The Minimum Legal Drinking Age

History, Effectiveness, and Ongoing Debate

TRACI L. TOOMEY, PH.D., CAROLYN ROSENFELD, M.P.H.,

AND ALEXANDER C. WAGENAAR, PH.D.

Minimum legal drinking age (MLDA) laws provide an example of how scientific research can support effective public policies. Between 1760 and 1765, 27 States lowered their MLDAs8 subsequently, scientists found that traffic crashes increased significantly among teenagers. Alcohol use among youth is related to many problems, including traffic crash es, d rowning s, vandalism , assaults , homicid es, suicides, teen age pregnancies, and sexually transmitted diseases. Research has demonstrated the effectiveness of a higher MLDA in preventing injuries and deaths among youth. Despite laws prohibiting the sale or provision of alcohol to people under age 21, minors can easily obtain alcohol from many sources. Encreased MLDA enforcement levels and deterrents for adults who might sell or provide alcohol to minors can help prevent

EY WORDS additional injuries and deaths among youth. K : minimum drinking age laws; evaluation; history of AOD public policy; public policy on AOD; adolescent; law enforcement; AOD availability; AOD sales; AODR (alcohol and other drug related) injury prevention; AODR mortality; traffic accident; drinking and driving

shows how science has influenced were being lowered (Wechsler and cience can play a critical role in MLDA policies in the past and sum­ Sands 1980). Scientists began study­ developing effective policies to marizes research contributing to the ing the effects of the lowered MLDA, address health issues, including ongoing debate on the MLDA. those focused on alcohol­related prob­ particularly focusing on traffic crashes, lems (Gordis 1991). In an ideal world, S public policy development would be TRACI L. TOOMEY, PH.D., is a research HISTORY OF THE MLDA based on the identification of a prob­ associate with, CAROLYN ROSENFELD,

lem and the scientific evidence of the Following the repeal of Prohibition, M.P.H., is a recent graduate of, and

factors that are most effective in reduc­ ALEXANDER C. WAGENAAR, PH.D., is nearly all State laws restricting youth ing that problem. In the real world, access to alcohol designated 21 as the a professor in the Division of Epi­ demiology, University of Minnesota, however, public policy results from minimum age for purchasing and con­ economic and political forces, which suming alcohol (Mosher 1980). Be­ Minneapolis, Minnesota.

occasionally combine with good sci­ tween 1970 and 1975, however, 29 This research was funded in part by ence. Minimum legal drinking age States lowered the MLDA to age 18, National Institute on Alcohol Abuse (MLDA) laws provide an example of 19, or 20. These changes occurred at and Alcoholism grants R01AA10426 how scientific research can support the same time that minimum­age limits and R01AA09142 to Alexander C. effective public policies. This article for other activities, such as voting, also Wagenaar.

VOL. 20, NO. 4, 1996 213

the leading cause of death among teen­ these other alcohol policies have only Opponents of the age­21 MLDA agers. Several studies in the 1970’s recently been evaluated (see Toomey et theorized that even if a higher MLDA showed that traffic crashes increased al. 1994 for a review of the research lit­ reduced alcohol use among minors, significantly among teenagers after the erature). Of all the alcohol control poli­ drinking rates and alcohol­related MLDA was lowered (Cucchiaro et al. cies, MLDA policy has been the most problems would surge among those 1974; Douglass et al. 1974; Wagenaar studied. Since the 1970’s, at least 70 age 21 and older. In other words, op­ 1983, 1993; Whitehead 1977; White­ studies have explicitly examined the ponents believed that a “rubber band” head et al. 1975; Williams et al. 1974). effects of either increases or decreases effect would occur: When youth turned With evidence that lower legal in the MLDA, with some studies using 21, they would drink to “make up for drinking ages were associated with more robust research designs than lost time” and thus drink at higher rates more traffic crashes among youth, citi­ others. MLDA policies may have been than they would had they been allowed zen advocacy groups led a movement evaluated sooner and more often for a to drink alcohol at an earlier age. A to restore the MLDA to 21 in all States variety of reasons, including: (1) a study by O’Malley and Wagenaar (Wolfson 1995). In response, 16 States growing concern about youth drinking (1991), however, refutes this theory. increased their MLDA’s between and driving; (2) availability of archived, Using a national probability sample, September 1976 and January 1983 time­series data on traffic crashes; (3) O’Malley and Wagenaar found that (Wagenaar 1983). Many States, how­ the fact that many States first low­ the lower rates of alcohol use due to a ever, resisted pressure from these ered, then raised, their MLDA’s; and high legal drinking age continued even groups and ignored Government in­ (4) preliminary research showing the after youth turned 21. centives to raise their MLDA’s (King large effects of changes in MLDA’s. Although the MLDA’s effect on 1987). The Federal Government be­ Thorough literature reviews by Wage­ youth alcohol consumption is impor­ came concerned about the safety both naar (1983, 1993), the United States tant, a key consideration is whether of youth in States that had lower General Accounting Office (1987), and the MLDA ultimately affects the rates MLDA’s and of youth who lived in Moskowitz (1989) provide detailed of alcohol­related problems. Alcohol neighboring States. Persons who were summaries of many of these studies. use among youth is related to numer­ below the MLDA in their own State MLDA laws have been evaluated ous problems, including traffic crashes, could drive across State borders to pur­ mostly in terms of how changing the drownings, vandalism, assaults, homi­ chase alcohol in a State with a lower MLDA affects rates of alcohol use cides, suicides, teenage pregnancies, MLDA and then return home, increas­ and traffic crashes among youth. and sexually transmitted diseases. Al­ ing the likelihood of being involved Methods used to study the effect of cohol use is reported in one­fifth to in traffic crashes. the MLDA on alcohol use have varied two­thirds of many of these problems Because the 21st amendment to the widely, contributing to differences in (Howland and Hingson 1988; Plant U.S. Constitution guaranteed States’ conclusions among studies. For exam­ 1990; Roizen 1982; Smith and Kraus rights to regulate alcohol, the Federal ple, some studies used convenience 1988; Strunin and Hingson 1992). As Government could not mandate a uni­ samples, such as students in introduc­ drinking rates increase or decrease, form MLDA of 21. Instead, in 1984 the tory psychology classes, whereas other rates of alcohol­related problems may Federal Government passed the Uni­ studies used sophisticated, random change in response. form Drinking Age Act, which provid­ sampling designs to obtain nationally

ed for a decrease in Federal highway representative samples. Wagenaar Decreases in Traffic Crashes funding to States that did not establish (1993) concluded that studies employ­ Using various research methods, at an MLDA of 21 by 1987 (King 1987). ing strong research and analytical de­ Faced with a loss of funding, the re­ signs typically observed increases in least 50 studies have evaluated the effect of changes in the MLDA on maining States returned their MLDA’s alcohol use among youth following a to age 21 by 1988. lowering of the MLDA. In contrast, traffic crashes (Wagenaar 1993). Some studies assessed policy changes in only when many States raised the MLDA, one State, whereas others analyzed the alcohol use among youth decreased. EFFECTIVENESS OF THE MLDA MLDA’s effect across multiple States. Beer is the alcoholic beverage of These studies evaluated the effect of choice for most youth. As a result, re­ Reductions in Drinking MLDA changes on a variety of out­ duced rates of alcohol use among youth comes, including total traffic crash Following the end of Prohibition, each after the MLDA was increased were fatalities for youth; drinking­driving State developed its own set of poli­ primarily evident in decreased rates of convictions; crashes resulting in in­ cies to regulate the distribution, sale, beer consumption (Berger and Snortum juries; and single­vehicle nighttime and consumption of alcohol. In addi­ 1985). Rates of wine and distilled tion to the MLDA, examples of other spirits use among youth did not change crash fatalities (the crashes most likely alcohol control policies include excise dramatically following the rise in the to involve alcohol). taxes, restrictions on hours and days MLDA (Barsby and Marshall 1977; Most studies on the effect of lower­ of sales, and server training. Many of Smart 1977). ing the MLDA found an increase in

214 ALCOHOL HEALTH & RESEARCH WORLD

MLDA: History, Effectiveness, and Ongoing Debate

traffic crashes and traffic deaths among youth each year (NHTSA 1989; Jones for selling alcohol to a minor. Further youth (Wagenaar 1993). Of the 29 et al. 1992). What is compelling is that research is needed to determine whether studies completed since the early 1980’s the effect of the higher MLDA is oc­ social sources are aware of their legal curring with little or no enforcement. A that evaluated increases in the MLDA, liability for providing alcohol to youth common argument among opponents and whether they perceive a high like­ 20 showed significant decreases in of a higher MLDA is that because lihood of facing penalties for doing so. traffic crashes and crash fatalities. Only many minors still drink and purchase Laws prohibiting the sale and pro­ three clearly found no change in traffic alcohol, an MLDA of 21 does not vision of alcohol to minors are not crashes involving youth. The remain­ work. The evidence shows, however, well enforced (Wagenaar and Wolfson ing six studies had equivocal results. that although many youth still consume 1995), and systems for enforcing the Based on results from research stud­ alcohol, they drink less and experience legislation vary by State. Typically, ies such as these, the National High­ way Traffic Safety Administration fewer alcohol­related injuries and deaths however, enforcement systems use than they did under lower MLDA’s both State administrative agencies, (NHTSA) estimated that in 1987 alone, 1,071 traffic crash fatalities (Wagenaar 1993). A more appropriate usually called State Alcohol Beverage discussion, therefore, is not whether Control (ABC) agencies, and local law were prevented because of the MLDA of 21 (NHTSA 1989). the MLDA should again be lowered enforcement agencies, such as police but whether the current MLDA can be departments and county sheriffs. En­ Since 1984 researchers have been investigating whether changes in the made even more effective. forcement of MLDA laws has focused Despite laws prohibiting the sale or primarily on penalizing underage MLDA also affect other alcohol­related problems. Of the four studies conducted provision of alcohol to people under drinkers for illegal alcohol possession age 21, minors throughout the United or consumption (Wagenaar and Wolf­ to date that focused on other social and health consequences of alcohol States can easily obtain alcohol from son 1995), an unintended and unantic­ many sources. Buyers who appear to ipated consequence of the MLDA use, three found an inverse relationship between the MLDA and alcohol­related be younger than 21 can successfully (Mosher 1995; Wolfson and Hourigan purchase alcohol from licensed estab­ in press). For every 1,000 minors problems: A higher legal drinking age was correlated with a lower number of lishments without showing age iden­ arrested for alcohol possession, only tification in 50 percent or more of 130 establishments that sell alcohol to alcohol problems among youth. The New York State Division of Alcohol­ their attempts (Forster et al. 1994, them have actions taken against them, 1995; Preusser and Williams 1992). and only 88 adults who purchase alco­ ism and Alcohol Abuse (1984) found a 16­percent decrease in rates of van­ In addition, although many youth hol for minors face criminal penalties. purchase alcohol themselves, most dalism in four States that raised the Wagenaar and Wolfson (1994) estimate youth indicate that they generally ob­ that only 5 of every 100,000 incidents MLDA. In a study of an increase of tain alcohol through social contacts of minors’ drinking result in a fine, the MLDA in Massachusetts, Hingson over age 21 (Wagenaar et al. 1996b; license revocation, or license suspen­ and colleagues (1985) did not find Jones­Webb et al. in press). These sion of an alcohol establishment. significant changes in the rates of non­ social contacts—who include friends, An in­depth review of enforcement motor­vehicle trauma, suicide, or homi­ siblings, parents, coworkers, and actions in 295 counties in 4 States cide. Smith (1986), however, found an strangers approached outside of alcohol (Kentucky, Michigan, Montana, and increase in non­traffic­related hospital establishments—purchase alcohol and Oregon) showed that in a 3­year period, admissions following decreases in the then either provide or sell it to minors. 27 percent of the counties took no ac­ MLDA in two Australian states. Jones Commercial establishments licensed tion against licensed establishments and colleagues (1992) found lower to sell alcohol, as well as social sources, for selling alcohol to minors, and 41 rates of death caused by suicides, motor face potential criminal penalties, fines, percent of those counties made no ar­ vehicle crashes, pedestrian accidents, license suspensions, and lawsuits for rests of adults who provided alcohol and other injuries in States with higher selling or providing alcohol to minors. to minors (Wagenaar and Wolfson MLDA’s. More research is needed to So why do they still supply alcohol to 1995). The States were selected for characterize the full effect of the youth? One reason is that policies are their diversity of alcohol­control MLDA on rates of alcohol­related not actively enforced. For policies to systems and availability of data. Al­ injuries and on problems other than deter specific behaviors effectively, though the majority of counties took at motor vehicle crashes. people must believe that they have least one action against alcohol estab­ some chance of being caught and that lishments and adults who provided

THE ROLE OF ENFORCEMENT they will face swift consequences for alcohol to youth, many did not take noncompliance (Gibbs 1975; Ross actions frequently. Research indicates that a higher MLDA 1992). Wolfson and colleagues (1996 b) As noted earlier, only a tiny propor­ results in fewer alcohol­related prob­ found that only 38 percent of the alco­ tion of incidents of minors’ drinking lems among youth and that the MLDA hol merchants they surveyed thought it results in fines or other penalties for of 21 saves the lives of well over 1,000 was likely that they would be cited establishments that sell alcohol. Some

VOL. 20, NO. 4, 1996 215

THE PUBLIC DEBATE OVER THE MLDA

The public debate over reducing change. Polls continue to show may be at lower risk of traffic the legal drinking age has remained strong public support for the drink­ crashes simply because they drive essentially unchanged since the min­ ing age of 21 (Wagenaar 1993a). less frequently than their U.S. imum legal drinking age (MLDA) counterparts. Other alcohol­related Question: Many Europeans let was first lowered in the 1970’s. problems are significant enough their children drink from an early Following are some frequently asked age, and European countries do in Europe that those countries are questions concerning the MLDA, not have the same alcohol­related examining the U.S. experience along with answers based on the problems that we do. Therefore, regarding MLDA policy and are research findings to date. initiating a debate over the most how can people claim that MLDA’s Question: If States are the only are a major factor in helping to appropriate age for legal access to alcohol (Wagenaar 1993a). entities that have the right to es­ prevent alcohol problems in the tablish a minimum drinking age, United States? In reviewing another country’s success with a given policy, one does Federal legislation concern­ Answer: Research confirms ing this policy area infringe on that European countries do experi­ cannot simply compare interna­ tional rates of alcohol­related prob­ State powers? ence alcohol­related problems. For Answer: The initial movement example, European countries have lems without assessing the role of factors that contribute to the prob­ to raise the MLDA to 21 was large­ rates of alcohol­induced diseases, ly fueled by citizen action groups such as cirrhosis of the liver, simi­ lems. Many cultural, political, and social conditions, which differ from in several States, which raised their lar to (or higher than) the United drinking ages before the Federal States (Single 1984). Drunk driv­ country to country, affect drinking rates. The most robust research, Government passed any legisla­ ing among youth may not be as although conducted in the United tion on the matter. Moreover, the great a problem in Europe; com­ States, has shown a strong inverse Federal Government encouraged pared with youth in the United relationship between MLDA and States to increase their MLDA’s to States, European youth obtain their alcohol consumption and its related 21 to reduce traffic crashes caused drivers’ licenses at an older age, problems: As MLDA increases, by people driving to States that are less able to afford automobiles, had lower MLDA’s. The Federal and more often use public trans­ alcohol­related problems among Government did not mandate the portation. Youth in Europe thus youth decrease. As MLDA changes

reasons that enforcement agencies minors (from 59 percent at baseline to servers to prevent illegal sales to youth, and (3) posting warning signs do not cite or arrest illegal providers 26 percent 1 year later) following an (Wolfson et al. 1996a,b). Wolfson include (1) perceived acceptance of enforcement campaign involving three and colleagues (1996a,b) found that underage drinking by community mem­ “sting operations” in which underage establishments adhering to these poli­ bers, (2) lack of community encour­ males attempted to purchase alcohol. agement to increase enforcement of In addition to increasing enforce­ cies were less likely to sell alcohol to young women who appeared to be the MLDA, and (3) lack of resources ment of the MLDA, other procedures (Wolfson et al. 1995). and policies can be implemented to under age 21 and who did not present age identification. Given the low level of enforcement improve the effectiveness of MLDA activity, it is not surprising that many laws. To ensure that adults do not sell adults do not hesitate to sell or give or provide alcohol to minors, both HE NGOING EBATE alcohol to minors. To create a deterrent public and institutional policies can be T O MLDA D effect, we need to increase the likeli­ developed that complement MLDA Despite an abundance of research hood of facing negative consequences laws (Wagenaar et al. 1996a). Alcohol demonstrating the effectiveness of the for illegally selling or providing al­ establishments, for example, can im­ age­21 MLDA at saving lives and re­ cohol to youth. One approach is to plement several policies and practices, ducing alcohol­related problems, sever­ encourage ABC and local law enforce­ including (1) requiring all alcohol al States are again considering lowering ment agencies to increase enforcement servers to receive responsible service their legal age limits for drinking. against illegal alcohol providers. Preus­ training on how to check age identifi­ Louisiana’s MLDA of 21 was recently ser and colleagues (1994) found dra­ cation and refuse sales to teenagers, challenged in court on the premise that matic reductions in alcohol sales to (2) establishing systems to monitor it violates the State’s constitutional

216 ALCOHOL HEALTH & RESEARCH WORLD

MLDA: History, Effectiveness, and Ongoing Debate

occur in Europe, researchers will searchers have come to believe ated with higher rates of drinking be able to determine more accurate­ that risk both to youth and to soci­ later in life. Research shows that ety in general can be reduced by ly the effect of a higher MLDA on when the MLDA is 21, people restricting people below age 21 alcohol­related problems among under age 21 drink less and con­ from drinking. European youth. tinue to do so through their early twenties. Those who are inclined Question: If a person is old Question: How can researchers to drink do not “make up for lost enough to serve in the military, be sure that the drop in rates of time” after turning 21 (O’Malley how can he or she not be old alcohol­related crashes among 19­ and Wagenaar 1991). enough to buy alcohol? and 20­year­olds following the —Carolyn Rosenfeld Answer: increase in the MLDA to 21 was Different activities have different ages of initiation: A related to MLDA policy? Answer: person can drive at age 16, vote in When the age­21 REFERENCES restriction was initiated, alcohol­ elections and serve in the military FELL, J.C. Drinking age 21: Facts, myths and at age 18, and serve as President at involved highway crashes declined fictions. American Association for Automotive among 18­ to 20­year­olds. This age 35. These restrictions are based Medicine Quarterly Journal 7(1):21–26, 1985. decline occurred with limited on the requirements of the specific O’MALLEY, P.M., AND WAGENAAR, A.C. Effects enforcement of the MLDA laws. activities (e.g., motor skills, capac­ of minimum drinking age laws on alcohol use,

The decline is therefore not at­ related behaviors and traffic crash involvement ity for judgment, and experience) tributable to drinking­driving en­ among American youth: 1976–1987. Journal of and take into account the risks and forcement and tougher penalties Studies on Alcohol 52(5):478–491, 1991. benefits of participation at different but directly results from lower SINGLE, E. International perspectives on alcohol as ages (Fell 1985). For example, re­ a public health issue. Journal of Public Health consumption levels (O’Malley search shows that at a given blood Policy June:238–259, 1984. and Wagenaar 1991). alcohol concentration, youth are WAGENAAR, A.C. Minimum drinking age laws.

more likely to be impaired than Question: If people cannot Encyclopedia of Drugs and Alcohol 3:1–5, 1993a.

adults. Underage drinking is WAGENAAR, A.C. Minimum drinking age and legally drink until they are 21, will alcohol availability to youth: Issues and research strongly related to serious public they just drink more when they needs. In: Hilton, M.E., and Bloss, G., eds. Eco­ health problems, including injuries reach the MLDA? nomics and the Prevention of Alcohol­Related

and death resulting from motor Answer: Research indicates Problems. National Institute on Alcohol Abuse

vehicle crashes, homicide, assault, that the opposite is true (Wagenaar and Alcoholism Research Monograph No. 25. NIH

and recreational injury. Conse­ 1993b). In fact, early legal access Pub. No. 93–3513. Bethesda, MD: the Institute,

quently, policymakers and re­ to alcohol (i.e., at age 18) is associ­ 1993b. pp. 175–200.

law regarding age discrimination. ONCLUSION C gests that to create long­term changes in youth drinking and alcohol­related Louisiana’s State Supreme Court The same arguments used to lower problems, strategies that change the concluded, however, that “. . . statutes the MLDA 20 years ago are being environment should be used. establishing the minimum drinking used today (see sidebar, pp. 216–217). Despite the MLDA of 21, minors age at a level higher than the age of Despite ongoing debates about the MLDA, research demonstrates the ef­ still have easy access to alcohol from majority are not arbitrary because they fectiveness of a higher MLDA in pre­ commercial and social sources. The substantially further the appropriate observed benefits of the MLDA have venting alcohol­related injuries and governmental purpose of improving deaths among youth. As the MLDA’s occurred with little or no active en­ forcement; simply by increasing en­ highway safety, and thus are constitu­ were lowered, rates of injuries and forcement levels and deterring adults tional” (Manuel v. State of Louisiana deaths increased; when the MLDA’s were raised, injuries and deaths signif­ from selling or providing alcohol to [La. 1996]). In other words, because icantly decreased. The benefit of using minors, even more injuries and deaths the MLDA was based on empirical environmental (i.e., external) approach­ related to alcohol use among youth ■ evidence that such laws saved lives, es, such as the MLDA, is further sup­ can be prevented each year. the court decided that the law was not ported by the fact that drinking rates arbitrary and thus did not violate were reduced even after youth turned REFERENCES

age 21. In contrast, individual approach­ BARSBY, S.L., AND MARSHALL, G.L. Short­term Louisiana’s constitution. Despite the es (e.g., school­based programs) have consumption effects of a lower minimum alcohol

Louisiana decision, the MLDA of 21 generated only short­term reductions in purchasing age. Journal of Studies on Alcohol also may be challenged in other States. underage drinking. This finding sug­ 38(9):1665–1679, 1977.

VOL. 20, NO. 4, 1996 217

ERGER AND NORTUM Problems. B , D.E., S , J.R. Alcoholic National Highway Traffic Safety Administration National Institute on Alcohol Abuse (NHTSA). The Impact of Minimum Drinking Age and Alcoholism Research Monograph No. 25. beverage preferences of drinking driving viola­ Journal of Studies on Alcohol Laws on Fatal Crash Involvements: An Update NIH Pub. No. 93–3513. Bethesda, MD: the tors. 46(3):232– 239, 1985. of the NHTSA Analyses. NHTSA Technical Re­ Institute, 1993. pp. 175–200.

UCCHIARO ERREIRA R AND ICHERMAN port No. DOT HS 807 349. Washington, DC: WAGENAAR, A.C., AND WOLFSON, M. Enforce­ C , S.; F , J., J .; S , The Effect of the 18­Year­Old Drinking Age on NHTSA, 1989. ment of the legal minimum drinking age in the A. Auto Accidents New York State Division of Alcoholism and United States. Journal of Public Health Policy . Cambridge, MA: Massachusetts Institute of Technology, Operations Research Alcohol Abuse. Drinking Age Laws and Vandal­ 15(1):37–53, 1994.

Center, 1974. ism Arrests: Preliminary Findings 1976–1981 . WAGENAAR, A.C., AND WOLFSON, M. Deterring

OUGLASS ILKINS AND LARK Albany: New York State Division of Alcohol­ sales and provision of alcohol to minors: A study D , R.L.; F , L.D.; C , F.A. The Effect of Lower Legal Drinking Ages on ism and Alcohol Abuse, May 1984. of enforcement in 295 counties in four states.

Youth Crash Involvement O’MALLEY, P.M., AND WAGENAAR, A.C. Ef­ Public Health Reports 110(4):419–427, 1995. . Ann Arbor: Univer­ sity of Michigan, Highway Safety Research fects of minimum drinking age laws on alcohol WAGENAAR, A.C.; GEHAN, J.P.; TOOMEY, T.L.;

Institute, 1974. use, related behaviors and traffic crash involve­ JONES­WEBB, R.; AND WOLFSON, M. “Organizing ment among American youth: 1976–1987. Jour­ FORSTER, J.L.; MCGOVERN, P.G.; WAGENAAR, for Alcohol Policy: What’s Needed and How nal of Studies on Alcohol 52(5):478–491, 1991. A.C.; WOLFSON, M.; PERRY, C.L.; AND ANSTINE, To Get It.” Presented at the Alcohol Policy X LANT P.S. The ability of young people to purchase P , M.A. Alcohol, sex and AIDS. Alcohol conference of the Alcohol Research Foundation, and Alcoholism 25(2/3):293–301, 1990. Toronto, Canada, May 1996a. alcohol without age identification in northeastern PREUSSER, D.F., AND WILLIAMS, A.F. Sales of WAGENAAR, A.C.; TOOMEY, T.L.; MURRAY, D.M.; Minnesota, USA. Addiction 89:699–705, 1994. SHORT, B.J.; WOLFSON, M.; AND JONES­WEBB, R. FORSTER, J.L.; MURRAY, D.M.; WOLFSON, M.; alcohol to underage purchasers in three New York counties and Washington, D.C. Journal Sources of alcohol for underage drinkers. Jour­ AND WAGENAAR, A.C. Commercial availability of Public Health Policy 13(3):306–317, 1992. nal of Studies on Alcohol 57(3):325–333, 1996b. of alcohol to young people: Results of alcohol REUSSER ILLIAMS AND EINSTEIN WECHSLER, H., AND SANDS, E.S. Minimum­age purchase attempts. Preventive Medicine 24: P , D.F.; W , A.F.; W , laws and youthful drinking: An introduction. 342–347, 1995. H.N. Policing underage alcohol sales. Journal of Safety Research 25(3):127–133, 1994. In: Wechsler, H., ed. Minimum Drinking­Age GIBBS, J.P. Crime, Punishment, and Deterrence. OIZEN Laws. Lexington, MA: Lexington Books, 1980. New York: Elsevier, 1975. R , J. Estimating alcohol involvement in serious events. In: Alcohol and Health Mono­ pp. 1–10. ORDIS G , E. From science to social policy: An WHITEHEAD, P.C. Alcohol and Young Drivers: graph 1: Alcohol Consumption and Related uncertain road. Journal of Studies on Alcohol Impact and Implications of Lowering the Drink­ Problems. Washington, DC: U.S. Govt. Print. 52(2):101–109, 1991. ing Age. Ottawa: Department of National Health Off., 1982. pp. 179–219. HINGSON, R.; MERRIGAN, D.; AND HEEREN, T. ROSS, H.L. Confronting Drunk Driving: Social and Welfare, Health Protection Branch, Non­ Effects of Massachusetts raising its legal drink­ Policy for Saving Lives. New Haven, CT: Yale medical Use of Drugs Directorate, Research ing age from 18 to 20 on deaths from teenage Bureau, 1977. University Press, 1992. homicides, suicide, and nontraffic accidents. WHITEHEAD, P.C.; CRAIG, J.; LANGFORD, N.; SMART, R.G. Changes in alcoholic beverage Pediatric Clinics of North America 32(1):221– AC RTHUR TANTON AND ERRENCE sales after reductions in the legal drinking age. M A , C.; S , B.; F , 232, 1985. R.G. Collision behavior of young drivers: Im­ American Journal of Drug and Alcohol Abuse OWLAND AND INGSON pact of the change in the age of majority. Jour­ H , J., H , R. Alcohol as a 4(1):101–108, 1977. risk factor for drownings. Accident Analysis nal of Studies on Alcohol 36:1208–1223, 1975. SMITH, D.I. Effect on non­traffic accident hos­ and Prevention 20(1):19–25, 1988. ILLIAMS ICH ADOR AND pital admissions of lowering the drinking age in W , A.F.; R , R.F.; Z , P.L.; OBERTSON JONES, N.E.; PIEPER, C.F.; AND ROBERTSON, L.S. two Australian states. Contemporary Drug Prob­ R , L.S. The Legal Minimum Drinking Age and Fatal Motor Vehicle Crashes. Washing­ The effect of legal drinking age on fatal injuries lems 13(4):621–639, 1986. American ton, DC: Insurance Institute for Highway of adolescents and young adults. SMITH, G.S., AND KRAUS, J.F. Alcohol and resi­ Journal of Public Health 82(1):112–115, 1992. Safety, 1974. dential, recreational, and occupational injuries: JONES­WEBB, R.; TOOMEY, T.L.; MINER, K.J.; A review of the epidemiologic evidence. Annual WOLFSON, M. The legislative impact of social

WAGENAAR, A.C.; WOLFSON, M.; AND POON, R. Review of Public Health 8:99–121, 1988. movement organizations: The anti­drunken­ driving movement and the 21­year­old drinking Why and in what context adolescents obtain STRUNIN, L., AND HINGSON, R. Alcohol, drugs, age. Social Science Quarterly 76(2):311–327, alcohol from older adults. Journal of Substance and adolescent sexual behavior. International 1995. Use and Misuse, in press. Journal of the Addictions 27(2):129–146, 1992. WOLFSON, M., AND HOURIGAN, M. Unintended KING, R.F. The politics of denial: The use of TOOMEY, T.L.; JONES­WEBB, R.J.; AND consequences and professional ethics: Criminal­ funding penalties as an implementation device for WAGENAAR, A.C. Policy: Alcohol. In: ization of alcohol and tobacco use by youth and social policy. Policy Sciences 20:307–337, 1987. Langenbugher, J.W.; McCrady, B.S.; young adults. Addiction, in press. Manuel v. State of Louisiana, No. 95–2189 Frankenstein, W.; and Nathan, P.E., eds. WOLFSON, M.; WAGENAAR, A.C.; AND HORNSETH, (La. July 2, 1996) [1996 WL 364739]. Annual Review of Addictions and Treatment. G.W. Law officers’ views on enforcement of MOSHER, J.F. The history of youthful­drinking Volume 3. New York: Pergamon Press, 1994. the minimum drinking age: A four­state study. laws: Implications for current policy. In: Wech­ United States General Accounting Office Public Health Reports 110(4):428–438, 1995. sler, H., ed. Minimum Drinking­Age Laws. Lex­ (USGAO). Drinking­Age­Laws: An Evaluation WOLFSON, M.; TOOMEY, T.; FORSTER, J.L.; ington, MA: Lexington Books, 1980. pp. 11–38. Synthesis of Their Impact on Highway Safety. WAGENAAR, A.C.; MCGOVERN, P.G.; AND OSHER GAO/PEMD–87–10. Washington, DC: USGAO, M , J.F. The merchants, not the customers: PERRY, C.L. Characteristics, policies, and Resisting the alcohol and tobacco industries’ 1987. practices of alcohol outlets and sales to under­ strategy to blame young people for illegal al­ AGENAAR Alcohol, Young Drivers, and W , A.C. age persons. Journal of Studies on Alcohol cohol and tobacco sales. Journal of Public Health Traffic Accidents. Lexington, MA: Lexington 57(6):670–674, 1996a.

Policy 16:412–432, 1995. Books, 1983. WOLFSON, M.; TOOMEY, T.L.; MURRAY, D.M.;

MOSKOWITZ, J.M. The primary prevention of AGENAAR ORSTER HORT AND AGENAAR W , A.C. Minimum drinking age and F , J.L.; S , B.J.; W , A.C. alcohol problems: A critical review of the re­ alcohol availability to youth: Issues and research Alcohol outlet policies and practices concerning Journal of Studies on Alcohol search literature. needs. In: Hilton, M.E., and Bloss, G., eds. Eco­ sales to underage people. Addiction 91(4):589– 50:54–88, 1989. nomics and the Prevention of Alcohol­Related 602, 1996b.

218 ALCOHOL HEALTH & RESEARCH WORLD