How To Reduce High-Risk College Drinking: Use Proven Strategies, Fill Research Gaps

Creating an Environment That Discourages High-Risk Drinking

Many of the following approaches have proven effective with college-age youth and young adults who may or may not have been students. They are included here because results from numerous carefully conducted studies and community trials suggest their potential value in reducing high-risk drinking among college students.

Summary of Relevant Research

Interventions that change the broader environment increase the likelihood of long-term reductions in alcohol use and alcohol-related problems (Bangert-Drowns, 1988; Moskowitz, 1989; Perry and Kelder, 1992; Rundall and Bruvold, 1988; Tobler, 1992; Toomey and Wagenaar, 2002). Individual drinking behavior is influenced by myriad environmental factors such as public and institutional policies and practices, economic factors, messages in the media, and social norms (Wagenaar and Perry, 1995). Reductions in alcohol use and related problems may be achieved by changing such environmental factors (Edwards et al., 1994; NIAAA, 1997; Toomey et al., 1993; Toomey and Wagenaar, 2002).

The Panel found strong to moderate evidence supporting the effectiveness of the following environmental approaches.

Public Policy

Laws designed to decrease alcohol-related harm in the general population have had considerable success. Public policies designed to reduce the commercial availability of alcohol have also shown promise in some areas.

Laws Affecting Consumption and Consequences
Laws affecting consumption and related problems include minimum legal drinking age laws, lowered blood alcohol concentration limits, and administrative license revocation.

Minimum Legal Drinking Age: In 1984, the Federal Government enacted the Uniform Drinking Age Act, which withheld Federal highway funds from States that failed to increase their minimum legal drinking age (MLDA) to age 21 (King, 1987). By 1988, all States had established an MLDA of 21. Because the MLDA had been at younger ages in many States, researchers had access to “natural” experiments to assess the effect of these policy changes on alcohol consumption and related problems among youth. As a result, the MLDA is the most well-studied alcohol control policy.

The Panel reviewed 48 published studies that assessed the effects of changes in the MLDA on indicators of alcohol consumption (Wagenaar and Toomey, 2002). Together the 48 studies analyzed a total of 78 alcohol consumption outcome measures (e.g., sales figures, self-reported drinking). The preponderance of evidence suggests that higher legal drinking ages reduce alcohol consumption.

In addition, the Panel reviewed 57 published studies that assessed the effects of changes in the MLDA on indicators of drunk driving and traffic crashes (Wagenaar and Toomey, 2002). These studies analyzed a total of 102 crash outcome measures including fatal crashes, drunk-driving crashes, and self-reported driving after drinking. Over half the studies found that a higher legal drinking age is associated with decreased rates of traffic crashes. There is also some evidence that higher drinking ages are associated with lower rates of other health and social problems such as suicide, homicide, and vandalism. The research also suggests that these results have been achieved with minimal enforcement of the law overall. As might be expected, the studies showed that increased enforcement produces greater results (Wagenaar and Toomey, 2002).

It is important to note that almost all the studies reviewed were conducted among general youth and adult populations; very few high-quality, college-specific studies exist. The review commissioned by the Panel compiled all identified published studies on drinking age from 1960 to 1999, a total of 132 documents (Wagenaar and Toomey, 2002). It also includes an indepth discussion of methodological issues and a table that presents the results of each study, coded for a range of variables. Appendix 3 of this report contains a summary of the review authors’ responses to arguments to lower the MLDA, which may be useful to college presidents and administrators who hear these sentiments from students, alumni, and others.

Lowered Blood Alcohol Concentration Limits: Studies also attest to the effectiveness in the general population of laws designed to reduce alcohol-related traffic crashes. For example, States that lowered legal blood alcohol concentration limits from 0.10 percent to 0.08 percent experienced a 6 percent greater post-law decline in alcohol-related fatal crashes in which drivers had blood alcohol levels of 0.10 percent or higher than States that retained the 0.10 percent standard. Estimates suggest that when all States adopt these laws, 400 to 500 fewer traffic fatalities will occur annually (Hingson et al., 2000; Shults et al., 2001; Voas et al., 2000). Laws making it illegal for drivers under 21 to drive after any drinking also have produced 9 to 24 percent declines in alcohol-related deaths and driving while intoxicated (DWI) (Hingson et al., 1994; Shults et al., 2001; Wagenaar et al., 2001).

Administrative License Revocation: In addition, legally mandated administrative license revocation for drinking-and-driving offenses and mandatory seat belt use have resulted in decreases in alcohol-related fatalities (Voas et al., 2000; Zador et al., 1989).

Restrictions on the Availability of Alcohol
A number of environmental strategies are available to reduce the social and commercial availability of alcohol to college students. The Panel’s review describes these options in detail, but most have not been studied, and evidence is limited or nonexistent regarding their effectiveness for college populations (Wagenaar and Toomey, 2002). Nevertheless, a few strategies for reducing commercial availability show some evidence of success. These include increasing the price, restricting the density of retail outlets, and limiting the hours and/or days of sale.

Increasing the Price of Alcohol: With the exception of MLDA, alcohol control policies affecting price of alcohol are the next most-studied alcohol policies (Toomey and Wagenaar, 2002; Wagenaar and Toomey, 1998). Studies of price in the general population indicate that as the price of alcohol increases, consumption rates decline (Clements and Selvanathan, 1991; Gao et al., 1995; Leung and Phelps, 1993; Österberg, 1995). However, the effect on consumption varies by culture, drinking level, age group, and type of alcohol (Coate and Grossman, 1988; Cook and Tauchen, 1982; Manning et al., 1995). For example, all types of drinkers appear to be affected by price, but the heaviest drinkers may be less affected by variations in price than other consumers (Manning et al., 1995). An exception to this trend occurs among young heavy drinkers. This group, which includes college students, may be more affected by price than heavy drinkers in the general population (Chaloupka and Wechsler, 1996; Godfrey, 1997; Kenkel, 1993; Sutton and Godfrey, 1995). Inverse relationships also exist between price of alcohol and several types of alcohol-related problems, including motor vehicle fatalities, robberies, rapes, and liver cirrhosis mortality (Cook and Moore, 1993; Cook and Tauchen, 1982; Ruhm, 1996).

Two major types of policies affect alcohol pricing: restrictions on happy hours or price promotions and placing excise taxes on alcohol. Evaluation of the impact of these policies on college populations is limited. One study of college students found that an increase in beer excise taxes had little effect on male college students’ consumption (Chaloupka and Wechsler, 1996). However, the authors noted that local excise tax may be a poor proxy for price differences among campuses.

Restricting Licenses for Retail Sales of Alcohol: Studies of the density or the number of alcohol licenses per population size have found statistically significant relationships among density of alcohol outlets, consumption, and related issues such as violence, other crime, and health problems. It is important to note that many of these studies use cross-sectional designs, which are weaker than randomized, controlled trials (Gliksman and Rush, 1986; Gruenewald et al., 1993; Ornstein and Hanssens, 1985; Scribner et al., 1995; Stitt and Giacopassi, 1992). Researchers who specifically studied college students found higher levels of drinking, drinking participation, and excessive drinking among underage and older college students when a larger number of businesses were selling alcohol within one mile of campus (Chaloupka and Wechsler, 1996). Numbers of outlets may be restricted directly or indirectly through policies that make licenses more difficult to obtain, such as increasing their cost.

Limiting Hours/Days of Sale: Evaluations of the effect of restricting hours and days of sale are mixed. A few studies suggest that changes in hours may decrease rates of problem drinking, cirrhosis mortality, and some types of alcohol-related problems such as traffic crashes and violence in the general population (Duffy and Pinot de Moira, 1996; Smith, 1986). Other studies indicate no changes in problems or a shift in the timing of problems from the original closing time to the new closing hour (De Moira and Duffy, 1995). Some (but not all) studies have found that an inverse relationship may exist between the number of days of sale and alcohol use and alcohol-related problems (Ligon and Thyer, 1993; Ornstein and Hanssens, 1985; Northridge et al., 1986; Smith, 1988).

Other Approaches to Limiting Availability
Underage youth do not always have to purchase alcohol themselves in order to drink. It is readily available from other youth and young adults in party situations, and it may also be provided by older adults who condone underage drinking.

A number of policy strategies have been developed to reduce social availability. Some address locations where alcohol can be consumed legally and include community bans on drinking in public places where large numbers of youth are likely to congregate. Colleges interested in limiting social availability can employ a variety of policy options ranging from campuswide bans on any alcohol use to designating sites and occasions where alcohol can and cannot be used. When alcohol is allowed, restrictions on how it is provided can reduce the likelihood that underage persons will be served. Restrictions include keg bans or keg registration, server training, and limitations on server practices such as prohibiting self-service and limiting the number of drinks served at one time. Education and enforcement are key to all these policies and to the effectiveness of existing laws that prohibit serving alcohol to persons under age 21 (Toomey and Wagenaar, 2002).

Media Approaches

The media are another important element of the environment that can influence college student drinking. Research addresses the effects of media on drinking from two perspectives: combating the negative impact of advertising from the alcohol industry and using the media constructively to create positive change.

Alcohol Advertising Bans
A recent report by the Federal Trade Commission (Evans and Kelly, 1999) concluded that underage individuals experience significant exposure to alcohol advertising. Researchers have also found that alcohol advertising increases awareness, which affects intentions to drink (Grube, 1993; Parker, 1998). This has led some public health groups to conclude that there is a link between advertising and alcohol consumption. The Robert Wood Johnson Foundation (1999), for example, identified alcohol advertising and marketing as environmental factors that help create problems of underage and high-risk college student drinking. Although there is limited evidence indicating that alcohol advertising has an effect on consumption (Goel and Morey, 1995) and related consequences such as highway fatalities (Saffer, 1997), methodological factors explain why more such evidence has not been found (Saffer, 2002).

Some advocates have pushed for reform of advertising practices or other restrictions on alcohol advertising (DeJong and Russell, 1995). Research and experience with advertising bans are limited (Saffer, 2002), but available evidence from the general population suggests that banning alcohol advertising appears to reduce alcohol abuse in some circumstances (Ornstein and Hanssens, 1985; Saffer, 1991). Partial bans on advertising in one or two media, however, are not effective and result in increased advertising in other media (Saffer, 2002).

Counteradvertising
Evidence from tobacco advertising suggests that counteradvertising that casts doubt on the credibility of an industry and its messages can be effective (Flay, 1987; Goldman and Glantz, 1998; Hu et al., 1995; Warner, 1981).

Media Campaigns
The public health community frequently uses media campaigns to convey information to the public. However, media initiatives can also support a range of other strategic objectives, including creating a climate of support for environmental change (DeJong, 2002). Most media campaigns to prevent or reduce college student drinking have been campus-based and have used a mix of posters, flyers, e-mail, and college newspaper advertisements. Recently some regional, State, and national organizations have implemented information, social norms, and advocacy campaigns to reduce hazardous college student drinking (DeJong, 2002).

Although evaluation data on individual campaigns are limited, the body of evidence supports the following general guidelines for using the mass media effectively to address college student drinking (DeJong, 2002).

• Conduct a strategic planning process, with strategic objectives that complement an institution’s larger goals and objectives. To select program goals and objectives, analyze the student drinking problems that the institution or town faces and consider a range of communication options. The typology in Appendix 2 offers some suggested options.
• Select the target audience, and define it in terms of its geographic, demographic, psychological, and problem-relevant characteristics to help create appropriately focused messages and materials. Conduct formative research with the audience to determine perceptions and message appeal.
• Develop a staged approach that recognizes the need to build toward behavior, norm, or policy change. For example, to promote personal behavior change, an audience may need to be led through a series of steps that include awareness, knowledge and beliefs, behavioral skills, self-efficacy (i.e., the conviction that individuals can master or maintain a behavior), and supports for sustaining change.
• Include a specific “call to action”; ask the audience to take a particular step.
• Select a message source or sponsor that is credible to the particular audience; be wary of choosing celebrity spokespeople, whose appeal or credibility may be fleeting.
• Choose a mix of media channels, including online resources, and provide a clear and consistent message.
• Conduct process and outcome evaluations.

Social Norms Approaches

Norms are social “facts” of life that help frame perceptions and influence behavioral choices (Festinger, 1954; Newcomb and Wilson, 1966; Sherif, 1972). Within the disciplines of social science, the term norm is used in two different but related ways. One refers to widely shared attitudes or expectations about how people in general or members of a social group ought to behave; that is, what constitutes acceptable behavior. The term also refers to the most common behavior actually exhibited in a social group; that is, the statistical average or most typical behavior of group members. The Panel considered approaches for affecting both types of norms.

Research shows that peers have the greatest influence on student norms. When peer norms appear to encourage immoderate drinking, consumption goes up (Lo, 1995; Perkins, 1986; Robinson et al., 1993). Regardless of gender, ethnic group, residential circumstance, and Greek affiliation, most students believe that their peers hold more permissive attitudes about drinking than they actually do. Likewise, they believe that their peers drink more heavily than they do (Baer and Carney, 1993; Baer et al., 1991; Perkins and Berkowitz, 1986, 1991; Perkins et al., 1999).

The strategy of communicating actual student norms to dispel myths, referred to as the “social norms approach,” is receiving increased attention due to its simplicity, cost efficiency, and effects. The basic idea is to convey the truth about what the majority of students actually think and do concerning alcohol consumption. This approach gives students a positive message. It says that the norms are safety, responsibility, and moderation because these are the thoughts and behaviors of most students on virtually every campus.

Social norms interventions can publicize data about actual drinking norms in orientation programs, student newspaper ads and articles, radio programs, lectures, campus poster campaigns, and other public venues (social norms marketing). These activities can clarify the misperceptions of the general student body and of those students at high risk for alcohol-related problems (Berkowitz, 1997; Haines and Spear, 1996; Johannessen et al., 1999; Perkins, 1997, 2002). Programs can also target the most problem-prone groups (e.g., first-year students, fraternity or sorority members, athletes) for special attention. Workshops can help these students confront their own misperceptions of peer use and can facilitate discussion about student norms as identified in group assessments and campuswide studies (Barnett et al., 1996). One university, for example, targeted social norms interventions to entire fraternities and sororities (Marlatt et al., 1995).

Initial results from programs adopting an intensive social norms approach are promising. Several institutions that persistently communicated accurate norms have experienced reductions of up to 20 percent in high-risk drinking over a relatively short time (Berkowitz, 1997; DeJong and Linkenbach, 1999; Haines, 1996, 1998; Haines and Spear, 1996; Johannessen et al., 1999). Together these findings provide strong support for the potential impact of the social norms approach. Although any case report in this literature could be challenged methodologically, the results of each study are remarkably consistent.

Panel Recommendations: What Colleges and Universities Can Do Now

The Panel recommends that colleges and universities:

• Pay careful attention to environmental factors on campus and in the community. They are extremely important in influencing college drinking behaviors both positively and negatively.
• Actively enforce existing age 21 laws on campus; they help decrease alcohol consumption.
• Use social norms interventions to correct misperceptions and change drinking practices. When discussing college drinking problems, do not inadvertently reinforce the notion that hazardous drinking is the norm. Help students understand that they have the right not to drink and to have negative feelings about the consequences they experience due to other students’ excessive drinking.
• Communicate the institution, the community, and the State’s alcohol policies to students and parents before and after students arrive on campus.
• Be cautious about making alcohol available on campus. In the general population, increased availability is associated with increased consumption.

Panel Recommendations: What Researchers Can Do To Address Gaps in Knowledge

The Panel recommends that researchers address the following questions to fill key gaps in knowledge:

• What is the effect of banning or stringently regulating alcohol on campus? Do problems simply move off campus? How are on- and off-campus cultures affected?
• Are parental notification policies effective? If so, what are the characteristics of effective parental notification programs? At what point should parents be notified for optimal results?
• What is the most effective type of campus disciplinary system for alcohol offenses? Should campus alcohol disciplinary systems and standards be extended to students who live off campus and in what circumstances? Should infractions be handled differently for those under 21 years of age?
• How does the academic environment affect student drinking patterns? For example, would high-risk drinking be reduced if more classes were scheduled on Fridays or academic expectations were increased (e.g., reducing grade inflation, increasing difficulty of classes and requirements)?
• What is the impact of substance-free housing on alcohol problems?
• What approaches effectively reduce alcohol problems within the Greek system? Does the presence of a live-in resident advisor reduce drinking? Does delaying rush reduce alcohol problems? Do risk management efforts make a positive difference?
• What are the key environmental characteristics that influence drinking? How should environmental characteristics and environmental change be measured?
• Do alcohol-free activities and venues reduce college alcohol problems? What factors (e.g., frequency, timing, type, planning) influence effectiveness?
• How are social norms campaigns most effectively used (e.g., in combination with other activities; to set the stage for more comprehensive initiatives)?

Previous | Back to Table of Contents | Next

Historical document
Last reviewed: 9/23/2005