National Institute on Alcohol Abuse and Alcoholism

No. 37 July 1997

----------------------------------------------------------------------

Youth Drinking: Risk Factors and Consequences

Despite a minimum legal drinking age of 21, many young people in the United States consume alcohol. Some abuse alcohol by drinking frequently or by binge drinking--often defined as having five or more drinks* in a row. A minority of youth may meet the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria for alcohol dependence (1,2). The progression of drinking from use to abuse to dependence is associated with biological and psychosocial factors. This Alcohol Alert examines some of these factors that put youth at risk for drinking and for alcohol-related problems and considers some of the consequences of their drinking.

Prevalence of Youth Drinking

Thirteen- to fifteen-year-olds are at high risk to begin drinking (3). According to results of an annual survey of students in 8th, 10th, and 12th grades, 26 percent of 8th graders, 40 percent of 10th graders, and 51 percent of 12th graders reported drinking alcohol within the past month (4). Binge drinking at least once during the 2 weeks before the survey was reported by 16 percent of 8th graders, 25 percent of 10th graders, and 30 percent of 12th graders.

Males report higher rates of daily drinking and binge drinking than females, but these differences are diminishing (3). White students report the highest levels of drinking, blacks report the lowest, and Hispanics fall between the two (3).

A survey focusing on the alcohol-related problems experienced by 4,390 high school seniors and dropouts found that within the preceding year, approximately 80 percent reported either getting "drunk," binge drinking, or drinking and driving. More than half said that drinking had caused them to feel sick, miss school or work, get arrested, or have a car crash (5).

Some adolescents who drink later abuse alcohol and may develop alcoholism. Although these conditions are defined for adults in the DSM, research suggests that separate diagnostic criteria may be needed for youth (6).

Drinking and Adolescent Development

While drinking may be a singular problem behavior for some, research suggests that for others it may be an expression of general adolescent turmoil that includes other problem behaviors and that these behaviors are linked to unconventionality, impulsiveness, and sensation seeking (7-11).

Binge drinking, often beginning around age 13, tends to increase during adolescence, peak in young adulthood (ages 18-22), then gradually decrease. In a 1994 national survey, binge drinking was reported by 28 percent of high school seniors, 41 percent of 21- to 22-year-olds, but only 25 percent of 31- to 32-year-olds (3,12). Individuals who increase their binge drinking from age 18 to 24 and those who consistently binge drink at least once a week during this period may have problems attaining the goals typical of the transition from adolescence to young adulthood (e.g., marriage, educational attainment, employment, and financial independence) (13).

Risk Factors for Adolescent Alcohol Use, Abuse, and Dependence

Genetic Risk Factors. Animal studies (14) and studies of twins and adoptees demonstrate that genetic factors influence an individual's vulnerability to alcoholism (15,16). Children of alcoholics are significantly more likely than children of nonalcoholics to initiate drinking during adolescence (17) and to develop alcoholism (18), but the relative influences of environment and genetics have not been determined and vary among people.

Biological Markers. Brain waves elicited in response to specific stimuli (e.g., a light or sound) provide measures of brain activity that predict risk for alcoholism. P300, a wave that occurs about 300 milliseconds after a stimulus, is most frequently used in this research. A low P300 amplitude has been demonstrated in individuals with increased risk for alcoholism, especially sons of alcoholic fathers (19,20). P300 measures among 36 preadolescent boys were able to predict alcohol and other drug (AOD) use 4 years later, at an average age of 16 (21).

Childhood Behavior. Children classified as "undercontrolled" (i.e., impulsive, restless, and distractible) at age 3 were twice as likely as those who were "inhibited" or "well-adjusted" to be diagnosed with alcohol dependence at age 21 (22). Aggressiveness in children as young as ages 5-10 has been found to predict AOD use in adolescence (23,24). Childhood antisocial behavior is associated with alcohol-related problems in adolescence (24-27) and alcohol abuse or dependence in adulthood (28,29).

Psychiatric Disorders. Among 12- to 16-year-olds, regular alcohol use has been significantly associated with conduct disorder; in one study, adolescents who reported higher levels of drinking were more likely to have conduct disorder (30,31).

Six-year-old to seventeen-year-old boys with attention deficit hyperactivity disorder (ADHD) who were also found to have weak social relationships had significantly higher rates of alcohol abuse and dependence 4 years later, compared with ADHD boys without social deficiencies and boys without ADHD (32).

Whether anxiety and depression lead to or are consequences of alcohol abuse is unresolved. In a study of college freshmen, a DSM-III (33) diagnosis of alcohol abuse or dependence was twice as likely among those with anxiety disorder as those without this disorder (34). In another study, college students diagnosed with alcohol abuse were almost four times as likely as students without alcohol abuse to have a major depressive disorder (35). In most of these cases, depression preceded alcohol abuse. In a study of adolescents in residential treatment for AOD dependence, 25 percent met the DSM-III-R criteria for depression, three times the rate reported for controls. In 43 percent of these cases, the onset of AOD dependence preceded the depression; in 35 percent, the depression occurred first; and in 22 percent, the disorders occurred simultaneously (36).

Suicidal Behavior. Alcohol use among adolescents has been associated with considering, planning, attempting, and completing suicide (37-39). In one study, 37 percent of eighth-grade females who drank heavily reported attempting suicide, compared with 11 percent who did not drink (40). Research does not indicate whether drinking causes suicidal behavior, only that the two behaviors are correlated.

Psychosocial Risk Factors

Parenting, Family Environment, and Peers. Parents' drinking behavior and favorable attitudes about drinking have been positively associated with adolescents' initiating and continuing drinking (41,42). Early initiation of drinking has been identified as an important risk factor for later alcohol-related problems (43). Children who were warned about alcohol by their parents and children who reported being closer to their parents were less likely to start drinking (42,44,45).

Lack of parental support, monitoring, and communication have been significantly related to frequency of drinking (46), heavy drinking, and drunkenness among adolescents (47). Harsh, inconsistent discipline and hostility or rejection toward children have also been found to significantly predict adolescent drinking and alcohol-related problems (46).

Peer drinking and peer acceptance of drinking have been associated with adolescent drinking (48,49). While both peer influences and parental influences are important, their relative impact on adolescent drinking is unclear.

Expectancies. Positive alcohol-related expectancies have been identified as risk factors for adolescent drinking. Positive expectancies about alcohol have been found to increase with age (50) and to predict the onset of drinking and problem drinking among adolescents (51-53).

Trauma. Child abuse and other traumas have been proposed as risk factors for subsequent alcohol problems. Adolescents in treatment for alcohol abuse or dependence reported higher rates of physical abuse, sexual abuse, violent victimization, witnessing violence, and other traumas compared with controls (54). The adolescents in treatment were at least 6 times more likely than controls to have ever been abused physically and at least 18 times more likely to have ever been abused sexually. In most cases, the physical or sexual abuse preceded the alcohol use. Thirteen percent of the alcohol dependent adolescents had experienced posttraumatic stress disorder, compared with 10 percent of those who abused alcohol and 1 percent of controls.

Advertising. Research on the effects of alcohol advertising on adolescent alcohol-related beliefs and behaviors has been limited (55). While earlier studies measured the effects of exposure to advertising (56), more recent research has assessed the effects of alcohol advertising awareness on intentions to drink. In a study of fifth- and sixth-grade students' awareness, measured by the ability to identify products in commercials with the product name blocked out, awareness had a small but statistically significant relationship to positive expectancies about alcohol and to intention to drink as adults (57). This suggests that alcohol advertising may influence adolescents to be more favorably predisposed to drinking (57).

Consequences of Adolescent Alcohol Use

Drinking and Driving. Of the nearly 8,000 drivers ages 15-20 involved in fatal crashes in 1995, 20 percent had blood alcohol concentrations above zero (58). For more information about young drivers' increased crash risk and the factors that contribute to this risk, see Alcohol Alert No. 31: Drinking and Driving (59).

Sexual Behavior. Surveys of adolescents suggest that alcohol use is associated with risky sexual behavior and increased vulnerability to coercive sexual activity. Among adolescents surveyed in New Zealand, alcohol misuse was significantly associated with unprotected intercourse and sexual activity before age 16 (60). Forty-four percent of sexually active Massachusetts teenagers said they were more likely to have sexual intercourse if they had been drinking, and 17 percent said they were less likely to use condoms after drinking (61).

Risky Behavior and Victimization. Survey results from a nationally representative sample of 8th and 10th graders indicated that alcohol use was significantly associated with both risky behavior and victimization and that this relationship was strongest among the 8th-grade males, compared with other students (62).

Puberty and Bone Growth. High doses of alcohol have been found to delay puberty in female (63) and male rats (64), and large quantities of alcohol consumed by young rats can slow bone growth and result in weaker bones (65). However, the implications of these findings for young people are not clear.

Prevention of Adolescent Alcohol Use

Measures to prevent adolescent alcohol use include policy interventions and community and educational programs. Alcohol Alert No. 34: Preventing Alcohol Abuse and Related Problems (66) covers these topics in detail. See the National Institute on Alcohol Abuse and Alcoholism's (NIAAA's) World Wide Web site at http://www.niaaa.nih.gov.

----------------------------------------------------------------------

Youth Drinking: Risk Factors and Consequences--A Commentary by
NIAAA Director Enoch Gordis, M.D.

Alcohol, the most widely used and abused drug among youth, causes serious and potentially life-threatening problems for this population. Although alcohol is sometimes referred to as a "gateway drug" for youth because its use often precedes the use of other illicit substances, this terminology is counterproductive; youth drinking requires significant attention, not because of what it leads to but because of the extensive human and economic impact of alcohol use by this vulnerable population.

For some youth, alcohol use alone is the primary problem. For others, drinking may be only one of a constellation of high-risk behaviors. For these individuals, interventions designed to modify high-risk behavior likely would be more successful in preventing alcohol problems than those designed solely to prevent the initiation of drinking. Determining which influences are involved in specific youth drinking patterns will permit the design of more potent interventions. Finally, we need to develop a better understanding of the alcohol treatment needs of youth. Future questions for scientific attention include, what types of specialized diagnostic and assessment instruments are needed for youth; whether treatment in segregated, "youth only" programs is more effective than in general population programs; and, irrespective of the setting, what types of specific modalities are needed by youth to increase the long-term effectiveness of treatment.

----------------------------------------------------------------------

*A standard drink is 12 grams of pure alcohol, which is equal to one 12-ounce bottle of beer or wine cooler, one 5-ounce glass of wine, or 1.5 ounces of 80-proof distilled spirits.

----------------------------------------------------------------------

----------------------------------------------------------------------

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES

Public Health Service * National Institutes of Health

By David Shepardson / The Detroit News

Dr. Paul Taheri, medical director of the University of Michigan Medical Centers' trauma and burn unit, wants to stop drunken driving accidents.

"We see more than 1,000 trauma victims per year, and 80 percent are automobile-accident related," Taheri said. "Of those, 50 percent -- or at least one a day -- is because of a drunk driver."

In an effort to scare teen-agers with a stiff dose of reality and steer them away from alcohol abuse, the University of Michigan is sponsoring a new program to keep adolescents off drugs and alcohol.

"These accidents are absolutely terrible and preventable," Taheri said.

Called Facing Alcohol Challenges Together, the program brings primarily high-risk youths and parents together to see the possible consequences of alcohol and drug abuse.

The medical center plans to serve about 250 young people per year. Many will attend the program as part of a court-ordered alternative sentencing program.

"Most of them will be referred to us by the court, especially for drunk-driving convictions," Taheri said. "If they don't come and bring their parents, they will have to carry out their sentence."

Taheri said the program was important because young drunken drivers show a high rate of reoccurrence.

"Between 50 (percent) to 80 percent of kids who drink and drive get caught again," he said.

More than 30 doctors, nurses and staffers at the hospital volunteer two afternoons every other week for the program.

The six-hour program spread over two half-days combines role playing and frank discussions about drugs and alcohol with a blunt look at the effects of traumatic accidents on the body. The teens also see the costs to the victim's family.

In one role-playing scenario, the youths witness a real nurse telling a mother of an accident victim that her child is dead. They hear a chaplain giving last rites to a pretend victim. They watch as hospital staff go over the bill with the parents.

"These are everyday, actual things that go on in the trauma unit," said Pam Pucci, a registered nurse at the trauma-burn unit and another coordinator of the program.

Parents and young people who attended the first session said they learned a lot.

"It was an unbelievable dose of reality," said Karen Nutting of Brighton, who went through the first run of the university-sponsored program Wednesday night with her daughter, Rachel.

Rachel, 12, said she thought the program could help youths resist peer pressure.

"There are kids in my neighborhood already caught in the drug web," she said. "They already have problems and they're still in middle school."

The program is based on a similar program that began a little over a year ago at Methodist Hospital in Indianapolis, Taheri said. University researchers will do follow-up interviews with the participants for several years to determine the program's effectiveness.

The Associated Press contributed to this report.

Simin Liu, M.D., M.S., and Robert D. Brewer, M.D., M.S.P.H., from the National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, Ga., and colleagues estimated how frequently adults in the U.S. drive while impaired by alcohol. Dr. Liu is now with the Harvard School of Public Health, Boston, Mass.

They write: "Despite the enactment and enforcement of stricter legislation in many states, 2.5 percent of survey respondents reported alcohol-impaired driving during the month before the interview. Based on these self-reports, we estimate that there were nearly 123 million episodes of alcohol-impaired driving among adults in the U.S. during 1993; nearly ten million of these events occurred among persons aged 18 to 20 years. This estimate is 82 times higher than the 1.5 million arrests for driving while intoxicated in the U.S. that year."

The study included 102,263 adults age 18 and older, from 49 states and Washington, D.C., who were surveyed by telephone for the Behavioral Risk Factor Surveillance System (BRFSS) in 1993.

The researchers found that there were 655 episodes of alcohol-impaired driving for each 1,000 adults. Alcohol-impaired driving was most frequent among men aged 21 to 34 years (1,739 episodes per 1,000 adults) and was nearly as frequent among men aged 18 years to 20 years (1,623 episodes per 1,000 adults), despite legislation in all states that prohibits the sale of alcohol to persons younger than 21.

The authors believe their results provide a conservative estimate of the prevalence of alcohol-impaired drivers because of the social stigma attached to reporting drinking and driving; incorrectly assessing whether they were impaired; and not including data from drivers younger than age 18, a group that has a high prevalence of alcohol-impaired driving.

The researchers write: "... We believe that BRFSS data on alcohol-impaired driving are useful for estimating the magnitude of the problem, monitoring temporal trends, developing programs and policies, and evaluating the effectiveness of interventions to prevent alcohol-impaired driving."

Aggressive Intervention Key to Preventing Drunk-Driving

Concerning possible interventions, the authors write: "Effective policies include prompt license suspension for persons arrested for driving while impaired and lowering the legal blood alcohol level to, at most, 0.08 grams/deciliter for adults and 0.02 grams/deciliter for drivers younger than 21 years of age. Since alcohol-impaired driving still occurs frequently among persons from 18 to 20 years of age we also recommend strict enforcement of minimum drinking age laws and the passage of 'zero tolerance' laws, which lower the legal alcohol concentration for drivers younger than 21 years of age.

"We also strongly encourage clinicians to be involved in the prevention of alcohol-impaired driving. In addition to supporting public policies, clinicians can screen patients for alcohol problems; obtain blood alcohol concentrations on injured patients; and provide patients with brief interventions, refer them for specialized treatment, or both, depending on the severity of their drinking problem."

They conclude: "Through this combination of legal and medical interventions, we can further reduce the unacceptable burden of injury and death from alcohol-related motor vehicle crashes and facilitate the early diagnosis and treatment of alcoholism."

According to the authors, injuries resulting from motor vehicle crashes are a leading cause of death in the U.S. among people one to 34 years old, and approximately 41 percent of the 40,676 traffic fatalities in 1994 were related to alcohol. Two of five people in the U.S. will be involved in an alcohol-related motor vehicle crash at some time during their lives.

Science News Press Releases for the week of January 8, 1997

It is well known that younger drivers are over-represented in driving fatalities due to drunk driving. My research focuses on changing older adolescent behavior with respect to drunk driving. Adolescents represent an important target group for several reasons. First, there is evidence indicating that the leading cause of death among young Americans is alcohol-related traffic accidents. Second, adolescents represent new drivers who are just embarking on a life behind the wheel of an automobile. The establishment of safe practices and orientations vis-a-vis drunk driving at this time is critical. Although it is the case that high school aged adolescents are under the legal age for alcohol consumption, estimates suggest that between 70% to 90% of all senior high school students experiment with alcohol. Thus, the reduction of drunk driving among this demographic group seems important.

My previous NIAAA funded research (with colleagues from the University at Albany, SUNY) has identified empirically the kinds of information that needs to be conveyed to teenagers in order to reduce drunk driving. Such information potentially could be conveyed to the teen by schools, peers (SADD), the media, parents . A review of school based treatments of drunk driving indicates that such treatments are limited in scope. Education about drunk driving typically occurs in mandatory health classes in which there is tremendous competition between topics (e.g., nutrition, alcohol consumption, drugs, sex) in terms of class coverage and class time. Drunk driving is typically addressed only superficially and in the context of more general lectures on alcohol. It seems unlikely that schools will devote large amounts of class time to a specialized topic such as drunk driving. Without special efforts on the part of schools to incorporate the kinds of educational materials that our previous research suggests is most effective, it is evident that other sources of information need to be developed.

One of the lines research I have been conducting will develop educational materials for parents of adolescents. It will teach parents how to develop good communication patterns with their teenager. It will teach them how to initiate communication with their adolescent about drunk driving, even when the family history is one of minimal parent-adolescent communication. The materials will teach parents what information will be most effective in convincing their teenager not to drive drunk and will teach parents the most effective ways of presenting this information to their teen. We will then examine the impact of this intervention on adolescent drunk driving behavior. There are several advantages to this approach. First, it will have the general effect of improving communication patterns between parents and teens. Second, it will permit parents to make value judgments about the kinds of information that their teen should be given. For example, most research on determinants of drunk driving focuses on the act of drunk driving per se (e.g., the increased risk of getting in a serious accident). Our research suggests that an important set of variables that impinge on drunk driving is how an individual construes alternative courses of action to driving drunk as well. When faced with a situation where he or she has consumed too much alcohol, an individual can drive drunk or pursue some other course of action (e.g., call a taxi, stay overnight, ask a friend for a ride home). If none of these alternatives appear viable or desirable, the individual is more likely to drive drunk, everything else being equal. It is possible to educate adolescents about what alternatives to driving drunk might exist and how to most effectively pursue these alternatives. However, our discussions with school administrators has indicated a reluctance to incorporate such information into school based programs. The primary objection is that by providing effective alternatives, one might be unwittingly encouraging adolescents to drink alcohol. This viewpoint holds that the risk to one's life by driving drunk is a deterrent to drinking alcohol and that by removing this deterrent, it is more likely that the teenager will drink alcohol, which is both illegal and undesirable. Administrators fear the controversy that might ensue from parents of students if such an approach is taken. With a parent based education approach, parents can be appraised of the potential relevance of alternatives to driving drunk and then make their own decisions about whether to address this issue and the kinds of alternatives that are acceptable to them.

The traditional stereotype among many lay persons and social scientists alike is that adolescence is a time when parents lose their influence on their children and that adolescent behavior is primarily a function of peer influences. This viewpoint is being increasingly challenged across a wide range of research domains. In addition to my research program on drunk driving, my colleagues at the University at Albany, Drs. James Jaccard and Patricia Dittus, have been actively studying parental influences on teenagers in the context of premarital sex and unintended pregnancy. Their research efforts have clearly shown that characterizations of minimal parental influence are based on data that are conceptually weak and methodologically suspect and that when approached from more compelling theoretical frameworks, parental influence on teen behavior can be substantial.

There is a growing body of social science literature on parent education programs in general and their effectiveness in influencing parental behavior. Much of this research is summarized in the recent Handbook on Parent Education. The forms of parent based interventions are varied, including school based programs, parenting conferences, written brochures on effective parenting, video-based programs of parenting, and parent teacher interactions, to name a few. Programs have been aimed at influencing such diverse child behaviors as school performance, sexual behavior, health behaviors, and physical development, to name only a few. It is evident from this literature that parenting education programs can be effective, but that they are not always so. I hope to contribute to this general body of knowledge by developing an approach to designing parent education programs aimed at changing specific adolescent problem behaviors. To the extent that we can show our approach produces tangible results in an area such as drunk driving, then this will encourage researchers to use the approach in other research domains to determine if it can form the skeleton for programs in other domains.

Only a few published accounts of the use of parent education program as a means of influencing drunk driving behavior in adolescence have been published in the scientific literature. Atkin reports a parent intervention program that led to increased concern on the part of parents for teen drunk driving and which increased communication between parents and teens about this topic. However, the program did not show evidence of effects on teen drunk driving behavior. McPherson developed a program to increase support networks for parents to discuss alcohol issues with their teens and to convey information about drunk driving and alcohol consumption. The results showed that parents tended to become more assertive about talking to teens and were more likely to monitor their teen's behavior with regard to drunk driving. These studies suggest that parent education programs can be effective in altering parental behavior, but there is little evidence that these effects filter through to the drunk driving behavior of adolescents. The focusof my research is distinct from previous parenting interventions in several ways. First, I have conducted extensive empirical research on our target adolescents focusing on cognitive, attitudinal, and personality variables that are likely to influence teen drunk driving. I have applied (and modified) a well developed theoretical framework based on over 15 years of decision theoretic work by Jaccard to the empirical analyses. This research has provided a list of variables that, if changed, are likely to impact on teen drunk driving behavior. It is these variables that will be the focus parental education efforts. Thus, the content of the research has a strong theoretical and empirical base that is directly tied to determinants of drunk driving of the adolescent target population. By contrast, past intervention efforts have not had this kind of empirical and theoretical base. Second, the educational materials will carefully take into account issues of adolescent development in the context of social, emotional, cognitive, moral, and physical development. Parents will be educated about adolescent development in each of these domains and given specific behavioral strategies for educating their teens about drunk driving in the context of basic adolescent development issues.

Although there are only a few studies focused on parent interventions and adolescent drunk driving, there are numerous studies that have used correlational paradigms to study the relationship between parental behaviors and teen drunk driving. For example, Beck observed that parents are more likely to attribute deviant behavior to friends of their children rather than their children themselves and that parents generally are not aware of the full extent of their teens' drinking habits and practices. Most parents admitted that they never talk to other parents about teen drinking and driving. Beck and Lockhart reviewed factors that can influence parent effectiveness in attempts to control adolescent drunk driving and present a theoretical framework for analyzing parental effectiveness. According to these authors, barriers that diminish the impact of parents include perceptions of low levels of empowerment and control, disaffiliation and lack of skills to communicate with their children, low levels of awareness, a lack of social support from other parents, and an increasing psychological distance from their children as they grow older. Beck and Lockhart review research from other research domains that suggest the importance of these variables. Beck, Summons, and Matthews report the results of focus groups with parents aimed at understanding issues related to adolescent alcohol consumption and drunk driving. They found that parents tend to be unaware of the extent of teen drinking, that many parents feel powerless to affect their teen's drinking behavior, that many parents feel a sense of isolation from other parents dealing with similar problems, and that parents are uncommitted to devoting large amounts of time to the problem in the context of formal workshops. DiBlaso applied social learning theory to the analysis of adolescent drunk driving behavior, examining the relationship between peer variables, parental variables, and self reports of drunk driving. He found support for a statistically significant association between numerous parental variables (e.g., disapproval of drunk driving and alcohol consumption, parental discipline strategies) and teen behavior. Jessor analyzed risky driving behavior in adolescents and found that such behavior was significantly related to parent-friend compatibility and the number of parental models for health reinforcing behavior. Klepp and Perry applied Problem Behavior Theory to the analysis of adolescent drunk driving and observed little utility of parent based variables in predicting drunk driving behavior. These studies, as well as others not reviewed here, generally point to the potential relevance of parents in influencing adolescent drunk driving behavior. Although there are some negative findings and evidence to suggest that parent communications with their teens are not frequent enough or satisfactory in quality, there does seem to be sufficient evidence to indicate that what a parent does and the type of relationship that a parent has with his or her teen can and does impact on drunk driving behavior.

As noted, parent intervention programs are relatively rare in the drunk driving domain. However, there is a much more substantial literature on the impact of parents and parent-based interventions focused on adolescent alcohol consumption, adolescent drug use, and adolescent sexual behavior. There is also a substantial body of literature on family systems approaches to the analysis of these behaviors. Space constraints do not permit a review of these literatures here, although overall, they affirm the promise of parent based education efforts.

In sum, there exists sufficient empirical data both in the area of drunk driving and related areas of adolescent problem behaviors to suggest that parents can play an important role in influencing drunk driving behavior. Based on data that I have collected, I believe that parental impact will be even greater if parent-teen communication can be encouraged and directed at the appropriate target variables identified by our empirical and theoretical analyses. The proposed research is significant in that it will be an important addition to the almost non-existent literature on parent interventions aimed at reducing adolescent drunk driving. It has the features of using a strong theoretical base, a strong empirical base that has already been collected and evaluated on the target populations, and it will present information taking into consideration developmental theory on adolescence.

Last Revised: 10/10/95