Preventing Drug Abuse: What Do We Know? (1993)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Concepts of Prevention To prevent drug abuse, the central question is: What individual and group factors need to be considered in designing interventions to be effec- tive? To answer that question, a series of related questions have been investigated: What elements affect the probability of onset, progression, severity, and cessation of drug use, abuse, and dependence? By what mechanisms do these factors work, in what combinations, and with what degrees of strength or determinacy? What interventions can be used to subject these probabilistic factors to preventive change? INTRODUCTION The research in this field has had to cope with great complexity, involv- ing multiple causal and conditioning pathways and factors that are influen- tial in some populations or environments but that appear far less salient in others. In trying to untangle this complexity, research has followed a num- ber of paths, some of which were ultimately abandoned as unfruitful. Over time, the field has increasingly become oriented to a few systematic ap- proaches that have survived tests of theoretical coherence and empirical plausibility. Although these approaches are not antagonistic or contradic- tory, they differ dramatically in emphasis. A more encompassing synthesis or integration of approaches is not realistically in view. Nevertheless, an overarching, three-part conceptual framework is helpful in understanding the current approaches, and it provides a good basis for considering their differences and commonalities. We refer to three general concepts as pre- disposing, enabling, and reinforcing elements. 45

46 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? Predisposing elements, the first part of the framework, are comprised of internalized individual characteristics (also called diatheses) and environ- mental exposures (conditions). Predisposing elements are in effect prior to the first encounter or opportunity to try illicit drugs. Predispositional logic holds that some subsets of individuals, by virtue of factors that they have acquired or been exposed to, are more vulnerable or more resistant to drug use, abuse, or dependence than individuals without such factors, or with less of them, all other things being equal. Potential predisposing elements may be genetically transmitted vulnerability in the form of certain temperamental or physiological characteristics; developmental deficits, such as failures in early socialization or a lack of self-esteem, which imply that interaction within the family is an important locus of concern; knowledge and beliefs concerning the hazards of drugs; the individual's own perceptions of a drug's ability to harm; moral beliefs and attitudes about drug consumption; or the individual's so- cial circumstances and prospects irrespective of family interaction. Second are enabling elements. These are decision-making and eco- nomic or other circumstances relating directly to individual behavior in the situation of opportunity to consume a drug. The major enablers are of two kinds: (1) the availability and accessibility of drugs and prevention or treatment resources in the community and (2) the individual's skills to de- fine and respond autonomously and effectively to problem situations such as the ones that drug availability presents. Knowledge or belief structures, self-perceptions, and skills may be transmitted interpersonally or through mass media. The distribution of both predispos- ing and enabling elements tends to be associated with socioeconomic class and ethnicity. The relationship of predisposing and enabling elements may be critical to understanding why the rates of onset of drug use may be similar in different groups but then diverge into sharply different rates of drug abuse and dependence. Third are reinforcing elements, which are the environmental (especially social and economic) contingencies that attach to drug-related behavior. Reinforcement may result from social recognition by a significant other or members of an important reference group, in the form of giving or with- holding approval (praise, prestige, esteem), disapproval (complaint, ridi- cule, or dislike), or intimacy; or earning money or acquiring property as a result of drug-related income. Major significant others and groups include parents (whose influence declines over time), peers (whose influence in- creases from childhood to adolescence); teachers; and job supervisors and coworkers (including military peers and superiors). Parents may retain greater influence than peers in some families. Like enabling elements, social reinforcers are distributed differently in different socioeconomic classes, ethnic groups, and residential zones (Green and Kreuter, 1991; Gottlieb and Green, 1987; Heckler, 1985; Jacob, 1987; Thomas, 1990~.

CONCEPTS OF PREVENTION 47 There are four major conceptual approaches to prevention: risk-factor, developmental, social influence, and community-specific. We briefly de- fine each of these approaches in the next few pages. We then proceed in the balance of the chapter to present a more thorough review of the respective literatures of the first three approaches. Since the community-specific ap- proach is still largely outside the drug prevention research literature, we defer discussion of this approach to the appendix. The Risk Factor Approach Three major schools of thinking and associated research about preven- tion emphasize one or more of these concepts of predisposing, enabling, and reinforcing elements. The first school speaks principally in terms of risk factors, a concept that is used extensively in the epidemiology of car- diovascular, cancer, and other chronic diseases (Bry et al., 1982; Newcomb et al., 1987~. This is the most comprehensive approach in terms of the range and number of factors considered; it is also the least theoretically structured and the least empirically focused. A risk factor is any observable (measurable) characteristic of the indi- vidual (including duration of exposure to specified environmental condi- tions) that has been shown to correlate significantly (in population or case- control studies) with a criterion behavior or outcome in this case, with the onset of illicit drug use, some threshold level of consumption, or the clini- cal occurrence of drug abuse or dependence. This specification makes the risk factor model more empirical than theoretical. The risk factor must precede or at least occur simultaneously with the drug behavior; that is, a risk factor must be a potential cause or precursor, not a direct or indirect effect or symptom, of the criterion behavior. Reciprocal causation between risk factors and criterion behaviors is not precluded; in fact, as discussed below, a mutually reinforcing feedback among problem behaviors is the common pattern. For example, the desire for peer approval may predispose a teenager to try marijuana with her friends, the reduced inhibition and the relaxation felt during use reinforces the behavior and predisposes her to another opportunity to use. Most of the risk factors studied, in terms of the conceptual framework just reviewed, count as predisposing elements. Interventions to prevent drug use following the risk factor approach tend to emphasize educational approaches to modify self-esteem, specific beliefs and attitudes concerning drug use, and related predisposing factors (Bry et al., 1982; Newcomb et al., 1987~. Risk factors are statistical or probabilistic: if an individual "has" the factor, his or her odds (that is, statistical risk) of having the outcome are higher than if the individual does not have the factor, all other things being equal. For example, if John thinks marijuana is harmless, then the odds that he will try it are higher than

48 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? if he thinks marijuana can hurt him. Risk factors are usually additive: that is, risks add up; the more of them that apply, the more probable it is that the criterion outcome "at risk" will be observed. Some risk factors are easier to change than others, and some risk factors may weigh more heavily (higher zero-order or partial correlations with the criterion) than others. Those that meet both of these criteria become more strategic targets for intervention. Some risks may interact or have "synergistic" effects, in which one factor statistically multiplies rather than simply adds to the effect of other factors; in other words, a may be a nonsignificant risk factor, b may be a nonsignificant risk factor, but a and b together may be a formidably sig- nificant risk factor. Thus, although a may be a significant risk factor, in the absence of b, its effect on drug use is minimal. An open question is whether risk factors are generic (i.e., to many drugs) or specific (to each drug family). The Developmental Approach The second school of thought about prevention is based on developmen- tal theory. This approach particularly emphasizes the character and dynam- ics of interaction over time within the family during early childhood and within environments such as the school, especially grades 1-6. It shares with some risk factor theories a concern with early developmental deficits or predisposing factors. It differs, however, from risk factor theories in its heavy concentration on characteristics of the family and school environ- ment that directly reinforce undesirable patterns of affect, belief, or (most important) behavior. Conversely, it also concentrates on environmental re- inforcement of the development of positive motivation, educational poten- tial, and prosocial behavior. The developmental approach articulates a more elaborately linked and structured set of factors than risk factor approach. It has a more diffuse target, however; instead of trying to identify and focus on individuals who are "high risk" as the object of preemptive intervention, the developmental approach tends to bracket more inclusive populations and more dimensions of lifestyle or behavior (more than drug use, that is) as the loci of long-term environmental and institutional change. Social Influence Approaches The third major school of thought about prevention really a family of related approaches involves research on social influence. It is the most tightly focused theoretically, and it is population-based. Increasing atten- tion is being given in social influence research to variations among demo- graphic and other groups. It recognizes the important role of peers in the initiation and progression of drug use.

CONCEPTS OF PREVENTION 49 The social influence model is based on four core components: (1) providing information on the negative social and short-term physiological consequences of smoking; (2) providing information on the social influ- ences to smoke- namely, peer, parents, and mass media; (3) correcting in- flated perceptions of smoking prevalence; and (4) training, modeling, re- hearsal, and reinforcement of methods to resist the social influences to smoke. Interventions largely concentrate on 6th through 10th grade students and are best known for aiming to prevent the onset of use by modifying enabling factors; in particular, increasing the knowledge of harmful effects and teaching specific resistance skills for resisting persuasive messages from peers and mass media. Cigarette smoking is the most thoroughly docu- mented health-related behavior in social influence theory, and most inter- ventions to increase resistance skills were originally developed and tested in the context of preventing the onset of smoking (Evans and Raines, 1982~. We have documented the relevance of smoking prevention to illicit drug use prevention in Chapter 1, in the discussion of gateway drugs and the se- quence of progression of drug involvement. An important variation on social influence approaches is the cognitive- behavioral model, which is based on the assumption that substance use results from the combined influences of social and psychological factors. Based on work by Schinke and colleagues on pregnancy prevention (Schinke and Gilchrest, 1977; Schinke, 1982), this approach has been adapted to smoking and other substances. The theoretical basis of the model is derived from both developmental and social learning theory. Alcohol and drug use is viewed as instrumental in meeting the developmental needs of youth (e.g., transition marker, reducing stress, peer group acceptance, establishing independence). The strategy for drug prevention emphasizes the develop- ment of enabling skills, the acquisition of decision-making and problem- solving skills to equip youth to make informed decisions about alcohol and drug use. The focus is on the development of cognitive, behavioral, and interpersonal skills. The approach is based on five core elements, which: deal with a wide range of problem situations through the use of a systematic problem-solving strategy, provide accurate information, teach coping strategies to relive stress and anxiety, develop assertiveness skills, and · develop self-instructional techniques for behavioral self-control. A final important stream of work is the life skills approach, which emphasizes the development of general life and coping skills, in addition to skills and knowledge related more directly to resisting peer influences to use substances (Botvin et al., 1980; Botvin and Eng, 19801. The program

so PREVENTING DRUG ABUSE: WHAT DO WE KNOW? focuses on teaching cognitive-behavioral skills that remedy psychological or behavioral deficits. The Life Skills Model program consists of three major components. A substance-specific component incorporates most of the information from the social influences approach. A second component addresses developing personal skills such as coping strategies, critical thinking, and decision-making skills and teaches the basic principles of behavior change. A third component develops social skills designed to improve interpersonal functioning. The Community-Specific Approach A fourth perspective attempts to encompass all of the prior three. We refer to this as the community-specific prevention approach. Community- specific prevention is receiving major attention in various fields of public health, particularly in preventing cigarette smoking and in controlling risk factors for cardiovascular disease, cancer, AIDS, teenage pregnancy, and other major health or related social problems. The conceptual foundations of drug abuse prevention historically have been imported from behavioral and social science research on cigarette smoking reduction and public health promotion generally. Large differences in the scale and nature of severe drug problems experienced in different communi- ties makes the community-specific approach seem especially applicable to drug abuse prevention, insofar as it is oriented to investigating population differences and community variations, and to mobilizing resources accord- ingly. The community-specific approach is, nevertheless, a barely culti- vated areas of drug abuse prevention research, within which the published work is not commensurate in scope with the risk-factor, developmental, and social-influence literatures. Therefore, we take this subject up in the appen- dix, which looks more generally to community-based health education to illuminate this important dimension. STUDIES OF RISK AND VULNERABILITY Much research attention has been focused on risk factors variables that exist before or during the typical age of onset of drug use (the second decade of life) and predict an elevated probability of developing abuse or dependence and on their mirror image, protective factors- those that seem to confer a degree of immunity against drug involvement. By and large, risk and protective factors are opposed ends of a set of continua, for ex- ample, impulsivity versus planning, strong versus weak family bonding (Jessor et al., 1992~. Risk and protective factors thus refer to relative degrees of vulnerability on a set of continua. Risk and protective factors may be characteristics of the individual or

CONCEPTS OF PREVENTION 51 of the environment. Individuals vary greatly in physical and behavioral responses to nearly all health-related exposures or opportunities; they also vary in the environments to which they are exposed. The study of such variations and how they affect the probability of health problems has been immensely important in the history of medicine and public health, so it is no surprise that this approach has been adopted in the drug area (Rennert et al., 1986). A salient finding about patterns of drug consumption, discussed in the previous chapter, is the fact that a much larger number of individuals use drugs some very briefly, some intermittently over a longer span of years, some regularly but at a modest level that does not increase over time than the number who progress to the clinical status of abuse or dependence. The infrequent and/or low-dose use of drugs is not a matter of indifference, because such use is illegal and can have serious consequences. Any level of use generates a degree of risk of progression to abuse or dependence as a result of internal reinforcement, and use by some is likely to model or reinforce abuse and dependence by others. But by definition, the conse- quences of use are much less hazardous for the individual, on average, than the consequences of abuse and dependence. Although users outnumber drug dependent and abusing individuals, the smaller number of the latter incur the majority of the social costs of drug problems. It is therefore important to give particular attention to the degree to which particular causes increase the probability of abuse or dependence over and above the inci- dence of drug use per se. There are indications that the processes leading to use may be differen- tiated from those leading to abuse and dependence. In particular, unusually early onset of drug use (that is, well before the average age of onset in the population) is a strong correlate of later abuse or dependence, although this is not an infallible marker (Kandel et al., 1986~. The early onset of ciga- rette smoking is of special interest, and early alcohol and marijuana onset are also of concern, because these tend to be gateways to other drugs. Most studies of drug-related risk factors have been exploratory rather than substantive, that is, they have employed small samples, followed up for abbreviated periods, and have inadequate disaggregation and control for gender, race/ethnicity, and socioeconomic status. There are, however, a few studies large enough to establish with a certain degree of confidence the relative importance of key factors, including longitudinal studies conducted by a number of research teams, including: Judith Brook and colleagues (Brook et al., 1990~; Brunswick (1988~; Elliott and colleagues (Elliott et al., 1989~; Jessor and colleagues (Jessor and Jessor, 19771; Kandel and col- leagues (Kandel et al., 1986~; Kaplan and colleagues (Kaplan, 1985; Kaplan et al., 1988~; Kellam and colleagues (Kellam et al., 1983~; Newcomb and gentler (1988, 19891; Pandina and colleagues (Pandina et al., 1984; Labouvie

52 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? et al., in press); Pentz and colleagues (Pentz et al., 1986~; and others. The following discussions draw heavily on these studies. We first review some of the literature that has focused on single risk factors; the yield of this literature is rather low, so we have been highly selective in attempting to represent it, pointing out major conclusions of studies on the role of genetic and congenital factors, personality characteristics, and socioeconomic neigh- borhood characteristics. We then review the results of studies on multiple risk factors that focus attention on the issue of how these risk factors inter- relate. Genetic and Congenital Predispositions Since psychoactive drugs are chemical agents that work inside the body, it is natural to think that biological factors, including biologically heritable factors, play some part in promoting or inhibiting the onset of drug use, abuse, and dependence. The evidence for this hypothesis, however, was indirect and slender at the time of the committee's review for all drugs except alcohol. For alcohol, the heritability of some tendency heavily modulated by environmental and developmental features appears reason- ably well established. The evidence for biological risk factors is of two kinds. First, different strains of animal species bred for laboratory studies vary in their predilec- tion or resistance to consuming alcohol and other drugs, and these prefer- ences can be altered over generations through selective breeding. (These preferences can also be altered through training; trained behaviors are not, of course, genetically transmissible, although quickness in learning is.) Second, there is evidence from behavioral-genetic and related studies with human populations. Most of this work pertains to alcoholism, al- though there is evidence from other pharmacogenetic and genetic epidemio- logical research indicating predispositions to other types of drug abuse and dependence (Institute of Medicine, 1989; Pickens and Svikis, 1988; Pickens et al., 1991~. Family and twin studies suggest that there is a genetic predis- position toward one of two typical patterns of alcoholism. Children with a biological parent who has developed clinical alcoholism, even if this parent had no role whatsoever in their childrearing (e.g., children adopted at birth), are at four- to tenfold greater risk of this outcome compared with matched children whose biological parents are without a clinical history of alcohol- ism (Cloninger et al., 1981; Goodwin, 1983~. One index of risk that has not been well studied is the magnitude of dissonance among biological, cognitive, and behavioral spheres of function- ing during the early second decade. It has been observed that girls who enter puberty early may not yet be equipped with a number of social and cognitive skills commensurate with biological maturation. They may there

CONCEPTS OF PREVENTION 53 fore be at increased risk for a number of adverse outcomes, perhaps for as long as a decade afterward, including drug and alcohol abuse, antisocial disorder, school dropout and unplanned pregnancy (Magnussen et al., 1986~. The age at menarche, as one biological marker of a host of anatomical, hormonal, and social changes, has been dropping steadily over the past 40 years, and social institutions have adjusted unevenly to these maturational developments. Overall, the place of biological heritage and biological mediation in explaining the onset of drug use, abuse, and dependence remains uncertain. Further human population research that attends as carefully to environmen- tal conditioning as to physiological measures is needed to evaluate the rela- tive role of neurochemical and other biological predisposing factors. A1- though it is premature to recommend trials of strategies for informing people of their possible risk based on family history of drug use, further analysis of the potential risks and benefits of such advice (e.g., the risks of labeling people and reduced self-esteem versus the benefits of reduced use of drugs) is justified in anticipation of improved biological markers of risk (Bamberg et al., 1990; Becker and Janz, 1987; Bensley, 1981; Childs, 1974; Hunt et al., 1986; Khowry et al., 1985; Zylke, 19873. Personality Characteristics Only a small number of the many personality characteristics that have been investigated in connection with drug use have shown significant re- sults as risk factors (Lang, 1983~. Among these few characteristics, the most positive evidence has accumulated in support of a psychological con- struct called sensation seeking. In contrast, such factors as depression, suicidal thoughts, and low self-esteem, all of which seem very plausible and often serve as commonsense assumptions underlying the design of drug abuse prevention efforts, do not stand up well under empirical investigation. Zuckerman (1979) described sensation seeking as a fundamental aspect of personality based in the neurochemistry of monamine oxidase. His four measures of sensation seeking seeking new experiences, seeking thrills or adventure, susceptibility to boredom, and disinhibition have been shown to correlate with a number of illicit activities, including alcohol and drug use, in adolescent and young adult populations (Bates et al., 1985; Huba et al., 1981~. In studies using the Rutgers longitudinal sample, sensation seek- ing and negative affectivity proved to have much larger effects on drug use, both independently and interactively, than positive affectivity. Newcomb and McGee (1989), using multivariate methods to probe results with the UCLA sample, found that sensation seeking had unexpectedly complex ef- fects, differing for males and females, with the most pronounced relation to high levels of alcohol use.

54 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? Many clinicians believe that specific emotional disorders, particularly depression and related distress, trigger or severely aggravate drug use, abuse, or dependence. The evidence in this direction is inconsistent. Kaplan (1985), Huba et al., (1986), Aneshensel and Huba (1983), and Labouvie (1986) all found that drug use is often preceded by emotional distress or depression. But the relieving effects of drug use on these states is short- lived. Newcomb and gentler (1988) found that alcohol use over time in a general population sample of adolescents was correlated with a reduction in depression, but no such correlation emerged linking other drug use to depression or other emotional distress. Elliott and Huizinga (1984) found that emotional problems and social isolation (feelings of loneliness) were moderately correlated with the level of use of alcohol, marijuana, and other illicit drugs in a general youth population sample. Dembo and col- leagues (1991) found a similar result among detainees in a juvenile deten- tion center. The most extreme level of depression is suicidal thinking and attempts. Suicide is the second leading cause of death among adolescents. However, drug use seems to be more a risk factor for suicide attempts than the other way around. Newcomb and gentler (1988) reported that adolescent use of "hard" drugs (beyond alcohol and marijuana) was associated with subse- quently increased suicidal thinking in young adulthood. The belief is widely held and intuitively appealing that a strong sense of self-esteem is a protective factor and lack of it a risk factor for adoles- cent drug use. There is no doubt that most cases of adolescent drug abuse or dependence that come to clinical attention are individuals who are short on self-esteem. The specific notion is that individuals with low self-esteem seek drugs in order to raise it (Kaplan, 19861. Numerous preventive inter- ventions have applied this theory by seeking to build up their participants' self-esteem, teaching them how to raise it, or expanding the opportunities for enhancing self-esteem in ways other than by taking drugs. Despite its attractions, the evidence for the self-esteem theory is mostly not supportive. In large studies such as White et al. (1986) and Kaplan et al. (1984), very weak correlations were observed between self-esteem and drug use, and these variables paled into insignificance under further statisti- cal manipulation. Even if self-esteem did seem to be an important risk factor for drug taking, the idea that it might be altered by any of the pro- gram measures ordinarily undertaken is problematic, denying or ignoring as it does commonly assumed determinants of self-esteem such as physical attractiveness (Simcha-Fagen et al., 19861. In summary, the search for specific personality risk factors for illicit drug taking has been mostly disappointing. Studies on sensation seeking, an active trait, have proven more promising than those focusing on more inward-turning characteristics such as depression and self-devaluation.

CONCEPTS OF PREVENTION 55 Socioeconomic Factors at the Neighborhood Level The epidemiologic evidence indicates that onset of illicit drug use oc- curs mainly through peer group contact and that rates of onset (as distinct from continued use) are at rather similar levels within economic and ethnic groups. We suspect that the illicit drug use and trafficking that occur in economically disadvantaged communities, which are disproportionately black, Puerto Rican, and Mexican-American, occur for many of the same reasons as in other segments of the population, but that these reasons are more intense. In the most depressed portions of these communities, there is an additional dimension associated with greater numbers of drug abusing and dependent individuals and high levels of violence: namely, for many poor, young minority men and women, illicit drug markets are key sources of employment and are perceived as a route to economic mobility. In order to be successful in selling drugs, it is necessary for these young people to encourage drug use aggressively among the most vulnerable members of the community and to be prepared to enforce and protect their transactions in an increasingly gun-ridden and anarchic environment. As Brunswick (1988) notes in her longitudinal study of several hundred youths from central Harlem: "An often overlooked cornerstone of hard drug use among young black males is that it is not only and perhaps not primarily a consumption and/or recreational behavior. It also serves eco- nomic functions of occupation and career for this group" (see also Johnson et al., 1985; Preble and Casey, 1966; Williams and Kornblum, 1986~. In a population subgroup in which employment opportunities are severely con- strained, and at a life stage at which economic independence is expected and required, the drug economy is one of the relatively few high-wage options that seem wide open (Reuter et al., 1990~. It is not known with certainty what distinguishes those who sell drugs in economically disadvantaged communities from the majority of their peers in these areas who, with similarly limited opportunities, shun drug involve- ment, or from those in the middle who use but do not sell drugs. The perception and fact of being socially distant from mainstream opportunities, at the same time needing money in order to survive, are important. But, in every ethnic group in subcommunities dominated by drug use and sales, families are the most important social unit particularly so given the pau- city of institutional infrastructure in most economically impoverished areas. Although drug users in poor minority subcommunities are predominantly from single-parent, female-headed households, the same is true of those adolescents who do not use drugs (Fitzpatrick, 1990~. Whether or not there is an intact nuclear family, the most important family inhibitions against drug use (either through predisposition or through reinforcement) may be the active involvement of multiple adults in the immediate or extended

56 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? family or even among nonfamily members in the lives of young people who are environmentally at risk (see, for example, Kellam et al., 1983; Zimmerman and Maton, 1992~. Another unknown is how differential aspects of African-American, Puerto Rican, Mexican-American, and other cultures serve as barriers to or pro- moters of drug use, as mediating factors in the initiation and conduct of drug use, and potential influences on the routes by which users can become drug free. Blount and Dembo (1984) assessed levels of alcohol and mari- juana use among approximately 1,000 Cuban and Puerto Rican youths in inner-city junior high schools, using questionnaires based on extensive eth- nographic work in these areas, which incorporated local cultural patterns by paying particular attention to perceptions of the "toughness" and level of drug involvement in the respondents' immediate neighborhoods. The re- sults provide a textured picture of the differing contingencies that inner-city youths confront. Participation in street culture during leisure hours was highly correlated with marijuana use, especially in the toughest neighborhoods (Blount and Dembo, 19849. The correlation between respondent and peer group mari- juana use was appreciably stronger in the tougher, more drug-involved neigh- borhoods. In other words, in tough neighborhoods, you are either with the pot smokers or not-it is rare to have close friends among abstainers and smokers at the same time. In contrast, alcohol use was not correlated with street culture it cut across neighborhood differences, and the positive as- sociation between respondent and peer group alcohol use was about the same everywhere. The attitudes, peer group relations, and adult role models of nonusers, alcohol-only, and alcohol-and-marijuana users were consis- tently different. Beyond these differences, the need to choose starkly be- tween friendships with tough kids who are usually marijuana users and friendships with nonusers was a fact of life in the toughest neighborhoods, one that youths in less combative zones even in the inner city-could more readily finesse, and one that was not present with respect to alcohol, regardless of neighborhood. Relationships Among Risk Factors Young people who engage in one form of health-compromising behav- ior are often engaged in other problem behaviors (Jessor and Jessor, 1977~. The co-occurrence of alcohol and other drug abuse with delinquency and criminal behavior is well established (Elliott et al., 1985; Hawkins et al. 1987; White, 1990~. From the perspective of temporal order (and thus relevant to predispositions), the first involvement in delinquent activity usu- ally predates illicit drug use. But findings from a number of longitudinal studies (e.g., Jones, 1968, 1971; McCord and McCord, 1962; Monnelly et

CONCEPTS OF PREVENTION 57 al., 1983; Ricks and Berry, 1970; Robins, 1966, 1978) suggest that drug use and antisocial behavior in adolescents have similar precursors: aggressive behavior, school conduct problems, poor grades, and, less certainly, shy- ness, anxiety, depression, and problems in peer relationships. Early alcohol and drug use along with violent or predatory behavior and early and aggres- sive sexual behavior seem to be part of a general pattern of rebellion and nonconformity variously called a "deviance syndrome," "antisocial person- ality," "conduct disorder," or "adolescent adjustment disorder." In an analysis based on a national longitudinal study of 11-17-year-old youths in 1976, Elliott and Morse (1987) demonstrated the interrelationship of drug use, delinquency, sexual activity, and pregnancy. They found that 71 percent of the males and 52 percent of the females who were using multiple illicit drugs were sexually active, compared with 10 percent of the males and 3 percent of the females who were not using any drugs. Along similar lines were results of a study of nearly 1,000 adolescents in Los Angeles in grades 7-9 who were resurveyed in grades 10-12 (Newcomb et al., 1986~. About 51 percent of the high school age sample had used mari- juana at some time. But only 22 percent of those with none of the risk factors identified (low grade point average, lack of religious participation, poor relationship with parents, early alcohol use, low self-esteem, lack of conformity, sensation seeking, perception of ease of obtaining drugs, per- ception of neutral or favorable norms concerning drug use) had used mari- juana, compared with 94 percent of those with 7 or more risk factors. These results were consistent for all other drugs and for higher levels of consump- tion. About 8 percent of the sample were using marijuana on a daily basis. Of youths in the sample with zero risk factors, however, less than 1 percent were daily marijuana users; of those with 7 or more risk factors, 56 percent were daily marijuana users. No single predisposing factor dominates these analyses; rather, move- ment toward drug problems seems to proceed by the accumulation of small and mutually supporting effects over time throughout early childhood and into the adolescent window of onset. The movement is a general drift toward adolescent problem behavior of various kinds and away from prosocial pursuits. If this drift across a continuum into increasingly problematic areas is indeed the principal type of causal process predisposing toward drug use, and particularly toward the higher (and more diversified) levels of consumption that mark abuse and dependence, then a preventive approach that attends systematically to a broad range of variables across a span of childhood years would be highly attractive. It is similar in this regard to the gradual accumulation of risk for heart disease and cancer from the cumula- tive effects of relatively innocuous discrete acts and gradually changing behavior patterns. Risk factor research thus seems to lead fairly directly to a developmental turn.

58 PREVENTING DRUG ABUSE. WHAT DO WE KNOW? Research Needs The study of multiple risk factors and their interaction appears to present substantial advances over attention to single factors or limited clusters of factors. This is not to say that more tightly focused studies should not be undertaken, but that such studies are best viewed as leading toward results that can be incorporated into larger-scale multivariate studies. There are needs for refinement of risk-factor research in several directions, but one in particular deserves emphasis here: methodological investment in improving techniques of measurement, particularly of environmental factors. A major reason for improved measurement is to avoid statistical biases (descriptive and inferential) in multivariate analyses. For example, factors such as personality traits are generally measured by multi-item scales ad- ministered to the individual and scored to identify the extent of individual variation from population parameters. In contrast, factors such as neighbor- hood quality, which urban researchers find can vary literally by the block in many areas, are usually measured at the level of the census tract or larger geographic swaths, using such proxies as average housing cost or popula- tion density, aggregated into quartiles, or loose "urbanicity" measures based on proximity to traditional city cores. The measurement error (in terms of an accurate index of the individual's experience) that accrues from averag- ing across many blocks and then assigning individuals into such large, often ill-fitting categories ensures that, even if neighborhood quality or other collective characteristics were a powerful influence on the individual's be- havior, these effects would be virtually precluded from statistical detection. This measurement bias would lead to false negative or Type II errors, in contrast to the likelihood that weak but transitory effects may be detected by finely calibrated personality variables that are measured at the individual level, leading to false-positive or Type I errors. THE DEVELOPMENTAL APPROACH A Model of Progressive Problem Behavior A four-stage model of behavioral problems accumulating across time, which draws together a large literature (Kumpfer, 1989), has been described by Schaps and Battistich (1991~. This model suggests that socialization deficits in early childhood lead young people to affiliate with peers opposed to traditional institutions (such as school), a tendency that culminates in social alienation and trouble with the law (and other conventional institu- tions of society) in late adolescence and adulthood. This model parallels the logical progression of drug use to abuse to dependence, in that a rela- tively small proportion of youths who embark on the path of drug use continue on to dependence.

CONCEPTS OF PREVENTION 59 In the first stage of the model, poor parenting (or, more generally, childrearing) practices in the family or among major alternative caretakers, which are evident during the preschool years, lead to low emotional attach- ment to parents, resistance to parental authority, early behavioral and emo- tional problems, and generalized developmental immaturity (poor attention span, poor impulse control). Negative parenting practices include low lev- els of parental affection, lack of concern and insensitivity to the child's needs, lack of supervision, hostility, rejection, and very inconsistent or pu- nitive discipline. If parenting practices to which the child is subject do not improve, these patterns of poor family bonding become more violent and reciprocal as the child grows beyond preschool. Although family economic conditions do not directly determine parenting practices, high levels of stress and disorganization degrade parenting per- formance, and these levels of stress are more common when family eco- nomic resources are scarce and when the neighborhood environment is it- self impoverished and disorganized. The effects of discrimination based on race or ethnicity add to these stressors. In the second stage, poor socialization in the family leads to emotional and conduct problems in school grades 1-3. Peers and teachers respond antagonistically to poorly socialized behavior, and the child in turn is beset by social isolation or rejection, anxiety, insecurity, and continued conflicts with authority. The course of this second stage is obviously affected by the ability of the classroom teacher to adapt to poorly socialized children and educe not simply a modicum of compliance but rather positive bonding with the school, its staff, and other students. In the third stage, middle to late elementary grades 3-6, persistent prob- lems in social adaptation result in decreased learning and poor grades. De- ficient academic performance in turn creates isolation from and rejection by more academically competent peers; problems in adaptation to school trans form into active alienation from school. It is among these youths that the early onset of tobacco or alcohol use, and in some instances marijuana as well, will occur. In the fourth stage, junior high school continuing on into high school, students disaffected from schooling firmly withdraw their efforts from aca- demic or any other school-organized pursuits, become more overtly rebel- lious, and associate with each other in increasing opposition to academi- cally competent and socially conventional peers, who reciprocate the hostility. An increasingly exclusive association with alienated peers intensifies into a school-oppositional peer group culture (Willis, 1977), characterized by ex- pressive rejection of the conventional social norms and values, continuing academic failure, alcohol and drug use, delinquent activities, sexual behav- ior resulting in pregnancy, and a higher probability of early school exit.

60 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? Schools themselves unintentionally further perpetuate this "clustering" of alienated adolescents by their policies of remedial education placement and detention activities that group these students together (Oetting and Beauvias, 1987~. Interventions that assume school-based peer ties and adult-student soli- darity will not be appropriate to committed members of the school-opposi- tional culture. Interventions that work as a reinforcer or accelerator of antidrug trends in the school-solidary culture may have null or even re- bound effects in the school-oppositional one. In most schools, oppositional norms characterize a marginalized, limited proportion of students. In some, these hold the allegiance of a large fraction or even the majority, for ex- ample, in "special schools" for disciplinary problems, schools for emotion- ally disturbed youths, and schools that experience drop-out rates prior to high school graduation of 50 percent or higher (Lorion et al., 1989~. The school-oppositional culture is resilient, a bed of resistance or rebel- lion that responds strongly to attempts to affect it; it "pushes back" in ways that rebound into the larger society. Efforts to vilify characteristic practices or rituals of oppositional groups may have the perverse effect of strengthen- ing those practices or amplifying the groups' sense of distance and rejec- tion. In cultures formed out of economic and normative marginalization, particularly within communities that are precariously bound to begin with, all identity appears to be formed around antimainstream attitudes; those involved, however, do in fact claim to hold many mainstream values despite some forms of denial or nonparticipation a good example being the drug dealer who says "I'm a businessman." Entry into the later stages of systematic, deep-seated deviance implies that earlier stages have probably occurred. But some children who become academically troubled or transfer all of their loyalty into school-opposi- tional culture have not experienced all of the earlier stages. School opposi- tion may not reflect alienation from family, for example, if the school is not generally integrated into a subculture, which is evidently the case in certain Native American and Mexican-American communities in metropolitan and rural areas. Nor will all the individuals at any one stage progress to later ones. In major longitudinal studies, no more than 30-40 percent of the early elementary children who displayed behavior problems engaged in antisocial behavior, delinquency, or drug abuse in adolescence (Robins, 1978~. Parenting practices can improve or deteriorate over time, as family structures change through divorce or remarriage, parents mature, marital discord emerges, etc. An unusually positive school experience may counter a poor home environ- ment; strong academic aptitude may prevail despite conduct problems; or uncompensated learning disabilities may erode initially successful academic work and school attachment.

CONCEPTS OF PREVENTION Taking the Model Seriously: Reforming the School 61 Despite the many sources of variance described above, the model of problem-behavior development has strong theoretical appeal and a variety of empirical supports. It is gaining increasing attention due to concern over the steady diminution in social attentiveness to children and a generalized social deficit in parenting, of which the more extreme cases of child abuse are only a fraction. Much has been written about the relative demise of the two-parent nuclear and the extended family (Schroeder, 1989), the disap- pearance of personal neighborhoods and other forms of continuous local community (Green, 1990), and the increasing separation of children and youth from adult workplaces and occupational pursuits (McMillan and Chavis, 1986~. These trends have resulted in the separation of children from adults in a way that is historically unprecedented. They have also served to limit caring, stable relationships between children and adults and to deprive chil- dren of meaningful exposure to a range of adult models and situations. It is largely through close relationships with adults mostly in the roles of parents, relatives, and teachers that children learn how to function as adults and develop motivation to take on adult responsibilities. As they are stripped of opportunities for such relationships, it is media portrayals to which they must increasingly turn for information about "what to become." The open, pluralistic character of American society and the great freedom that this potentially provides to select personal behavior is, in a sense, lost on children whose ideas and aspirations are increasingly encapsulated within a peer social system whose culture is heavily oriented to electronic media figures- surreal, postural, and fantastic especially when they are antago- nistic to schools and other conventional institutions. The societal trends are pervasive, cutting across virtually all demo- graphic categories. The observation is becoming increasingly common that vast numbers of American children are now "at risk" (see Carnegie Council on Adolescent Development, 1989; National Commission on the Role of the School and the Community in Improving Adolescent Health, 1990~. There is undoubtedly variation in the degree to which children are lacking in sustained adult connections and guidance, are excluded from exposure to responsible adult roles, and are living in environments saturated with oppor- tunities for problem behaviors. It is probable that such conditions now prevail in extreme forms for many and in milder ones for most children, and that widespread experimentation with problem behaviors, including drug abuse in one form or another, may prove endemic, even though waves of such behavior will advance and recede. American schools have changed less in the past few generations than have the other major socializing institutions. Indeed, it appears highly problematic that schools have changed so little in the face of dramatic

62 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? changes elsewhere. Most schools, rather than trying to compensate for the growing deficiencies in students' lives, are Reemphasizing personal rela- tionships between children and teachers (Carnegie Council on Adolescent Development, 1989~. T Instead, their focus is on rigor and efficiency, in reaction to recent concerns about academic achievement, particularly in science and mathematics. Strengthening child-adult relationships is simply not viewed as a priority in how schools are organized and how teacher time is allocated. The typical classroom is structured, impersonal, and formal (Goodlad, 1984), and students are given little opportunity to take guided responsibility for their own learning or to learn service to others. Other school characteristics compound this problem. Curricula heavily focused on developing basic cognitive skills and acquiring facts provide students few opportunities to demonstrate mastery, to see connections with "real life," or to develop the higher-order cognitive skills and social compe- tencies they will need to experience satisfying interpersonal relationships, to resist dysfunctional social pressures, and to take on adult roles. Most schools rely on competitive evaluation systems and pervasive use of extrin- sic rewards, practices that adversely affect many children's sense of compe tence, self-esteem, intrinsic motivation for learning, and actual performance (Deci and Ryan, 1985~. As a result, students' motivation to learn seems to be declining; many students see classroom work as meaningless and not worth the effort to succeed (Ames, in press; Zimiles, 1986~. The impersonalcompetitive classroom atmosphere alienates many as they progress through school, leading to negative perceptions of self-worth, reduced academic efforts, more frequent misbe- havior. Schools may respond by imposing an even heavier "curriculum of control" (Knitzer et al., 1990), and the downward spiral ensues. In recognition of these and other problems, some recent thinking in education has begun to shift toward a greater concern with developmental relevance (Katz, 1989), promotion of intrinsic motivation (Deci and Ryan, 1985; Nicholls, 1989), the active role of the learner as a "maker of mean- ing" (Resnick, 1989', attention to social and moral development as a legiti- mate aspect of the curriculum (Ryan, 1986), and the importance of whether the school is a "caring community" (Carnegie Council on Adolescent De- velopment, 1989~. Although these perspectives are gaining attention, for the most part they have not been translated into research and practical applications. To the degree that an interaction among several influences determines the occurrence of problem behaviors (Goodstadt, 1986; Huba et al., 1980), preventive interventions should provide a set of mutually reinforcing posi- tive influences that affect all of the relevant socializing agencies (the peer group, the family, the school, the wider social community). This is in contrast to the notion that only one or two primary variables should be

CONCEPTS OF PREVENTION 63 addressed (e.g., lack of accurate knowledge about drugs, poor resistance or assertiveness skills, early antisocial habits of behavior; see Durlak, 1985; Klitzner et al., 1985~. Of course, a multilevel intervention strategy is much more demanding than one concentrating on one or two variables. One important strand of prevention is focused on reforming the school. This reform movement views prevention not as a circumscribed, limited- duration, add-on module of curriculum designed to contravene certain nega- tive possibilities (Moskowitz, 1987a, 1987b) but as a comprehensive effect of an entire climate of school experience that facilitates and promotes posi- tive, effective socialization. The content of this reform includes revision of organizational structures, classroom management practices, school policies, teacher-student relationships, and instructional approaches with the inten- tion of fostering children's social, personal, and academic development. These reforms are intended to commence with the first school exposure in the primary grades, so that the preventive effects are fully transmitted well before the early second decade when the onset of problems such as illicit drug use which problems are most persistent and least amenable to reme- dial intervention occur. Research Needs Research is particularly needed on the role of school organization, envi- ronment, norms, policies, and social processes and their effects on problem behaviors such as drug and alcohol use, abuse, and dependence. The school as a social institution has received much less attention in research on drug abuse prevention than have the characteristics of individual children, their families, and their peer groups. Psychological paradigms have dominated the prevention research in drug abuse; sociological paradigms have been less influential in this as in other fields of health behavior. Prevention research needs to be diffused across the preschool and el- ementary levels as well as secondary school ages; the balance of concentra- tion has been badly off kilter in the direction of middle and junior high school cohorts, in which the unprevented problems manifest themselves. Only when research is focused on this longer period can we identify critical stages and factors of development if there are any for problems that persist and become increasingly serious in adolescence and hence do a better job of selecting optimal times, types, and intensities of intervention. SOCIAL INFLUENCE AND SOCIAL LEARNING In Chapter 1, we reviewed evidence concerning the role of cigarette smoking as a gateway to further drug consumption. The relationship es- tablished between smoking and other drug use passes various important

64 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? tests of causality: appropriate temporal ordering; a substantial level of correlation, which does not vanish under multivariate analysis; a clearly described and well-studied set of intervening mechanisms (particularly, in this instance, differential access to systems of distribution); the existence of scalable dose-response relationships; and, finally, demonstration that the relationship holds across varying population groups, such as those of differing socioeconomic status. The committee took this not as evidence that cigarette smoking inevitably causes drug use, but as evidence that the prevention of smoking could help forestall, if not prevent, the onset of drug use. Even if cigarettes did not hold this special salience for the onset of illicit drug use, significant attention would have to be given to smoking in this report. For cigarette smoking, due to its well-established role in the genesis of lung cancer, heart disease, and numerous other health problems, has been subject to some of the best-known and well-documented public health promotion and disease prevention campaigns of the last 40 years (see Warner, 1977~. Cigarettes were a major focus not only of mass media programs but also pioneering large-scale experiments in cardiovascular risk reduction beginning in the early 1970s (the Stanford 3-community and 5- community studies by Farquhar and associates [1990] and the North Karelia project in Finland reported by Puska and colleagues [1981; 19853~. The large-scale study of smoking reduction continues today with the city-level COMMIT and state-level ASSIST trials supported by the National Cancer Institute. Smoking was also the focus of an influential school-based prevention program conducted and reported by Evans and colleagues (Evans and Raines, 1982), which has become the model for a succession of closely watched school-based drug abuse prevention programs organized by researchers and conducted along experimental and quasi-experimental lines in the 1980s. The national "Just Say No" campaign publicized by Nancy Reagan leaned on this line of research for its justification. Flay (1987) has defined four generations of such studies, differing in the scale of experimentation, rigor of design, and quality and intensity of measurement: (1) the early pilot studies by Evans and colleagues; (2) more extensive pilot experiments by research groups based at Stanford and Minnesota (McAlister et al., 1980~; (3) substantial field experiments by the latter teams and others in Scandinavia (Puska et al., 1985) and Los Angeles (Johnson et al., 19861; and (4) long- term multisite programs such as the Waterloo trials in Canada (Flay et al., 1985), the Kansas City and Indianapolis STAR studies of the USC Mid- western Prevention Project (Pentz et al., 1989), and the RAND Corporation's Project ALERT (Ellickson and Bell, 1990~. One might add to this last generation a series of more comprehensive school health curriculum evalua- tions directed not specifically at drug abuse prevention but including at

CONCEPTS OF PREVENTION 65 least prevention of smoking onset as a dependent variable (Cornell and Turner, 1 985; Connell et al., 1 9851. Many programs are theory based, specifying which risk factors or me- diating variables they are trying to change and measuring whether these are in fact changed by program exposure. Studies of social influence interven- tion studies have measured changes in information, in specifically instructed interactive skills, and in native expectations regarding alcohol, tobacco, and drug use. MacKinnon et al. (1991) analyzed the first year of the Kan- sas City STAR program and found that a large share of the observed desir- able effects were best explained by changes in normative expectations among program-exposed youth. The fundamental work of Evans and colleagues (Evans, 1976; Evans et al. 1978, 1981) relied heavily on McGuire's (1964) "social inoculation" and "resistance to persuasive communication" theories for background. They drew most heavily, however, on Bandura's (1977, 1982, 1986) theories of social learning and his prescriptions for enhancing perceived self-efficacy: (1) specifying very explicit and proximal goals of training in this case, resistance skills; (2J promoting accomplishments of performance through participation and practice, (3) providing models of successful behavior in this case, peer models; and (4) providing task-specific feedback to reinforce and validate successful performance. The most fully developed, research-based, social-influence programs are cast from a single mold. Virtually all are based on a core of junior high or middle school classroom lessons given by regular teachers, trained "peer leaders," or specialized health educators. The curriculum runs through a sequence of modules attending to predisposing, enabling, and reinforcing factors, with central attention to the development of resistance behaviors against the initial opportunity to use drugs (tobacco, alcohol, or marijuana) in a peer group context. Ellickson et al. (1988:vi-vii) give a cogent sketch of a typical lesson plan, the 7th and 8th grade ALERT program: The first two lessons are intended to develop motivation to resist by sharp- ening students' perception of the seriousness of drug use and by revealing their personal susceptibility to the harmful effects of such use [predisposi- tional factors]. The next three lessons focus on resistance skills helping students to identify pressures to use drugs, counter prodrug messages and learn how to say "no" to both internal and external pressures Enabling factors]. The final three sessions reinforce the earlier content and clarify the benefits of resistance. During the eighth grade, students receive a three-session booster curriculum designed to reinforce resistance skills learned the previous year "reinforcing factors]. The curriculum provides multiple opportunities for student participation- role playing, question and answer techniques, small group activities, indi- vidual and group practice in saying "no," and written exercises.

66 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? There is some diversity among social influence researchers in how nar- rowly or broadly the programs are defined. Pentz et al. (1989) have pro- posed embedding the school-based curriculum within more comprehensive school and community efforts, for example, efforts to invigorate school antidrug policies and to mobilize community-wide awareness and support. Most of the research, however, has been focused on the curriculum compo- nent. There are differences here as well concerning the degree to which there is an emphasis on building general social competence or skills (such as assertiveness) in addition to ones targeted specifically at resisting peer- stimulated drug onset. This division between targeting proximal variables that will affect drug behavior but not (according to design) much else versus generic training that may have effects in many directions is characteristic of the larger school health education field, which has moved increasingly from categorical toward comprehensive programming (Green and Iverson, 1982; Kolbe and Iverson, 1983~. Research Needs A particular problem with social influence models is the implicit as- sumption that school-based influence encompasses all young people. The needs for recognition of many youths, especially economically disadvan- taged children in inner cities, are not well enough served by the schools to lead them to look to schools or even to their peers within the school frame- work for practical or moral instruction. These youths largely define them- selves by their street peer loyalties, not by school district lines. Peer influ- ences, as defined in research literature, are too often generalized as though all adolescents were culturally homogeneous; there is not enough research that recognizes the specific features of ethnic and street culture (Becker et al., 1989~. The foundations of social influence theory were in relatively small- scale social psychological studies, and more of these are needed now to extend our understanding of influence processes. More fundamental re- search is needed on small groups with a variety of youth-cultural affilia- tions. The careful studies in the l950s and 1960s of institutionalized street gangs, including attempts to change them, are a model worth recon . . SIC .erlIlg. SUMMARY Three principal approaches in drug abuse prevention research emerge from the recent past: the study of risk factors, the study of developmental sequences, and the study of social influence. It is helpful in seeing how these approaches relate to each other to note their differential emphasis on

CONCEPTS OF PREVENTION 67 predisposing, enabling, and reinforcing elements or variables in the respec- tive theories and methods of inquiry. The risk factors under study include biological, personality, and socio- economic variables. In general, under longitudinal study, risk factors seem to operate as individually small but cumulative causes of criterion behav- iors. These studies generally suggest prevention strategies based on identi- fication of the high-risk youths, those for whom many such factors apply. Studies of risk factors are hobbled by measurement deficiencies with re- spect to environmental variables in particular, and methodological invest- ments and improvements in this respect are needed. The developmental approach involves a more structured, sequential model of poor early parenting, school maladjustment, academic deficiency, and gravitation toward school-oppositional groups, which are seedbeds of illicit drug use and other disorderly and problem behaviors. This approach incor- porates the general sense that there is a weakening of family bonds through- out the population and that primary schools, which may be more amenable to intervention particularly experimental intervention than family units, should be a key locus of study. The study of social influences, largely in junior high school popula- tions, has also been based on a highly structured theory derived from the concept of self-efficacy and its roots in social learning. While these theo- retical foundations have been extensively researched and appear robust in many ways, there has not been enough study of the differentiated social and normative world of early adolescence. This applies particularly to the emergence and significance of norms strongly antagonistic to schools and to the per- ception by adolescents of prodrug or antidrug norms in their peers. These are critical reinforcing environments that may make or break intervention strategies, so it is critical to build a more systematic understanding of them. REFERENCES Ames, C. in The enhancement of student motivation, In M.L. Maehr and D.A. Kleiber, eds., press Advances in Motivation and Achievement, Vol. 5. Greenwich, Conn.: JAI Press. Aneshensel, C.S., and G.J. Huba 1983 Depression, alcohol use, and smoking over one year: a four-wave longitudinal causal model. Journal of Abnormal Psychology 92:134-150. Bamberg, R., R.T. Acton, J.M. Roseman, R.C.P. Go, B.O. Barger, C.J. Vanichanan, and R.B. Copeland 1990 The effect of genetic risk information and health risk assessment on compliance with preventive behaviors. Health Education 21(2):26-32. Bandura, A. 1977 Social Learning Theory. Englewood Cliffs, N.J.: Prentice-Hall. Bandura, A. 1982 Self-efficacy mechanisms in human agency. American Psychologist 37:122-147.

68 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? Bandura, A. 1986 Social Foundations of Thought and Action: A Social Cognitive Theory. Englewood Cliffs, N.J.: Prentice-Hall. Bates, M.E., E.W. Labouvie, and H.R. White 1985 The effect of sensation seeking needs on alcohol and marijuana use in adoles- cence. Bulletin of HTE Society of Psychologists in Addictive Behaviors 5:29-36. Becker, M.H., and N.K. Janz 1987 On the effectiveness and utility of health hazard/health risk appraisal in clinical and non-clinical settings. Health Services Research 22:537-551. Becker, S.L., J.A. Burke, R.A. Arbogast, M.J. Naughton, I. Bachman, and E. Spohn 1989 Community programs to enhance in-school anti-tobacco efforts. Preventive Medi- cine 18:221-228. Bensley, L.B. 1981 Health risk appraisals in teaching health education in colleges and universities. Health Education 12:31-33 Blount, W.R., and R. Dembo 1984 The effect of perceived neighborhood setting on self-reported tobacco, alcohol, and marijuana use among inner-city minority junior high school youth. The Inter- national Journal of the Addictions 19(2): 175- 198. Botvin, G.J., and A. Eng 1980 A comprehensive school-based smoking prevention program. Journal of School Health 50:209-213. Botvin, G.J., A. Eng, and C.L. Williams 1980 Preventing the onset of cigarette smoking through life skills training. Preventive Medicine 9: 135- 143. Brook, J.S., D.W. Brook, A.S. Gordon, M. Whiteman, and P. Cohen 1990 The psychosocial etiology of adolescent drug use: a family interactional approach. Genetic, Social, and General Psychology Monographs 116:111-267. Brunswick, A.F. 1988 Drug use and effective distress: a longitudinal study of black youth. Advances in Adolescent Mental Health 3:101-125. Bry, B.H., P. McKeon, and R.J. Pandina 1982 Extent of drug use as a function of number of risk factors. Journal of Abnormal Psychology 91 :273-279. Carnegie Council on Adolescent Development 1989 Turning Points: Preparing American Youth for the 21st Century. New York: Carnegie Corporation. Childs, B. 1974 A place for genetics in health education and vice versa. American Journal of Human Genetics 26:120-135. Cloninger, C., M. Bohman, and S. Sigvardsson 1981 Inheritance of alcohol abuse: cross fostering analyses of adopted men. Archives of General Psychiatry 38:861-868. Connell, D.B., and R.R. Turner 1985 The impact of instructional experience and the effects of cumulative instruction. Journal of School Health 55:324-331. Connell, D.B., R.R. Turner, and E.F. Mason 1985 Summary of findings of the school health education evaluation: health promotion effectiveness, implementation and costs. Journal of School Health S5:316-321. Deci, E.L., and R.M. Ryan 1985 Intrinsic Motivation and Self-Determination in Human Behavior. New York: Ple- num.

CONCEPTS OF PREVENTION 69 Dembo, R., L. Williams, J. Schmeidler, E.D. Wish, A. Getreu, and E. Berry 1991 Juvenile crime and drug abuse: a prospective study of high risk youth. Journal of Addictive Diseases 11:5-31. Durlak, J.A. 1985 Primary prevention of school maladjustment. Journal of Consulting and Clinical Psychology 53:623-620. Ellickson, P.L., and R.M. Bell 1990 Drug prevention in junior high: a multi-site longitudinal test. Science 247:1299- 1305. Ellickson, P.O., R.M. Bell, M.A. Thomas, A.E. Robyn, and G.L. Zellman 1988 Designing and Implementing Project ALERT: A Smoking and Drug Prevention Experiment. Santa Monica, Calif.: The RAND Corporation. Elliott, D.S., and D. Huizinga 1984 The Relationship Between Delinquent Behavior and ADM Problem Behaviors. Paper prepared for the ADAMHA/OJJDP State of the Art Research Conference on Juvenile Offenders with Serious Drug/Alcohol and Mental Health Problems, Bethesda, Md. Elliott, D.S., and B.J. Morse 1987 Drug Use, Delinquency, and Sexual Activity. Unpublished paper. Elliott, D.S., D. Huizinga, and S.S. Aceton 1985 Explaining Delinquency and Drug Use. Beverly Hills, Calif.: Sage Publications. Elliott, D.S., D. Huizinga, and S. Menard 1989 Multiple Problem Youth: Delinquency, Substance Use, and Mental Health Prob lems. New York: Springer-Verlag. Evans, R.I. 1976 Smoking in children: developing a social psychological strategy of deterrence. Preventive Medicine 5:122-127. Evans, R.I., and B.E. Raines 1982 Control and prevention of smoking in adolescents: a psychological perspective. In T.J. Coates, A.D. Peterson, and C. Perry, eds., Promoting Adolescent Health: A Dialog on Research and Practice. New York: Academic Press. Evans, R.I., R.M. Rozelle, M. Mittelmark, W.B. Hansen, A. Bane, and J. Havis 1978 Deterring the onset of smoking in children: knowledge of immediate psychologi- cal effects and coping with peer pressure, media pressure, and parent modeling. Journal of Appl~ed Social Psychology 8:126-135. Evans, R.I., R.M. Rozelle, S.E. Maxwell, B.E. Raines, C.A. Dill, T.J. Guthrie, A.H. Henderson, and P.C. Hill 1981 Social modeling films to deter smoking in adolescents: results of a three-year field investigation. Journal of Applied Psychology 66:399-414. Farquhar, J.W., S.P. Fortmann, J.A. Flora, C.B. Taylor, W.L. Haskell, P.T. Williams, N. Maccoby, and P.D. Wood Effects of community-wide education on cardiovascular disease risk factors the Stanford 5-city Project. Journal of the American Medical Association 264:359- 365. 1990 Fitzpatrick, J. 1990 Drugs and Puerto Ricans in New York City. Pp. 103-126 in R. Glick and J. Moore, eds., Drugs in Hispanic Communities. New Brunswick, N.J.: Rutgers University. Flay, B.R. 1987 Social psychological approaches to smoking prevention: review and recommenda tions. Pp. 121-180 in Advances in Health [Education and Promotion, Vol. 2. Greenwich, Conn.: JAI Press Inc.

70 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? Flay, B.R., K.K. Ryan, J.A. Best, K.S. Brown, M.W. Kersell, J.R. d'Avernas, and M.P. Zanna 1985 Are social psychological smoking prevention programs effective? The Waterloo study. Journal of Behavioral Medicine 8:37-59. Goodlad, J.I. 1984 A Place Called School. New York: McGraw-Hill. Goodstadt, M.S. 1986 School-based drug education in North America: what is wrong? What can be done? Journal of School Health 56:278-281. Goodwin, D. 1983 Alcoholism. Pp. 195-213 in R. Tarter, ea., The Child at Psychiatric Risk. New York: Oxford University Press. Gottlieb, N.H., and L.W. Green 1987 Ethnicity and lifestyle health risk: some possible mechanisms. American Journal of Health Promotion 2:37-45. Green, L.W. 1990 The revival of community and the public obligation of academic health centers. Pp. 148-164 in R. Bulger, R.E. Bulger, and S. Reiser, eds., The Role of the Aca- demic Health Center in Humani ing Medicine. Ames, Iowa: University of Iowa Press. Green, L.W., and D.C. Iverson 1982 School health education. Annual Review of Public Health 3:321-338. Green, L.W., and M.W. Kreuter 1991 Health Promotion Planning: An Educational and Environmental Approach. Palo Alto, Calif.: Mayfield. Hawkins, J.D., D.M. Lishner, and R.F. Catalano 1987 Schooling and delinquency. In E.H. Johnson, ea., Handbook on Crime and Delin- quency Prevention. Westport, Conn.: Greenwood Press. Heckler, M. 1985 Report of the Secretary's Task Force on Black and Minority Health. Washington, D.C.: U.S. Department of Health and Human Services. Huba, G.J., J.A. Wingard, and P.M. gentler 1980 Applications of a theory of drug use to prevention programs. Journal of Drug Education 10:25-38. Huba, G.J., J.A. Wingard, and P.M. gentler 1981 A comparison of two latent causal variable models for adolescent drug use. Jour- nal of Personality and Social Psychology 40:180-193. Huba, G.J., M.D. Newcomb, and P.M. gentler 1986 Adverse drug experience and drug use behaviors: a one-year longitudinal study of adolescents. Journal of Ped iatric Psychology 11 :203-219. Hunt, S.C., R.R. Williams, and K.K. Barlow 1986 A comparison of positive family history definitions for defining risk of future disease. Journal of Chronic Diseases 39:809-821. Institute of Medicine 1989 Prevention and Treatment of Alcohol Problems: Research Opportun~ties. Com mittee to Identify Research Opportunities in the Prevention and Treatment of Al cohol-Related Problems. Washington, D.C.: National Academy Press. Jacob, J. 1987 Black America, 1986: an overview. In The St:ate of Black America. New York: National Urban League. Jessor, R., and S.L. Jessor 1977 Problem Behavior and Bsychosocial Development: A Longitudinal Study of Youth. New York: Academic Press.

CONCEPTS OF PREVENTION 71 Jessor, R., J.E. Donovan, and F.M. Costa 1992 Beyond Adolescence: Problem Behavior and Young Adult Development. New York: Academic Press. Johnson, B.D., P.J. Goldstein, E. Preble, J. Schmeidler, D.S. Lipton, B. Spunt, and T. Miller 1985 Taking Care of Business: The Economics of Crime by Heroin Abusers. Lexing- ton, Mass.: Lexington Books. Johnson, C.A., W.B. Hansen, L.M. Collins, and J.W. Graham 1986 High-school smoking prevention: results of a three-year longitudinal study. Jour- nal of Behavioral Medicine 9(5):439-452. Jones, M.C. 1968 Personality correlates and antecedents of drinking patterns in adult males. Journal of Consulting and Clinical Psychology 32:2-12. Jones, M.C. 1971 Personality antecedents and correlates of drinking patterns in women. Journal of Consulting and Clinical Psychology 36:61-69. Kandel, D.B., O. Simcha-Fagan, and M. Davies 1986 Risk factors for delinquency and illicit drug use from adolescence to young adult- hood. Journal of Drug Issues 16:67-90. Kaplan, H.B. 1985 Testing a general theory of drug abuse and other deviant adaptations. Journal of Drug Issues 15:477-492. Kaplan, H.B. 1986 Social Psychology Self-Referrent Behavior. New York: Plenum. Kaplan, H.B., S.S. Martin, and C. Robbins 1984 Pathways to adolescent drug use: self-derogation, peer influence, weakening of social controls, and early substance use. Journal of Health and Social Behavior 25:270-289. Kaplan, H.B., R.J. Johnson, and C.A. Bailey 1988 Explaining adolescent drug use: an elaboration strategy for structural equation modeling. Psychiatry 51:142- 163. Katz, L.G. 1989 Engaging Children's Minds: The Project Approach. Norwood, N.J.: Ablex. Kellam, S.G., C.H. Brown, B.R. Rubin, and M.E. Ensminger 1983 Paths leading to teenage psychiatric symptoms and substance abuse: developmen- tal epidemiological studies in Woodlawn. In. S.B. Guze, F.J. Earls, and J.E. Barrett, eds., Childhood Psychopathology and Development. Vienna, Va.: Center for Advanced Health Studies. Khowry, M.J., C.A. Newill, and G.A. Chase 1985 Epidemiological evaluation of screening for risk factors: application to genetic screening. American Journal of Public Health 75:1204-1208. Klitzner, M., M. Blasinsky, K. Marshall, and U. Paquet 1985 Determinants of Youth Attitudes and Skills Towards Which DrinkinglDriving Pre- vention Programs Should be Directed, Vol. 1. Report No. DOT HS 806-903. Springfield, Va.: National Technical Information Service. Knitzer, J., Z. Steinberg, and B. Fleish 1990 At the Schoolhouse Door: An Examination of Programs and Policies for Children with Behavioral and Emotional Problems. New York: Bank Street College of Education. Kolbe, L.J., and D.C. Iverson 1983 Integrating school and community efforts to promote health: strategies, policies and methods. Hygie: International Journal of Health Education 3:40-47.

72 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? Kumpfer, K.L. 1989 Effective Parenting Strategies for High Risk Youth and Families: Literature Re- view. Social Research Institute, Graduate School of Social Work, University of Utah. Labouvie, E.W. 1986 Alcohol and marijuana use in relation to adolescent stress. International Journal of the Addictions 21:333-345. Labouvie, E.W., R.J. Pandina, and V. Johnson in press Developmental trajectories of substance use in adolescence: differences and pre dictors. International Journal of Behavioral Development. Lang, A.R. 1983 Addictive personality: a viable construct? Pp. 157-236 in P.K. Levison, D.R. Gerstein, and D.R. Maloff, eds., Commonalit~es in Substance Abuse and Habitual Behavior. Lexington, Mass.: D.C. Health and Company. Lorion, R.P., D. Bussell, and R. Goldberg 1989 Identification and Assessment of Youths at High Risk of Substance Abuse. Rockville, Md.: Office of Substance Abuse Prevention. MacKinnon, D.P., C.A. Johnson, M.A. Pentz, J.H. Dwyer, W.B. Hanson, B.R. Flay, and E.Y.I. Ward 1991 Mediating mechanisms in a school-based drug prevention program: first-year effects of the Midwestern Prevention Project. Health Psychology 10(3):164-172. Magnussen, D., H. Statin, and V. Allen 1986 Differential maturation among girls and its relation to social adjustment: a longi- tudinal perspective. In P. Baltes, D. Featherman, and R. Lerner, eds., Life Span Development and Behavior. Hillsdale, N.J.: Erlbaum. McAlister, A.L., C. Perry, J. Killen, L.A. Slinkard, and N. Maccoby 1980 Pilot study of smoking, alcohol, and drug abuse prevention. American Journal of Public Health 70-719-721. McCord, W., and J. McCord 1962 A longitudinal study of the personality of alcoholics. In D.J. Pittman and C.R. Snyder, eds., Society, Culture, and Drinking Patterns. New York: Wiley. McGuire, W.J. 1964 Inducing resistance to persuasion. In L. Berkowitz, ea., Advances ~n Experimental Social Psychology, Vol. 1. New York: Academic Press. McMillan, D.W., and D.M. Chavis 1986 Sense of community: a definition and theory. Journal of Community Psychology 14:6-23. Monnelly, E.P., E.M. Hartl, and R. Elderkin 1983 Constitutional factors predictive of alcoholism in a follow-up of delinquent boys. Journal of Studies on Alcohol 44:530-537. Moskowitz, J.M. 1987a The Primary Prevention of Alcohol Problems: A Critical Review of the Research Literature. Berkeley, Calif.: Prevention Research Center. Moskowitz, J.M. 1987b School Drug and Alcohol Policy: A Preliminary Model Relating Implementat~on to School Problems. Berkeley, Calif.: Prevention Research Center. National Commission on the Role of the School and the Community in Improving Adolescent Health 1990 Code Blue: Uniting for Healthier Youth. Alexandria, Va.: National Association of State Boards of Education, and Chicago, Ill.: American Medical Association. Newcomb, M.D., and P.M. gentler 1988 Consequences of Adolescent Drug Use. Beverly Hills, Calif.: Sage Publications.

CONCEPTS OF PREVENTION 73 Newcomb, M.D., and P.M. gentler 1989 Substance use and abuse among children and teenagers. American Psychologist 44:242-248. Newcomb, M.D., and L. McGee 1989 Alcohol use and other delinquent behaviors: a one-year longitudinal analysis controlling for sensation-seeking. Criminal Justice Behavior 16:345-369. Newcomb, M.D., E. Maddahian, and P.M. gentler 1986 Risk factors for drug use among adolescents: concurrent and longitudinal analy- sis. American Journal of Public Health 76(5):525-531. Newcomb, M.D., E. Maddahian, R. Skager, and P.M. gentler 1987 Substance abuse and psychosocial risk factors among teenagers: associations with sex, age, ethnicity, and type of school. American Journal of Drug and Alcohol Abuse 13(4):413-433. Nicholls, J.G. 1989 The Competitive Ethos and Democratic Education. Cambridge, Mass.: Harvard University Press. Getting, E.R., and F. Beauvais 1987 Common elements in youth drug abuse: peer clusters and other psychosocial factors. Journal of Drug Issues (Spring): 133-151. Pandina, R.J., E.W. Labouvie, and H.R. White 1984 Potential contributions of the life span developmental approach to the study of adolescent alcohol and drug use. Journal of Drug Issues 14:253-270. Pentz, M.A., C. Cormack, B. Flay, W.B. Hansen, and C.A. Johnson 1986 Balancing program and research integrity in community drug abuse prevention: project STAR approach. Journal of School Health 56:389-393. Pentz, M.A., J.H. Dwyer, D.P. MacKinnon, B.R. Flay, W.B. Hansen, E.Y.I. Wang, and C.A. Johnson 1989 A multi-community trial for primary prevention of adolescent drug abuse: effects on drug use prevalence. Journal of the American Medical Association 261:3259- 3266. Pickens, R., and D. Svikis, eds. 1988 Biological Vulnerability to Drug Abuse. NIDA Research Monograph 89. Rockville, Md.: National Institute on Drug Abuse. Pickens, R.W., D.S. Svikis, M. McGue, D.T. Lykken, L.L. Heston, and P.J. Clayton 1991 Heterogeneity in the inheritance of alcoholism. A study of male and female twins. Archives of General Psychiatry 48(1):19-28. Preble E., and J.J. Casey 1966 Taking care of business: the heroin user's life on the street. International Journal of the Addictions 4: 1 -24. Puska, P., A. McAlister, J. Pekkola, and K. Koskela 1981 Television in health promotion: evaluation of a national programme in Finland. International Journal of Health Education 24:2-14. Puska, P., A. Nissinen, J. Tuomilehto, J.T. Salonen, K. Koskela, A. McAlister, T.E. Kottke, N. Maccoby, and J.W. Farquhar 1985 The community-based strategy to prevent coronary heart disease: conclusions from the ten years of the North Karelia Project. Annual Review of Public Health 6: 147-193. Rennert, M.P., M.J. Telch, J.W. Farquhar, and J.D. Killen 1986 Disease and risk factor clustering in the United States: the implications for public health policy. In lnteg'-ation of Risk Factor Interventions: Two Reports to the Office of Disease Prevention and Health Promotion. Washington, D.C.: U.S. Department of Health and Human Services.

74 PREVENTING DRUG ABUSE: WHAT DO WE KNOW? Resnick, L.B., ed. 1989 Knowing, Learning, and Instruction: Essays in Honor of Robert Glaser. Hillsdale, N.J.: Lawrence Erlbaum. Reuter, P., R. MacCoun, and P. Murphy 1990 Money from Crime. A Study of the Economics of Drug Dealing in Washington, D.C. Publication R-3894-RF. Santa Monica, Calif.: The Rand Corporation. Ricks, D., and J.C. Berry 1970 Family and symptom patterns that precede schizonphrenia. In M. Roff and D.F. Hicks, eds., Life History Research in Psychopathology. Minneapolis, Minn.: Uni- versity of Minnesota Press. Robins, L.N. 1966 Deviant Children Grown Up: A Sociological and Psychiatric Study of Sociopathic Personality. Baltimore, Md.: Williams & Wilkins. Robins, L.N. 1978 Sturdy childhood predictors of adult antisocial behavior: replications from longi- tudinal studies. Psychological Medicine 8:611-622. Ryan, K. 1986 The new moral education. Phi Delta Kappan (November):228-233. Schaps, E., and V. Battistich 1991 Chapter in E. Goplerud, ea., A Practical Guide to Substance Abuse Prevention in Adolescent. OSAP Prevention Monograph 8. DHHS Publ No. (ADM) 91-1725. Washington, D.C.: U.S. Government Printing Office. Schinke, S.P. 1982 A school-based model for teenage pregnancy prevention. Social Work Education 4:32-42. Schinke, S.P., and L.D. Gilchrest 1977 Adolescent pregnancy: an interpersonal skill training approach to prevention. Social Work in Health Care 3:159-167. Schroeder, P. 1989 Champion of the Great American Family. New York: Random House. Simcha-Fagen, O., J.C. Gersten, and T.S. Langer 1986 Early precursors and concurrent correlates of patterns of illicit drug use in adoles- cents. Journal of Drug Issues 16(1):7-28. Thomas, S.B. 1990 Social change theory: applications to community health. Pp. 45-65 in N. Bracht, ea., Health Promotion at the Community Level. Newbury Park, Calif.: Sage Publications. Warner, K.E. 1977 The effects of the anti-smoking campaign on cigarette consumption. American Journal of Public Health 67(7):645-650. White, H.R. 1990 The drug use-delinquency nexus in adolescence. In R. Weisheit, ea., Drugs, Crime and the Criminal Justice System. Champaign, Ill.: Anderson. White, H.R., V. Johnson, and A. Horowitz 1986 An application of three deviance theories to adolescent substance use. Interna- tional Journal of the Addictions 21(3):347-3 66. Williams, T.M., and W. Kornblum 1986 Growing Up Poor. Lexington, Mass.: Lexington Books. Willis, P. 1977 Learning to Labor: How Working Class Kids Get Working Class Jobs. New York: Columbia University Press.

CONCEPTS OF PREVENTION 75 Zimiles, H. 1986 The changing American child: the perspective of educators. Pp. 61-84 in T.M. Tomlinson and H.J. Walberg, eds., Academic Work and Educational Excellence. MSSE Yearbook. Berkeley, Calif.: McCutcheon Publishing. Zimmerman, M.A., and K.I. Maton 1992 Lifestyle and substance use among male African-American urban adolescents: a cluster analytic approach. American Journal of Community Psychology 20:121- 138. Zuckerman, M. 1979 Sensation Seeking: Beyond the OptimalLevel of Arousal. Hillsdale, N.J.: Erlbaum. Zylke, J. 1987 Once identified, will high-risk families make lifestyle changes, lower coronary disease rate? Medical News and Perspective. Journal of the American Medical Association 258(4):