing internalizing and externalizing of problems in the short term and up to six years following the intervention (DeGarmo, Patterson, and Forgatch, 2004; Forgatch, Beldavs, et al., 2008; Tein, Sandler, et al., 2004, 2006; Zhou, Sandler, et al., 2008; Martinez and Forgatch, 2001). The principles of richly reinforcing desirable behavior and minimizing punishment are practices that may go a long way toward reducing problem behaviors among young people (see also Chapter 11).
Durlak and Wells (1997) reviewed 177 interventions targeted at reducing behavioral and social problems in children and adolescents, including both prevention and mental health promotion interventions. They found significant mean effects for programs that modified the school environment, helped children negotiate stressful transitions, and provided individually focused mental health promotion. Most of these programs both significantly increased competencies and significantly reduced problems.
Catalano, Berglund, and colleagues (2002, 2004) identified 25 youth development programs that focused on building positive constructs, such as social, emotional, and cognitive competence; self-determination; and self-efficacy. They concluded that the programs showed evidence of improving measures of positive development and reducing a range of problem behaviors, such as risky sexual behavior, alcohol and drug use, violence, and aggression. For example, Raising Healthy Children (Catalano, Mazza, et al., 2003), an extension of the successful Seattle Social Development Program, focuses on promoting positive youth development by improving classroom and family support for prosocial behavior. A trial matched 10 schools and randomized first or second grade students to the Raising Healthy Children intervention or a no-intervention group. At 18-month follow-up, program participants had higher teacher-rated academic performance and commitment to school, lower antisocial behavior, and higher social competency. Participants also showed less increase in the use of alcohol and marijuana in their middle school years (Brown, Catalano, et al., 2005).
Similarly, in a meta-analytic review of 237 school-based mental health promotion programs, Durlak, Weissberg, and colleagues (2007) reported improvements in aspects of positive development (e.g., social-emotional skills, prosocial norms, school bonding, positive social behavior), as well as reductions in problem outcomes (e.g., aggressive behavior, internalizing symptoms, substance use). Kraag, Zeegers, and colleagues (2006) reviewed 19 trials of school-based programs that teach coping skills or stress management through relaxation training, social problem solving, or social adjustment and emotional self-control. Although there was significant heterogeneity in methodological quality, they found large pooled effect sizes for both enhanced coping skills and reduced stress symptoms.
A recent evaluation by the RAND Corporation of a widely implemented after-school program, Spirituality for Kids, demonstrated a causal link between spiritual development and resilience. In a randomized trial involving 19 program sites, the program showed medium to large effects on positive behaviors, such as adaptability and communication, and small to medium effects on behavioral problems, such as attention problems, hyperactivity, and withdrawal (Maestas and Gaillot, 2008).
Embry (2004) has suggested that the dissemination of a set of simple behavior-influence procedures, or “kernels,” would be helpful for parents, teachers, health care providers, and youth workers in fostering positive development among children and adolescents. Examples include praise notes (Gupta, Stringer, and Meakin, 1990; Hutton, 1983; Kelley, Carper, et al., 1988; McCain and Kelley, 1993), peer-to-peer tutoring (Greenwood, 1991a, 1991b), the Beat the Timer game (Adams and Drabman, 1995), and some of the skills that are used in parent–child interaction therapy (Eyberg, Funderburk, et al., 2001) and other caregiver training approaches. Others have similarly called for the study of core components of programs to facilitate their implementation in schools and other community settings (e.g., Greenberg, Feinberg, et al., 2007). Discerning generic principles that are common to diverse interventions could foster their broader use.
Illustratively, because they achieve their preventive effects through promotion of family and child competencies, several programs discussed earlier in this report, including the Promoting Alternative Thinking Strategies (PATHS) curriculum (see Box 6-7), Fast Track (see Box 6-9), and Life Skills Training (see Box 7-4), as well as the Big Brothers Big Sisters Program (see Box 7-6) are frequently cited as successful promotion and prevention programs; they have also been recommended by Blueprints for Violence Prevention.
Lifestyle Factors That Promote Mental Health and Prevent Mental, Emotional, and Behavioral Disorders
Evidence from a small but growing set of observational and interventional studies indicates that modifications in a number of lifestyle factors, including sleep, diet, activity and physical fitness, sunshine and light, and television viewing, can promote mental health. Of these factors, the opportunity is perhaps strongest for the salutary effects of adequate sleep and certain nutritional elements, such as adequate iron content in the diet. In many cases, intervention studies related to lifestyle factors have documented physical health benefits. Given the strong connections between physical and mental health, improvements in both may be achievable using common approaches.
Attempts to modify lifestyle factors can appropriately be centered on families and the activities of the medical care community, promoted in the context of schools and community organizations, or accomplished through policy decisions. It should be noted that in many families, there are substantial barriers to promotion and prevention related to lack of knowledge, as well as factors that interfere with healthy decisions, such as poverty, neighborhood stresses, family tensions, and a general lack of child supervision.
While there is a commonsense element to interventions aimed at improving modifiable lifestyle factors, future efforts must rigorously document the promotion and prevention outcomes of their adoption. Promotion of mental health early in young people’s lives using such universal strategies that are feasible, inexpensive, and scientifically compelling holds great promise.
Sleep deprivation and sleep-related breathing disorder (SBD) are linked to emotional and behavioral problems that include hyperactivity, inattention, impulsivity, mood lability, and aggression (Institute of Medicine, 2006c; Rosen, Storfer-Isser, et al., 2004; Wolraich, Drotar, et al., 2008). Hyperactivity and attention disorders are associated with two other sleep disorders—restless leg syndrome and periodic limb movement disorders (Chervin, Hedger Archbold, et al., 2002).
Given that 20 percent or more of children have sleep problems, the contribution of SBD and other sleep problems to behavioral disorders is potentially enormous, though largely underrecognized. Interventions to improve sleep duration and quality must be rigorously assessed to determine their potential for improving emotional and behavioral outcomes. For example, a program to screen all children in primary care based on a history of snoring, interrupted sleep, and insufficient hours of sleep could be followed by a behavioral assessment using validated instruments and behavioral interventions as indicated. Studies are needed to demonstrate that the treatment of obstructive sleep apnea with tonsillectomy and adenoidectomy or other measures reduces the occurrence of behavioral consequences. A more general proposed approach to healthy sleep is the establishment of a multimedia public education campaign targeting specific populations, such as children, their parents, teachers in preschool and elementary school, college students, and young adults (Institute of Medicine, 2006c). The intent of such a campaign would be awareness concerning the consequences of insufficient or disrupted sleep, leading to identification of these problems and reestablishment of healthy sleeping patterns.
Diet and Nutrition
Adverse emotional and behavioral outcomes for children have long been linked to dietary factors. However, many suggested nutritional interventions have little or no evidence base. Prenatal nutrition was addressed in Chapter 6. Postnatal nutrition factors include hunger, undernutrition, and failure to thrive, which have been linked to cognitive and behavioral consequences (Dykman and Casey, 2003). Other factors that may be more modifiable include knowledge about optimal food intake and content, which can be addressed with education.
Breastfeeding has been studied extensively concerning its relevance to emotional and behavioral health. On the one hand, mounting evidence suggests that breastfeeding can contribute to enhanced cognitive capabilities independently of confounding factors (Kramer, Aboud, et al., 2008). While the IQ effect is modest in most studies, intelligence is a protective factor for MEB disorders and related problems. On the other hand, the weight of evidence at this time does not support superior behavioral outcomes for children who have been breastfed (Kramer, 2008). Based on current information, breastfeeding should be promoted for many reasons, but prevention of MEB disorders in childhood or in later life is not one of them.
Avoidance of nutritional deficiencies is important for promotion of mental health. High on the list of critical nutritional elements is iron. Children shown to have severe chronic iron deficiency in infancy score lower on measures of mental and motor functioning and are rated by both parents and teachers after 10 years of follow-up as more problematic in the areas of anxiety, depression, social problems, and attention problems (Lozoff, Jimenez, et al., 2000). This study is one of several that suggests an important relationship between iron deficiency and subsequent behavior. A concern, of course, is that iron repletion does not reverse long-term adverse outcomes and that iron deficiency remains very common in the United States (e.g., Schneider, Fuji, et al., 2005). U.S. Hispanic children and overweight children are particularly vulnerable (Brotanek, Halterman, et al., 2005). Strategies for avoiding iron deficiency include iron supplementation of exclusively breastfed babies (Dallman, Siimes, and Steckel, 1980), avoidance of prolonged bottle feeding (Brotanek, Halterman, et al., 2005), and routine testing of certain populations of infants for iron deficiency in the course of medical care. Given the magnitude of potential adverse outcomes, systematic efforts to inform parents of childbearing age about the importance of adequate iron intake for both mother and child should be adopted and sustained at the national level.
Attention has been focused for the past decade or two on the omega-3 fatty acid content of prenatal maternal diets and diets for children postnatally. Low levels of DHA and EPA—omega-3 fatty acid products—and corresponding high levels of arachadonic acid have been shown in animal studies to be detrimental to brain development (Innis, 2008) and are related to indices of brain inflammation (Orr and Bazinet, 2008). Cognitive and some behavioral consequences of this imbalance have been described in animals and correlated with effects on cell membranes in the central nervous system (Mahieu, Denis, et al., 2008). In human studies, alterations in omega-3 fatty acid levels have been associated with cardiovascular disease; stroke; cancer; cognition problems; and a number of behavioral problems, including attention deficit disorders, depression, autism, and suicide.
A number of randomized trials of omega-3 supplementation for mothers during gestation or for infants indicate benefits for cognitive and motor skills, including language development. These improvements could serve as protective factors for MEB disorders. Trials of the effects of omega-3 supplementation on aggression have also been conducted. Studies involving children have had mixed results, with three studies demonstrating a reduction in some symptoms of ADHD and related problem behaviors (Richardson and Montgomery, 2005; Richardson and Puri, 2002; Sinn and Bryan; 2007); one showing a reduction in hostility and aggression, primarily among girls (Itomura, Hamazaki, et al., 2005); two showing no effect on aggressive or disruptive behavior (Hirayama, Hamazaki, and Terasawa, 2004; Voigt, Llorente, et al., 2001); and one finding only limited effectiveness (Stevens, Zhang, et al., 2003). While not yet conclusive, however, the available evidence warrants well-designed experimental trials of the impact of omega-3 in preventing depression and behavioral disorders involving aggression.
The majority of randomized controlled trials of omega-3 supplementation have focused on its use to treat adults with mental disorders. Although two recent meta-analyses report evidence for the potential value of omega-3 supplementation, particularly for depression (Freeman, Hibbeln, et al., 2006; Lin and Su, 2007), another suggests that the effects are negligible (Appleton, Hayward, et al., 2006). All concur, however, regarding the troublesome variability of results; the heterogeneity and poor quality of many studies; and the need for large-scale, well-designed and -executed studies to permit conclusive statements.
Other associations between dietary content and MEB disorders are focused on the potential effects of allergenic foods and large boluses of sugar on the occurrence of ADHD (Wolraich, 1998). More study in this area is warranted.
Exposure to neurotoxins, such as lead and mercury, is a significant risk during gestation (see Chapter 5). Postnatal exposures are also of concern. Blood levels of neurotoxins in childhood are correlated with cognitive deficits and MEB disorders, including ADHD and conduct disorder (Braun, Kahn, et al., 2006; Braun, Froehlich, et al., 2008). Evidence has accumulated that blood lead levels once thought to be safe (>10 mg/ml) can be detrimental to infants (Canfield, Henderson, et al., 2003). Protection against exposure to lead, as well as other potential neurotoxins whose effects are not as well documented, is deserving of greater national attention, and demands the concerted efforts of medical caregivers, environment health specialists, community organizations, and lawmakers, as well as regulatory officials at all levels of government.
Physical Fitness and Exercise
Physical fitness and exercise are widely recognized as important modulators of stress, and there is some evidence of their effectiveness for the treatment of depression (Craft, Freund, et al., 2008). A meta-analysis of exercise interventions targeting depression and anxiety, primarily in college students, showed significant positive effects related to depression and positive but not significant effects related to anxiety (Larun, Nordheim, et al., 2006). However, the 16 available trials were of low methodological quality. A clear relationship between physical fitness and exercise and the prevention of MEB disorders in children is even less well documented. Given the clear relationship between exercise and stress, however, both general and medical education for children and their families should include discussion of appropriate exercise and advocacy for overall family fitness.
Extended television viewing has been linked to the occurrence of ADHD (Christakos, Zimmerman, et al., 2004) and limiting television time for children as a preventive measure has received increasing attention. The American Academy of Pediatrics recommends no television viewing for children under two years of age and no more than two hours a day thereafter. Exposure of children to violence through television and other media has been linked to conduct problems in children and adolescents (Bushman and Huesmann, 2006; Huesmann, Moise-Titus, et al., 2003). Attempts to reduce exposure of children to violence have had very little effect on the content of entertainment programming, and management of this risk falls largely to in-home restriction.
Exposure to adequate sunlight and light in general may affect mental health. Vitamin D deficiency can occur because children today are outside for shorter periods of time and are often protected by sunscreen. Vitamin D may have effects not only on bone mineralization, but also on immunity to infectious agents. Vitamin D plays an important role as well in brain development and function. Subtle effects of vitamin D deficiency on behavior have been suggested, but a causal relationship has not been firmly established (McCann and Ames, 2008). Whether prevention of vitamin D deficiency truly contributes to mental health in childhood deserves further study. Furthermore, limited exposure to light is related, in some individuals, to the occurrence of seasonal affective disorder. More brightly lit classrooms are associated with fewer classroom problems for children with ADHD (Kemper and Shannon, 2007).
CONCLUSIONS AND RECOMMENDATIONS: CHAPTERS 6 AND 7
This and the preceding chapter have documented substantial progress since the 1994 IOM report in approaches to prevention in multiple developmental stages. The strength of evidence related to prevention of symptoms and incidence of externalizing disorders and problem behaviors has significantly increased, particularly through school-based interventions. There is emerging evidence that preventive interventions not only can reduce symptomatology, but also can reduce the number of new cases of depression. And there is promising evidence of the potential to intervene in the lives of young people in the early stage of schizophrenia, prior to full-blown disorder.
Many programs that have been tested in multiple randomized controlled trials demonstrate efficacy, and an increasing number have demonstrated effectiveness in real-world environments. Increasing numbers of programs are culturally adapted and, while still relatively limited, some have been tested with multiple racial, ethnic, or cultural groups. It is no longer accurate to argue that emotional and behavioral problems cannot be prevented or that there is no evidence for the prevention of MEB disorders experienced during childhood, adolescence, and early adulthood.
Conclusion: Substantial progress has been realized since 1994 in demonstrating that evidence-based interventions that target risk and protective factors at various stages of development can prevent many problem behaviors and cases of MEB disorders.
Interventions variously target strengthening families by modifying discipline practices or parenting style; strengthening individuals by increasing resilience and modifying cognitive processes and behaviors of young people themselves; or strengthening institutions, such as schools, that work with young people by modifying their structure or management processes. Parenting and family-based interventions have demonstrated positive effects on reducing risk for specific externalizing disorders, for multiple problem outcomes in adolescence, for reducing prevalence of diagnosed MEB disorders, and for reducing parenting and family risk factors.
Conclusion: Interventions that strengthen families, individuals, schools, and other community organizations and structures have been shown to reduce MEB disorders and related problems. Family and early childhood interventions appear to be associated with the strongest evidence at this time.
Interventions based in schools have demonstrated positive effects on violence, aggressive behavior, and substance use and abuse. Emerging evidence has indicated the potential for a positive impact of some of these interventions on academic outcomes. Communities have a role in supporting preventive interventions and in developing responses that address community needs and build on community needs.
Conclusion: Community-based organizations, particularly schools and health care providers, can help prevent the development of MEB disorders and related problems.
Although an increasing number of interventions have shown positive results related to reductions in the incidence or prevalence of MEB disorders, most measure highly relevant risk and protective factors but do not measure disorders per se.
Conclusion: Preventive interventions can affect risk and protective factors strongly associated with MEB disorders. Future research must determine the full impact of these interventions on MEB disorders.
Preventive interventions have increasingly demonstrated positive effects on multiple outcomes, but the range of outcomes assessed is also limited. The same type of intervention may demonstrate positive effects on different outcomes, given the limited nature of the outcomes assessed. Similarly, although academic outcomes are likely to be important to schools considering adoption of preventive interventions, because there is some indication of positive effects on academic achievement, this has been assessed in only a few studies. Inclusion of a broader range of outcomes could help in the identification of potential iatrogenic effects that can meaningfully inform the development of future interventions.
Recommendation 7-1: Prevention researchers should broaden the range of outcomes included in evaluations of prevention programs and policies to include relevant MEB disorders and related problems, as well as common positive outcomes, such as accomplishment of age-appropriate developmental tasks (e.g., school, social, and work outcomes). They should also adequately explore and report on potential iatrogenic effects.
Although there are now multiple, well-tested interventions, the effect sizes for most interventions are small to modest. Similarly, though several studies have now demonstrated results with strong empirical designs and statistical techniques, meta-analyses consistently highlight the methodological weaknesses of many studies. As discussed in Chapter 10, this is not because of a lack of appropriate methodological techniques. There is a convergence among both meta-analyses and individual studies suggesting that interventions are more effective for participants with elevated risk, including for participants in many universal interventions. However, most interventions have been tested with a single cultural group, and few have been tested in community-wide interventions that reach large numbers of at-risk youth. Continued rigorous research is needed to improve the reach of current interventions and to expand interventions that are culturally relevant and responsive to community priorities (see Chapter 11).
Conclusion: Although evidence-based interventions are now available for broad implementation in some communities, there is a need to increase the effectiveness of prevention programs and to develop interventions that reach a larger portion of at-risk populations.
Recommendation 7-2: Research funders should strongly support research to improve the effectiveness of current interventions and the creation of new, more effective interventions with the goal of wide-scale implementation of these interventions.
Mass media and the Internet present a potential opportunity to reach large numbers of young people with readily disseminable interventions. Although the currently available evidence does not support particular interventions, this is an area that warrants additional research. Mass media also offers the potential to address concerns related to stigma that serve as a barrier to prevention.
Recommendation 7-3: Research funders should support research on the effectiveness of mass media and Internet interventions, including approaches to reduce stigma.
Although the research base of preventive interventions has expanded significantly, there are several groups or settings that have not been represented in this expansion. With the exception of college populations, very little research has been done related to young adulthood. Adolescence is also less well represented than earlier developmental periods. In addition, there has been limited research following young people across developmental stages. Although there is converging evidence that approaches that combine multiple interventions, such as family and school interventions, have greater effects, this is a relatively new area of inquiry.
Recommendation 7-4: Research funders should address significant research gaps, such as preventive interventions with adolescents and young adults, in certain high-risk groups (e.g., children with chronic diseases, children in foster care) and in primary care settings; interventions to address poverty; approaches that combine interventions at multiple developmental phases; and approaches that integrate individual, family, school, and community-level interventions.
In addition, as discussed in the chapters that follow, achieving the widespread benefits of evidence-based preventive interventions will also require further research on how to train those who implement interventions, how to influence organizations to adopt evidence-based interventions and to implement them with fidelity, and establishing an infrastructure with the capacity to implement and evaluate proven approaches. These problems might seem to be political and beyond the purview of public health and the behavioral sciences. However, policy decisions and the public support needed to influence those decisions are matters of human behavior. Just as a behavior like cigarette smoking is seen as something to change because it is a risk factor for cancer and heart disease, the lack of public understanding and support for prevention can be seen as a risk factor for societal failure to prevent problem development in childhood and adolescence. Research on how to generate public support for the implementation of evidence-based practices is a next logical step in the centuries-long struggle of the public health community to improve human well-being.
National Academies Press (US), Washington (DC)
National Research Council (US) and Institute of Medicine (US) Committee on the Prevention of Mental Disorders and Substance Abuse Among Children, Youth, and Young Adults: Research Advances and Promising Interventions; O'Connell ME, Boat T, Warner KE, editors. Preventing Mental, Emotional, and Behavioral Disorders Among Young People: Progress and Possibilities. Washington (DC): National Academies Press (US); 2009. 7, Prevention of Specific Disorders and Promotion of Mental Health.