Underage drinking is a serious problem for our society. From reports in the media, one gets the impression that it is getting worse ever year and that even casual teenage drinking carries with it devastating implications for our youth, including increasing the alcoholism rate of those who drink early and even death. Do the statistics support these stories?
Can you honestly say that you never drank alcohol during high school? If so, you are in a distinct minority: nearly 80% of high school seniors admit to at least trying a drink and about 50 percent report having consumed alcohol during the last month. While these numbers are down from their peaks in the late 70s and early 80s, a large majority of young people still drink regularly long before they reach the legal age of 21.
Clearly, efforts to eliminate underage drinking have failed. It is also plain that many successful Americans have a history of high school and college alcohol use - sometimes involving of periods of excess. If youthful drinking were a genuine bar to political office, higher education or career advancement, government, academia and industry would grind to a halt.
Nonetheless, the media and some activist groups, government agencies and foundations constantly promote the idea that most teen drinking is extremely risky and carries a high probability of causing alcoholism or death. For example, an August 2004 press release from the Office of Juvenile Justice and Delinquency Prevention and the National Liquor Law Enforcement Association screams, "The carnage caused by underage drinking in America is unrelenting," and notes that alcohol kills "6.5 more kids than all other drugs combined."
Similarly, a recent CNN medical segment ["Teenage binge drinkers are more likely to become adult alcoholics" 6/16/04"] reported that a study had found that teen binge drinking was linked to an increased risk for alcoholism. CNN mentioned that 70% of teens binge drink - but did not note that only 3.81% of adults are alcoholics. The channel's physician-reporter Sanjay Gupta ended his report with a conclusion not backed by the data he cited, saying, "So really, the message to parents out of this is even a small drink, even only a sip early on in life could be a problem later on."
Headlines like the San Francisco Chronicle's "Teenage Drinking a U.S. Epidemic" [2/26/02] and the New York Post's "Boozed-Up Teens In Dangerous Liaisons," [2/7/02] shout along with the advocates' PR campaigns, often including hyped-up quotes from people like the National Center on Addiction and Substance Abuse (CASA)'s head Joe Califano, such as "Alcohol is the fatal attraction for many teens."
As a result, politicians call for ever stricter legislation; activists say that alcohol advertising is targeting teens and should be banned; and schools enact "zero tolerance" policies that expel kids for doing once what the vast majority of American adults did many times at their age.
Do these efforts make sense? Are we using our resources wisely by trying to prevent all teens from taking a sip till 21-or could we better spend our money by focusing on reducing the genuine harm that can occur when some young people drink to excess?
Costs and Consequences
According to a 1999 study conducted by the Pacific Institute on Research and Evaluation (PIRE), about 3,500 deaths per year are caused by drinkers under the age of 21. Homicides account for the greatest number of these deaths (roughly 1,600 per year), followed by drunk driving (1,400), and teen suicides (260). The remaining 300-odd deaths are caused by other accidents, drownings, burns, and alcohol overdoses.
Interestingly, the proportion of deaths judged to be alcohol-related was lower than one might expect from the hysterical coverage teen drinking is usually given in the media. For example, just under 20% of car accident deaths caused by drivers 21 and younger are related to intoxication-and this proportion has been dropping for the last 20 years.
The PIRE researchers also estimated that 21% of murders committed by youth were caused by alcohol. This was based on research showing that 42% of killers report being intoxicated at the time of their crime and on other studies which found that while half of alcohol-linked killings would probably have occurred even if alcohol hadn't been consumed by the perpetrator, the other half would not have taken place if the killer had been sober. The figures for suicide were even lower, with just 12% of male suicides and eight percent of those committed by females attributable to alcohol.
But while the PIRE analysis looked at deaths caused by intoxicated youth, it didn't separate out which of those deaths occurred among other young people. Another analysis (1) tried to calculate this number, seeking to determine the number of deaths of youth aged 10-19 related to alcohol.
Using 1995 figures, it found a very similar number to that compiled by PIRE: 3300 deaths of young people could be attributed to alcohol-related incidents including homicides, suicides, car accidents and alcohol overdoses .
From this report and the census population estimate of teens aged 10-19 in 1995, a very rough estimate of the risk of alcohol-related death in this age group can be calculated. The odds of a teen dying an alcohol-related death come to approximately .00009 per year or 9 in 100,000 (2) . If one calculates the figure only among those who admit to drinking in the last month, the figure is still less than 2/100ths of one percent.
This still represents a significant proportion of the untimely deaths among youth: the total number of deaths among those aged 10-19 in 1995 was 14,600, meaning that 23% of adolescent deaths that year can be blamed on alcohol. And while the risk of death is relatively low, the risk of injury is not. The PIRE study estimated that about 1 million non-fatal injuries were caused by underage drunk drivers and 1 million assaults were committed by youth under the influence of alcohol. About 5,100 injuries were related to drunken suicide attempts and 40,000 to the results of alcohol poisoning. These estimates relied on statistics from the CDC's National Hospital Ambulatory Medical Care Survey.
The PIRE study estimated the costs related to underage drinking at $52.8 billion per year (3). The vast majority of these costs - $38.5 billion - were related to loss of quality of life due to deaths and serious injuries from car accidents and violent assaults. These figures are calculated by determining what people are willing to pay to reduce their risk of incurring such harms - they do not derive from pay-outs in lawsuits or actual government expenditures. Actual lost income and other non-health related costs related to youth drinking added up to $10.6 billion, while medical costs amounted to $3.6 billion.
There are two major conflicting schools of thought in alcohol policy. One focuses primarily on reducing alcohol consumption rates throughout the population: I will refer to them here as the "dry" group, using the terminology of Robin Room, a top alcohol expert and the Director of the Center for Social Research on Alcohol and Drugs at Stockholm University. This group includes most American advocates like the National Center on Addiction and Substance Abuse and the Center on Alcohol Marketing and Youth (CAMY). The "dry" group pushes for measures like high alcohol taxes, restrictions on advertising and selling times and high drinking ages. They tend to favor most measures that restrict access to alcohol-up to and including absolute prohibition in some cases (though most do not believe this is actually feasible).
The second group, the "wets" claim that these kinds of "prohibitionist" measures actually increase damage by, for example, driving young drinkers underground where they are more likely to drink dangerously. They support efforts like lowering the drinking age in order to allow young people to learn to drink moderately. They believe such measures would encourage youth to seek help (rather than avoid doing so for fear of being caught) if they do get into trouble. "Wets" include groups like the Drug Policy Alliance and experts like author Stanton Peele.
There is some evidence to support both perspectives: Those who believe in lowering overall consumption cite studies linking higher alcohol taxes and a higher drinking age with fewer drunk driving deaths, lower crime rates (4) and even lower rates of sexually-transmitted disease.
Those who believe these measures increase rather than reduce harm point to studies showing that youth drinking was declining (5) long before the U.S. adopted a national drinking age of 21 in 1987. They also say that the states that waited longer to raise their drinking ages did not have more drinking by young people than those that did it sooner (6).
The "wets" claim that a uniform drinking age is responsible for the reduction in drunken driving seen after the U.S. adopted a national drinking age of 21, not the older age level itself. When drinking ages vary between states, youth often travel to across state lines to the state with the lower drinking age to get drunk-leading to an increased likelihood of drunk driving and to longer trips made by intoxicated drivers. From this perspective, the reduction in crash deaths seen after America adopted a national drinking age of 21 was caused by the elimination of cross-border traffic, where drinking ages varied, and not by fewer young people drinking and driving.
The "wets" also note that at the same time as the drinking age was raised to 21, national media campaigns simultaneously stigmatized drunk driving and pushed the idea of a "designated driver"-and that this, too, could have helped drive down alcohol-related crashes while leaving youth alcohol consumption rates untouched. In fact, this group claims that when underage drinking is illegal, it becomes more attractive to youth as "forbidden fruit" (7). The need to hide drinking may also encourage youth to binge in order to reduce their chances of being caught in possession of alcohol
There is almost certainly some truth in both positions: a high drinking age may deter some people who otherwise would have drunk heavily from doing so, while pushing others to do so in a more dangerous manner.Either way, the fact that the vast majority of youth are drinking regularly long before it is legal to do so suggests that current policy is a failure at its intended goal of keeping alcohol out of the hands of most young people.
Zero Tolerance Policies
Throughout the 90s and the early 00s, fears about alcohol and teens were continually stoked by the media and by organizations with a stake in pushing the notion that drinking is a great danger to youth.
For example, there was the study - Teen Tipplers: America's Underage Drinking Epidemic - released by the Center on Addiction and Substance Abuse (CASA) in 2002 claiming that people under 21 drunk 25% of all the alcohol consumed in the United States. This prompted dozens of headlines in newspapers across the country about a teen alcohol "epidemic" that year. (Unfortunately, CASA failed to take into account that the government deliberately over-sampled teens, and as a result its analysis misrepresented their proportion in the population.).
There were also numerous reports claiming that young people were increasingly engaged in binge drinking, many stemming from research by Henry Weschler at the Harvard School of Public Health. For example, the Washington Post headlined one article, citing the researcher, "Drinking Lessons; As Alcohol Problems Grow, Colleges Seek New Remedies" [4/16/02]. But these accounts rely on such a liberal definition of binging (which Wechsler himself promotes) that it includes those with blood alcohol levels (BACs) well below the legal limits for driving.
Rather than seeing a binge as being a spree of drinking lasting a day or more, as many people would colloquially define it, alcohol researchers have now classified a binge as being a situation in which someone takes "5 or more drinks on one occasion." This means that a college student who has a drink an hour during a long party-and never reaches intoxication-is a "binge drinker." Not surprisingly, this has lead to a majority of high school seniors and college kids being defined as bingers.
Given such studies and the increasing calls for action by activists and concerned parents, educators and policy makers decided that something had to be done. It was already illegal for teens to drink, so high schools began setting "zero tolerance" policies. These took off in the wake of the Columbine high school shootings, and targeted alcohol and drugs along with weapons, even though none of the school shootings were linked to alcohol or recreational drugs.
These policies meant that if a high schooler was caught-sometimes even off-campus-with one drink on one occasion, he or she would be permanently expelled. For example, in 1998, four teens from a Colorado school who had never been in trouble before, were expelled from their high school for having admitted to drinking in a motel room on one occasion. [Monument County Tribune, 10/1/98]. In another case, a California teen became suicidal after being expelled for drinking off-campus one day during his lunch hour (8). He never completed high school. In some cases, the mandatory punishment is less severe: the teen is sent to treatment instead.
While the legal drinking age is a matter of debate amongst alcohol policy experts, even the most conservative alcohol policy groups like CASA do not support extreme zero tolerance, recognizing that expulsion from school does more damage to a kid's future than most drinking incidents do. Not only does expulsion give kids more time to drink (and a reason to drown their sorrows), research suggests that higher education itself reduces the odds of long-term alcoholism (9).
Even mandated treatment poses problems if it is used indiscriminately. While teens with genuine drinking problems may benefit, those who were just unlucky enough to get caught experimenting can be harmed by being grouped with kids with more severe problems (10). Forcing kids to admit that they are alcoholics in order to successfully complete treatment-as most teen treatment programs do-may also be dangerous. Telling teens, who are often confused about their identities to begin with, that they have the lifelong disease of alcoholism (which they are told, carries a 90% chance of relapse) can be a self-fulfilling prophecy.
Almost all experts now agree-from CASA to the ABA to the ACLU- that zero tolerance policies have more risks than benefits (though CASA supports tough policies if kids are given treatment, not expulsion). Drinking during the school day is certainly unacceptable; but not all underage drinkers are drinking for the same reasons: One kid may be an alcoholic, while another may have parents going through a nasty divorce, and yet another is a shy straight-A student trying to be more sociable. A school's response should take these circumstances into account and not sacrifice one child's future in hopes of deterring others.
There's just no evidence that supports a deterrent effect for these policies and a great deal which suggests that they can do harm. For example, researchers at Harvard's Civil Rights project found that high rates of suspension from school (which are linked with zero tolerance because it mandates such punishment for first offenses) are linked with high rates of juvenile incarceration. Previous studies found no evidence that zero tolerance policies improved school performance or atmosphere (11).
So what does the research support in terms of policies and programs that reduce youth drinking and the harms related to it?
Some experts like Stanton Peele, author of the classic text, "The Meaning of Addiction" point to wine-drinking cultures like those in Italy, France and Spain for hints on how to handle alcohol policy better. In these cultures, alcohol is essentially a food-it is part of a meal and children learn to drink with their parents from a very early age. Drunkenness is not tolerated. Though rates of daily drinking are high in these cultures, rates of binging and other alcohol problems have, historically, been low. Alcohol itself is viewed as a neutral substance; it is drunkenness and not drinking that is seen as the problem.
In contrast, northern drinking cultures see alcohol as almost a mystical substance-with great power for good or ill. Drinking is an activity in and of itself; alcohol is consumed in bars not at family meals, and it is often positively associated with masculinity. In such settings, drunkenness is frequently the goal of drinking. Northern drinking cultures typically have lower rates of daily drinking but higher rates of binging, alcohol-related problems and complete abstinence. The U.S. has generally followed the northern pattern (12).
Lately, however, drinking-to-get-drunk behavior has become popular in southern Europe (possibly as a result of large influx of young drink-oriented tourists from northern countries), proving that these cultures are not totally immune from alcohol-related problems (13).
Nevertheless, the southern, demystified view of alcohol as part of a meal may be worth adopting. Viewing alcohol as a "demon" allows those who drink to excess to avoid responsibility for their behavior and place the blame on the substance. By contrast, seeing it as something most people manage responsibly re-enforces self-control.
Further, teens raised to drink responsibly at the table are less likely to see alcohol as a badge of adulthood and rebellion. Allowing them to learn their limits in a safe space may also help ward off risky experiments in more dangerous settings later. As a result, permitting teens to drink with their parents at home and in restaurants at any age with adult supervision may make sense.
As for purchasing alcohol, Canada has adopted a legal age of 19, making alcohol illegal for high school students but allowing virtually all college kids to legally drink. This means that at least some of their drinking is supervised and that there is no incentive to binge to avoid being caught possessing alcohol.
Not surprisingly, the lower legal drinking age means that more Canadian students drink than American students; however, those who imbibe display more moderation than their U.S. counterparts. For example, 54% of American students who drank in the last week drank heavily, while only 42% of Canadian students in that group did likewise (14). And despite Canada's lower drinking age, drunk driving fatality rates are virtually the same as in the U.S. (in 1999, for example, both were 40%) (15).
Dealing with Excess
Though taking a pragmatic view of alcohol consumption can help reduce its use as a symbol of teen rebellion, it will not eliminate alcohol abuse or alcoholism. Those who support keeping the drinking age high cite research finding that the later one starts drinking, the lower the risk of developing dependence. For example, one large study found that those who reported starting drinking at 14 or younger are four times more likely to become alcoholics than those who said they started at 20 or older (16).
Such studies define the start of drinking as the age when a person begins drinking regularly-as opposed to having just a sip or two. They do not look at the context of early drinking-so they don't offer evidence on whether drinking wine with the family at dinner would increase risk in the same way as drinking with 12-year-old friends in the schoolyard might.
The studies also cannot determine causality: there is evidence that the youngest heavy drinkers are more antisocial, more likely to have a family history of alcoholism and more troubled generally than their peers (17), so it may well be that these things increase both the odds of early drinking and dependence. Rat studies do suggest that drinking at a very young age may alter brain development, however, so the causal picture is quite complex.
Whatever the case, raising the drinking age has not reduced the number of youth who drink at the earliest ages: in fact, the percentage of youth in eighth grade and lower who report drinking jumped dramatically just after the drinking age was raised to 21 (from 27% before 1984 to 36% by 1993) before returning to its previous level. Measures that show promise in reducing youthful drinking and its consequences include:
Identifying college students who are drinking heavily and teaching them about the effects of alcohol in a non-judgmental setting. G. Alan Marlatt of the University of Washington has found simply educating randomly selected young heavy drinkers dramatically reduced their alcohol-related problems compared to a control group of student bingers who were not given his classes. The results were still significant four years later (18).
It is critical that these classes be framed as giving students options for managing drinking-if students believe that the only choice available is quitting entirely, they simply won't turn up. Furthermore, if the classes don't recognize the pleasure young people seek from drinking and focus entirely on potential negative consequences, the material will often just be dismissed.
Teaching young people to understand what researchers call the "biphasic" effect of alcohol can be particularly helpful. Alcohol initially produces energy and excitement, but later causes tiredness and fatigue. Drinking more doesn't reproduce the early effects-but many youth don't understand this. When they do, moderation is more likely and achievable.
Helping students recognize that their peers are not drinking as heavily as they may believe they are.The "social norms" approach to reducing youthful drinking involves teaching people that "everybody really isn't doing it," and that most of those who are, are not indulging in the excessive amounts their peers may think they are.
Research finds that people are highly influenced by what they believe others are doing, and that they also tend to take more notice of extreme, flamboyant behavior. In combination, these factors produce teens who see a few peers vomiting in trash cans on weekends and decide that most of the campus must be getting rip-roaring drunk.
Studies have found that when youth are educated about these effects and about how much their peers are actually drinking, heavy alcohol use can be reduced by between 25-44% (19) (Note: while one recent study called this approach into question, it had serious methodological problems and wasn't actually designed to evaluate social norms).
Media campaigns that offer practical solutions. The "designated driver" ad campaigns of the 1970s and 80s were some of the most effective public health ads ever broadcast. After they were introduced, the idea caught on with drinkers, while also creating a valued role for people who wanted to socialize with drinkers but did not want to drink themselves. Other ads and articles worked to make drunk driving socially unacceptable-while the designated driver offered a solution for those who wished to drink. The proportion of driving deaths attributable to drinking fell from 60% in 1982 to 40 % in 1999 .
Interestingly, ad researchers found that while commercials showing the devastating effects of drunk driving accidents had a large emotional impact and won industry awards, humorous ads teaching women how to get the car keys away from their intoxicated boyfriends were better at changing people's behavior. Media campaigns are far more effective when they offer people actions to take and not just avoid (20).
Conclusion: A Question of Values
How to deal with youth drinking is a question that is heavily value-laden in American society. The risks associated with drinking by young people are not necessarily greater than those associated with "worthwhile" activities like sports-but we do not often offer teens the equivalent of football helmets when they participate in activities that we tend to morally condemn.
This makes evaluating the real risks extremely difficult. When people consider teen drinking, the default assumption is that it is harmful, valueless, and that anyone who would consider allowing it must have bad intentions. This often precludes rational discussion of certain options.
For example, some suburban parents secretly hold parties where they allow teens to drink, and they take their car keys and allow them to stay overnight so that driving cannot occur. In many states, this is illegal. But does it really encourage excess and condone teen drinking-or does parental supervision by its very nature reduce the odds of over-indulgence?
There have been cases where teens have died for lack of medical attention when those around them didn't seek help because they feared educational or parental consequences. In 2001, 17-year-old high school football star passed out after being punched at an unsupervised, illicit drinking party in an upscale community in Westchester, New York. He was not taken to the hospital and died, possibly as a result of the delay in care. The other teens present said they didn't get help because being caught drinking by parents or school officials would have had devastating consequences.
Yet because parentally-supervised parties are driven underground, there is little research on this subject. One study, published in the Journal of Adolescent Health in 2004 (21) did find that teens whose parents held parties where drinking was allowed were twice as likely to binge drink as those whose parents did not. They were also twice as likely to have had at least one drink in the month before they were surveyed.
But the same study also found that adolescents who drank at home with their parents were half as likely to have had a drink in the last month and only one-third as likely to binge - drink as those who had not done so. And the study didn't determine whether the kids who were at parties supervised by parents had lower risks of drunk driving accidents, injuries or alcohol poisoning.
Without more research, it's impossible to figure out what this paradoxical data says about parental supervision of drinking by adolescents. And if more research were done, the measures for success would probably be disputed. Some would see failure if teen drinking were to increase due to parental license, while others would see success in fewer deaths, school dropouts and accidents.
After the Westchester case, the community soon enacted even tougher policies. Yet the media coverage never questioned whether tougher policies may have been part of the problem to begin with.
Reporters who cover teen drinking often take the values hidden in the discussion of risks related to teen alcohol use for granted. But in order to develop effective policies for dealing with underage alcohol abuse, these assumptions need to be exposed, questioned, and challenged so that parents and politicians can explore a whole range of solutions and make the best choices for their kids.
1. Health and the American Child Part 1: A Focus on Mortality Among Children - Risks, Trends, and Priorities for the Twenty-First Century, May 1999: www.phpab.org (link to doc itself is broken)
2.The actual odds will increase per year throughout adolescence, as older teens are at greater risk both for drinking and for drinking-related deaths-but even at their peak, these still remain a small fraction of one percent.
3.The abstract says 58.4 billion but the chart adds up to 52.8, and that's the number cited in a later version by the Department of Justice.
4. Smith, DI and PW Burvill, "Effect on Juvenile Crime of Lowering the Drinking Age in Three Australian States," British Journal of Addiction, 83:181-188, 1987.
5. see especially p. 168
6. Hingson, Ralph, Merrigan, Daniel, and Heeren, Timothy. Effects of Massachusetts raising its legal drinking age from 18 to 20 on deaths from teenage homicide, suicide and nontraffic accidents. Pediatric Clinics of North America, 1985, 32, 221-233.
7. Engs, Ruth C., and Hanson, David J. Reactance theory: A test with collegiate drinking. Psychological Reports, 1989, 64, 1083-1086.
8. Personal conversation with anonymous Tuston High father, Maia Szalavitz.
9. Dawson DA, Grant BF, Chou SP, Pickering RP Subgroup variation in U.S. drinking patterns: results of the 1992 national longitudinal alcohol epidemiologic study. J Subst Abuse. 1995; 7(3):331-44.
10.Dishion, TJ; McCord, J.; and Poulin, F. When interventions harm: Peer groups and problem behavior. American Psychologist 54(9):755-764, 1999.
11.Skiba, R. J., & Knesting, K. (2001). Zero tolerance, zero evidence: An analysis of school disciplinary practice. In R.J. Skiba & G.G. Noam (Eds.), New directions for youth development (no. 92: Zero tolerance: Can suspension and expulsion keep schools safe?) (pp. 17-43). San Francisco: Jossey-Bass./p>
12.Personal conversation, Robin Room, Director, Center for Social Research on Alcohol and Drugs, Stockholm University. List of his publications here. See also: Levine, H. G. Temperance Cultures: Alcohol as a Problem in Nordic and English-Speaking Cultures. In: Lader, M., et al. The Nature of Alcohol and Drug-Related Problems. New York: Oxford University Press, 1992 pp. 16-36.
13. See also Liat Radcliffe and Adam Piore, "Is Europe Drinking Too Much? For a new generation of young people-especially teenagers-bingeing is all the rage" Newsweek International, 11/3/03 .
14. Kuo M, Adlaf EM, Lee H, Gliksman L, Demers A, Wechsler H., Addiction. 2002 Dec;97(12):1583-92. "More Canadian students drink but American students drink more: comparing college alcohol use in two countries."
15. See Mothers Against Drunk Driving http:www.madd.ca/library/madd_alcohol_facts.pdf and http://www.madd.org/stats/0,1056,1298,00.html
16. Grant BF Dawson DA Age of onset of alcohol use and its association with DSM-IV alcohol abuse and dependence: Results from the National Longitudinal Alcohol Epidemiologic Survey J Substance Abuse 9:103-110, 1997.
17. Chassin L, Pitts SC, Prost J Binge drinking trajectories from adolescence to emerging adulthood in a high-risk sample: predictors and substance abuse outcomes.J Consult Clin Psychol. 2002 Feb;70(1):67-78.
18. Baer JS, Kivlahan DR, Blume AW, McKnight P, Marlatt GA. Brief intervention for heavy-drinking college students: 4-year follow-up and natural history. Am J Public Health. 2001 Aug; 91(8):1310-6.
19. Haines, M. and G. Barker. "The NIU Experiment: A Case Study of the Social Norms Approach," (2003) in The Social Norms Approach To Preventing School And College Age Substance Abuse: A Handbook For Educators, Counselors, And Clinicians, Ed. H. Wesley Perkins. San Francisco: Jossey-Bass.
20.E. Maibach RL Parrott, eds. Designing Health Messages: Approaches from Communication Theory and Public Health Practice. Sage Publications: 1995.
21.Foley KL, Altman D, Durant RH, Wolfson M. Adults' approval and adolescents' alcohol use. J Adolesc Health. 2004 Oct;35(4):345.e17-26.