Skip Navigation
College Drinking Prevention - Changing the Culture

NIAAA College Materials

What Colleges Need to Know Now: An Update on College Drinking Research

A Call to Action: Changing the Culture of Drinking at U.S. Colleges

Panel Reports

College Drinking Statistical Papers

College Fact Sheet for Parents


Reducing Alcohol Problems on Campus: A Guide to Planning and Evaluation

Whole College Catalog

Prevention Curriculum

NIAAA Alcohol Alert

Tips For Cutting Down on Drinking

Alcohol Alert #68 Young Adult Drinking

Helpful Tools

In the News


Link to Us

E-mail this Page

Print this Page

High-Risk Drinking in College: What We Know and What We Need To Learn

Understanding College Drinking From a Multidimensional Perspective

Research consistently shows that no single factor determines whether a college student will misuse alcohol. Multiple developmental, individual, and environmental factors influence this outcome, both individually and interactively. These factors include public policy; social and institutional structures such as law enforcement; market mechanisms; legal availability of alcohol; economic availability, including retail price and disposable student income; social integration into college life; family roles and influence of family background and peers; family history of alcoholism; student belief system and personality; expectancies regarding alcohol’s effects; and the social context in which drinking takes place.

Developmental Factors

The problems of college drinking are, in part, a product of development. The college years are a time of transition that involve multiple adjustments including a reexamination of identity, exploration of new social relationships, and changes in living situations. However, this time is also one of potential transition for those who do not attend college. To the extent that high-risk drinking and drinking problems can be identified among college students and same-age nonstudents, it is possible that behavior usually associated with “the college years” is actually characteristic of “the years of late adolescence and early adulthood” more generally. For both college and noncollege youth, this period of life involves greater personal freedom and independence, increased involvement in intimate relationships, and freedom from the responsibilities that marriage, family life, and the workplace typically entail.

When considered in a developmental framework, college students face multiple, challenging transitions (Schulenberg and Maggs, 2002), including:

Pubertal and physical development. Hormonal changes, physical development, and societal expectations lead to increased interest in sexual relationships and in use of alcohol. Adolescents begin to look like adults, and they may desire adult status and privileges such as the right to drink.

Brain development. During adolescence, a major “remodeling” of the brain occurs in most species. This remodeling includes not only the formation of new synaptic connections in certain neural systems, but also the pruning of synaptic connections in specific neural systems. Research findings about the behavioral and developmental characteristics of human adolescents are supported by findings from animal studies (Spear, 2002). For example, reminiscent of human adolescent behavior, adolescent rats are often hyperactive and explore more vigorously relative to rats of other ages (Spear et al., 1980). In addition, both human adolescents and adolescent laboratory rats show an enhanced hormonal and physiological response to stressors (Bailey and Kitchen, 1987; Meaney et al., 1985; Ramaley and Olson, 1974; Rivier, 1989; Walker et al., 1986, 1998). What makes these similarities interesting in the present context is other research that suggests that exploratory behavior and stress may be important factors involved in the tendency of human adolescents to drink heavily (Baer et al., 1987; Deykin et al., 1987; Pohorecky, 1991; Tschann et al., 1994; Wills, 1986). These and other related findings are provocative; however, additional study is needed to determine whether and how biological brain remodeling and adolescent behavioral and developmental characteristics are associated with alcohol initiation, use, and misuse. Further research is also needed to establish whether adolescence represents a period of particular vulnerability to alcohol neurotoxicity (Brown et al., 2000; DeBellis et al., 2000; Tapert and Brown, 1999).

Cognitive and moral development. Normative cognitive changes in this period include the increased ability to think abstractly, view issues as relative rather than absolute, and make judgments based on higher-level, universal principles such as justice and equality, rather than “arbitrary” rules. Due in part to these cognitive developments, some college students view adult-imposed prohibitions against youthful alcohol use skeptically. They know that adults use alcohol and may consider age-based restrictions unfair and discriminatory.

For many older adolescents and young adults, the decision to drink is a rational one. For others, norms supporting excessive consumption, combined with inexperience, often lead to risky drinking behavior.

Identity domain transitions. College is a time when students, through the exploration of philosophies, lifestyles, relationships, and behaviors, eventually make commitments to an integrated set of personal beliefs, values, and goals. Such exploration of identity is normal and healthy but may increase experimentation with risky behaviors, including alcohol consumption.

Transformations in relationships with the family of origin. College students experience increasing autonomy and independence from their parents. Ideally, increasing independence should occur in the context of continued family support and attachment. Indeed, the quality of a college student’s relationship with his or her parents may actually improve when the adolescent moves out, even though the quantity of parental interactions decreases, thus reducing day-to-day parental influence. Having older siblings may increase the college student’s alcohol expectancies and consumption pattern, especially if he or she looks up to an older sibling who drinks.

Transformations in relationships with peers. Older adolescents spend more time with their peers, and many are susceptible to peers’ suggestions that they engage in risky behaviors, including excessive drinking. Cultural myths about campus drinking may increase use and misuse of alcohol, especially when alcohol use is considered a fundamental part of social relationships and socializing.

A developmental perspective encourages the examination of alcohol use and heavy drinking in relation to normative developmental tasks and transitions in college students’ lives. Because conceptual models can be useful in relating developmental transitions to health risks, including risks from alcohol misuse, five that have been usefully applied to the problem of college student drinking are included here (Schulenberg and Maggs, 2002).

  1. The overload model postulates that multiple developmental transitions may overwhelm the individual’s coping capacities, resulting in increased health-risk behaviors such as heavy drinking.
  2. The developmental mismatch model posits that developmental transitions may decrease the match between individuals’ needs or desires and opportunities in the new contexts in which they find themselves, resulting in an increase in health-risk behaviors (e.g., as a form of compensatory behavior or self-medication).
  3. The increased heterogeneity model states that developmental transitions may exacerbate individual differences in ongoing health-risk trajectories. Thus, a college student who already has an emotional or psychological problem may have difficulty negotiating the challenge of a new transition and may turn to heavy drinking as a form of self-medication.
  4. The transition catalyst model states that health-risk behaviors may help in or be fundamental parts of negotiating certain developmental transitions. Thus, a college student may drink heavily in the belief that drinking may lead to new friendships, romantic and/or sexual relationships, and social bonding.
  5. The heightened vulnerability to chance events model postulates that individuals undergoing developmental transitions may seek out novel experiences, thereby increasing their vulnerability to both the positive and negative effects of chance events. For example, a student who typically does not drink to excess may participate in heavy episodic drinking as a rite of passage in college, placing himself or herself in a position of heightened vulnerability to alcohol’s damaging effects.

Models such as these can often enhance our understanding of developmental phenomena and provide guidance for theoretically based research.

The particular challenge of the college student’s first year. The first year in college represents a social and developmental milestone for all college students, whatever their background or type of institution. This transition is often so difficult to negotiate that about one-third of first-year students fail to enroll for their second year of college (Upcraft, 2000). There is some anecdotal, although not much empirical, evidence that the first 6 weeks of enrollment are critical to first-year student success. Due to the changing demographics among students attending college, first-year experiences may be variable (see box below).

Many college students today are not as academically prepared for college as they should be, as reflected in the finding that about 29 percent of today’s first-year college students are enrolled in remedial reading, writing, or mathematics courses (King, 1998). Many first-year college students are also under financial pressure; only about 20 percent of undergraduates between the ages of 18 and 22 are pursuing an exclusively parent- or student-financed education (National On-Campus Report, 1992).

First-year college students also appear to be more academically “disengaged” than those in years past, according to 1999 freshman norms based on the responses of 261,217 students at 462 U.S. 2- and 4-year colleges and universities (Sax et al., 1999). In 1999, fully 40 percent of first-year students reported feeling frequently “bored in class”—a record number. In 1985, that percentage was 26 percent. Along with reporting boredom in class, a record high of 63 percent of college freshmen came late to class frequently or occasionally in 1999 compared to 49 percent in 1966. The percentage of first-year students who overslept and missed a class or appointment rose to 36 percent in 1999 from 19 percent in 1968.

In addition to academic disengagement, a record number of entering college students—30 percent—reported feeling frequently “overwhelmed by all I have to do” in 1999 compared to a low of 16 percent in 1985 (Sax et al., 1999). Although occasional feelings of anxiety have long characterized the majority of entering college students and are considered a normal part of this developmental transition, the percentage of students who report feeling “frequently” overwhelmed has grown steadily over the past 15 years, with 39 percent of women and 19 percent of men reporting this level of stress in 1999. A possible factor contributing to the growing stress among incoming college students is the record proportion who report “some” or a “very good likelihood” of working full-time while attending college—25 percent in 1999 compared to 16 percent in 1982 (Sax et al., 1999).

Trajectories of binge drinking. Distinct trajectories of binge drinking during the transition to young adulthood (ages 18 to 24) have been documented by Schulenberg and colleagues (1996) in a four-wave study. These researchers identified six trajectories of binge drinking that applied to all but 10 percent of the sample:

  1. Chronic (two or more binge drinking episodes in the last 2 weeks across all four waves);
  2. Decreased (started like the chronic group in high school, and then decreased binge drinking across the four waves);
  3. Increased (very little binge drinking in high school, and increased binge drinking across the four waves, catching up to the chronic group);
  4. Fling (very little binge drinking in high school, followed by a rapid increase and then a decrease across the four waves);
  5. Rare (very little binge drinking across the four waves); and
  6. Never (no binge drinking across the four waves).

Trajectories vary according to gender, ethnicity, and college student status in ways consistent with findings already described in this report. For example, compared to men, women are underrepresented in the chronic and increased groups and overrepresented in the never group. Compared to most ethnic minorities, Whites are overrepresented in all groups except the never group.

When considering the problem of heavy drinking during adolescence, and especially during the transition to young adulthood, it is essential to examine different trajectories over time. A key reason to be concerned with differential change in alcohol use over time is that a given level of use at any given time could result from a number of different trajectories, with some being far more troublesome than others.

The challenges faced by colleges and universities encompass more than the behavioral issues linked to transition. In many cases, they must also respond to a subset of alcohol problems that began during high school or earlier. Although the college experience does not “cause” these drinking problems per se, they are frequently identified in college and thereby become a “college problem.”

Individual Student Factors

No one individual profile describes college students who drink. Many factors specific to the individual influence how a particular college student views and uses alcohol. Differences in personality, social relationships, beliefs, attitudes, psychological needs, and responses to alcohol have all been studied to explain why some people use alcohol more than others. The challenge for researchers is to integrate these variables and develop multivariate models that can explain the relationship between patterns of drinking and combinations of risk factors and outcomes (Baer, 2002).

Genetic vulnerability and family factors. Alcohol problems run in families, with the best available research indicating that both genetic and environmental components contribute to risk (McGue, 1999). Although extensive genetic research is under way, the mechanisms of genetic and family risk remain unclear. Approximately 10 percent of college students report growing up in a home where a parent abused alcohol. These children of problem drinking parents exhibit a bimodal pattern of drinking behavior, with higher than normal odds of past-year abstinence or binge drinking (Weitzman and Wechsler, 2000). Although there is little evidence that children of alcoholics metabolize alcohol differently than others, they may be more sensitive to the early, stimulating, and stress-dampening effects of alcohol and less sensitive to the delayed, subjectively assessed depressant and motor effects of alcohol (Newlin and Thomson, 1990; Schuckit, 1998; Sher, 1991; Wood MD et al., 2001).

Personality. Decades of research have failed to identify an “addictive personality.” However, certain personality traits have been related to drinking habits. For example, sensation-seeking has been related to higher rates of consumption, while religiosity has been related to lower rates. Personality traits are typically seen as mediating or moderating the relationship between biological, psychological, social, and environmental factors and subsequent alcohol use and misuse.

There is strong and consistent research evidence linking problem drinking with impulsivity and disinhibition, moderate evidence of links with neuroticism and emotionality, and mixed evidence for a link with sociability and being extraverted (Sher et al., 1999). Consistent with this research is the finding that two problem drinking patterns appear to be dominant among college students. These are (1) a pattern of heavier drinking related to impulsivity and sensation-seeking and (2) a pattern of heavier drinking associated with negative emotional states.

Beliefs about alcohol. Learning about alcohol can occur at very young ages, before alcohol is consumed, as children observe others drinking (often their parents). There is good evidence that beliefs about alcohol are related to the initiation of drinking behavior. It is common for a college student to begin drinking based on what he or she has observed adults doing to cope with stress, enliven a party, or relieve boredom. Although most of this belief structure is in place before college, the college environment presents a stimulus structure that can reinforce prior beliefs—for example, that drinking is fun and makes people sexier.

Students vary considerably in their perceptions and expectations of whether alcohol is a positive or negative influence on behavior. Their reasons for drinking also vary and have been linked to the management of specific emotional states, such as feeling unhappy, with drinking seen as a form of self-medication for these feelings. For some, a belief becomes a linkage to alcohol use. For example, if a student believes that alcohol will make him or her more sociable, he or she may drink for that reason. A person who drinks to manage negative emotional states might use alcohol to cope with stress, relieve depression or social anxiety, or boost low self-esteem. It should be noted, however, that in the absence of beliefs that alcohol produces certain psychological states—stress reduction and/or mood elevation—the relationship between stress and depression and alcohol use is not found (Cooper et al., 1995; Kushner et al., 1994).

Religiosity. Several large, multicampus studies show that students who are more religious and more committed to traditional values drink less than their peers who are less religious (Engs et al., 1996; Wechsler et al., 1995a).

The influence of prior drinking, peers, and family. For some students, alcohol use in high school has already set the stage for college drinking, with an enabling environment on campus supporting precollege drinking behavior. A study of 140 college campuses found that the frequency of binge drinking in high school predicted the frequency of binge drinking in college (Wechsler et al., 1995a).

Peer use is one of the strongest correlates of adolescent alcohol use (Bucholz, 1990; Jacob and Leonard, 1994). Young people tend to select peers who drink like they do and to influence each other to drink (Curran et al., 1997; Kandel, 1986).

A number of parenting practices including parental conflict, insufficient monitoring of adolescent behavior (e.g., not knowing where children are at night), and poor communication have also been associated with adolescent drinking problems (Barnes, 1990; Jacob and Leonard, 1994).

Environmental Factors

Although the existing literature on the influence of collegiate environmental factors on student drinking is limited, a number of environmental influences working in concert with other factors may affect students’ alcohol consumption (Presley et al., 2002). Students are not passive members of the college community; campus culture interacts with personality and experiential variables to influence the use and misuse of alcohol. Some potentially influential environmental factors are listed below.

The social scene.The college years are marked by social activity with much student drinking occurring at small and large parties. Indeed, the social environment on campus and social processes appear to play a critical role in influencing drinking in college (Baer, 1993; Maggs, 1997).

College organizational aspects.Several aspects of a college’s organization are associated with student drinking. Among them are the following:

  • Historically black colleges and women’s colleges. Historically Black colleges and universities and women’s colleges tend to have lower rates of excessive drinking compared to predominantly White and coeducational institutions (Dowdall et al., 1998; Meilman et al., 1994, 1995). More research is needed to determine whether attendance at a historically Black or women’s college mitigates against excessive alcohol use.

  • Presence of a Greek system on campus. The presence of a Greek system on campus increases the likelihood of heavy alcohol use. Similarly, participation by individual students in fraternities or sororities tends to increase the likelihood that they will drink heavily. Living in a Greek house, belonging to a Greek organization, and intent to join the Greek system are all correlated with higher rates of binge drinking, frequency of drinking, and negative consequences associated with drinking (Klein, 1989; Lo and Globetti, 1993; Wechsler et al., 1996; Werner and Greene, 1992). Among members of fraternities and sororities, the rate of binge drinking (according to the 5/4 definition) is 65 percent; among those living in fraternity and sorority houses, the rate is 79 percent (Wechsler et al., 2000b). What is not known, however, is whether and to what extent fraternities and sororities attract those who are more inclined to drink excessively and whether and to what extent such behavior is a result of participation in the Greek system (Borsari and Carey, 1999). Probably both scenarios are occurring to some extent. One study, for example, found that a much higher percentage of male students who were binge drinkers in high school became members of fraternities in college and that among women who did not binge in high school, those who joined a sorority were much more likely to start binging in college than those who did not join a sorority (Wechsler et al., 1996). It should also be noted that while the presence of a Greek system is associated with higher rates of binge drinking on campus, there are colleges that have no Greek system and a high percentage of binge drinkers.

  • Importance of athletics on campus. Multi-institutional research has found that the importance of athletics on campus and student involvement in athletics are positively associated with higher rates of excessive drinking (Leichliter et al., 1998; Nelson and Wechsler, 2001; Wechsler et al., 1997a). A March 1999 symposium sponsored by The Higher Education Center for Alcohol and Other Drug Prevention focused on the fact that college athletes are more prone to the adverse consequences of alcohol than are nonathletes. Research has also shown that athletes who are members of the Greek system are at even greater risk for heavy drinking (Meilman et al., 1999). However, no study to date has looked at this issue with respect to campuses that are both Greek and focused on athletics to discern how these two factors, when linked, relate to overall campus alcohol consumption and campus culture.
  • Two-year versus 4-year institutions. Data from 2- and 4-year colleges and universities show that students at 2-year institutions reported lower average weekly consumption of alcohol and lower rates of binge drinking than students at 4-year schools (Presley et al., 1993, 1995, 1996a, 1996b).
  • Substance-free residence halls and campuses. Research indicates that living in a substance-free residence hall has a protective effect and is associated with a lower likelihood of binge drinking in college for students who did not binge in high school (Wechsler et al., 2001b). In addition, students living in substance-free dorms experienced fewer secondhand effects than students living in unrestricted housing. Another study found that college students on campuses that ban alcohol were 30 percent less likely to binge drink and were more likely to abstain from alcohol (Wechsler et al., 2001a). Furthermore, fewer students on campuses that ban alcohol experience secondhand effects from others’ drinking than those on campuses that do not ban alcohol.

Physical properties of college campuses. Several physical properties of college campuses are associated with college student drinking, including the following:

  • Commuter versus noncommuter schools. If a college is primarily a commuter institution, alcohol consumption among its students tends to be lower. Commuters living at home are more likely to be lighter drinkers than students who live on campus (O’Hare, 1990; Wechsler et al., 1994, 1998, 2000b). The Core survey found differences in drinking levels between students who lived in on-campus versus off-campus housing (Presley et al., 1996a). The average number of drinks per week and the number of binge-drinking episodes were higher for on-campus as compared to off-campus residents. On-campus residents who drank the most lived in a fraternity or sorority house (Presley et al., 1993; Wechsler et al., 2000b). Students living at home appear to be more likely to drink in night clubs and bars, whereas residence hall students are more likely to drink in large, mixed-gender groups in their own residences. These findings are not surprising. Although parents and peers are both influential in defining standards of drinking, peers appear to be more influential in terms of affecting actual drinking behavior (Fromme and Ruella, 1994).
  • School size. Students at smaller schools consume greater amounts of alcohol on an average weekly basis than students at larger schools (Presley et al., 1993, 1995, 1996a, 1996b). This may be partially explained by the fact that larger schools are likely to have more commuter students who tend to drink less (see above). Because school characteristics such as size are correlated so closely with other institutional characteristics, such as public versus private sponsorship, religious affiliation, and location (rural, small town, suburban, urban), it is difficult to disentangle the influences of these characteristics.
  • Location. Alcohol consumption rates in colleges vary by region. It has been consistently shown that students at schools in the Northeast section of the United States, followed by those in the North Central region, consume more alcohol and have higher binge drinking rates than students at colleges in other sections of the country (Presley et al., 1993, 1995, 1996a, 1996b; Wechsler et al., 1994, 1998, 2000b). These regions also have the highest rates of occasional heavy use and annual and 30-day use among young adults generally (Johnston et al., 2001a, 2001b). There is also anecdotal evidence that students on rural campuses drink more than students on urban or suburban campuses. The CAS data show that binge rates of rural/small town campuses are consistently higher than those of urban/suburban campuses (for example, 49 percent versus 42 percent in 1999), although the differences are not statistically significant.

Alcohol pricing. Researchers agree that higher alcoholic beverage prices and higher taxes result in less drinking; however, the magnitude of consumer response is more difficult to specify. Using econometric estimates in a policy simulation analysis, one study found that increases in alcoholic beverage prices would lead to substantial reductions both in the frequency of alcohol consumption by youth and in heavy drinking among youth (Chaloupka, 1993). The study also concluded that the effects of excise tax hikes on alcohol exceeded the effects of establishing the uniform legal drinking age of 21 in all States studied. When the research was expanded to include not only the monetary price of alcoholic beverages, but also the other “costs” of heavy drinking, including time spent obtaining alcohol and legal costs associated with drinking-related behavior, it found that drinking by youth is price-sensitive (Chaloupka et al., 1998). Increases in total cost can significantly reduce consumption (Chaloupka et al., 1998). Data from the CAS indicate that price significantly affected underage drinking and binge drinking by female students but not by male students (Chaloupka and Wechsler, 1996). Another analysis of CAS data found that underage, as compared to legal age, college students were more likely to obtain alcohol very cheaply and that paying a low price per drink or a set fee for “all you can drink” is associated with heavy episodic drinking (Wechsler et al., 2000a).

Outlet density and drinking venues. Research has shown that (Gruenewald, 1999):

  • Population growth leads to a greater number of alcohol retail outlets;
  • Greater numbers of alcohol retail outlets translate to greater alcohol use; and
  • Greater use of alcohol results in more alcohol-related problems.

One study found that when alcohol outlet concentrations increase and multiple drinking venues exist, both long-term and short-term drinking problems increase (Gruenewald, 1999). Another study indicated that level of drinking, drinking participation, and binge drinking are all significantly higher among college students when there are a greater number of outlets licensed to sell alcohol near campus (Chaloupka and Wechsler, 1996). A third study found that parties, dates, and socializing, along with being with friends, are the most common situations where heavy student drinking occurs, suggesting that a reduction in alcohol outlets might affect student drinking levels in social situations (Clapp et al., 2000). Although this was a single-institution study, it was well designed and explored some of the environmental variables that may put college students at risk for alcohol misuse.

There is no doubt that social availability affects drinking on campus. Social availability is defined in this context as actual, easy access to alcohol, such as at beer-keg parties where heavy drinking is the norm; participating in drinking games (see Drinking Games); and attendance at gatherings where older students obtain alcohol for younger students.

A study that followed 319 young adults throughout 4 years of college and for 1 to 3 years afterward found that during the college years members of Greek societies consistently drank more heavily than their non-Greek peers (Sher et al., 2001). Statistically controlling for previous alcohol use did not eliminate the higher consumption level among Greeks. However, Greek status did not predict heavy drinking levels postcollege. Study results suggest that perceived social norms associated with drinking in the Greek system are largely responsible for the prevalence of heavy drinking among fraternity and sorority members. In short, heavy drinking among Greeks is the norm and Greeks perceive their peers as supportive of a heavy-drinking lifestyle. However, once out of a Greek environment in which heavy drinking is normative and encouraged, these young adults’ drinking patterns are similar to those of their non-Greek peers (Sher et al., 2001).

Strategies for Filling Gaps in Knowledge: Understanding Drinking in College From a Multidimensional Perspective

Research on drinking among college students must take into account the multiple developmental, individual, and environmental factors (and their interactions) that appear to affect whether and how much college students drink.

The research literature in this field is large but of uneven quality. To improve this situation and build on some of the excellent work in this area, future research efforts should test interactive and mediating models of multiple risk factors based on theory, address developmental processes, and use additive models of multiple risks to identify those students at highest risk for alcohol-related problems.

Developmental processes. Developmental transitions represent windows of opportunity for effecting change and the college student is dealing with a number of such transitions. More studies are needed on the value of intervening at critical transitional periods with developmentally appropriate prevention strategies to reduce excessive drinking. Intervention strategies should be implemented not only on an individual level, but also on a contextual level aimed at changing group social norms. Students themselves should be partners in this process.

There is also a need for more research on the unique characteristics of the adolescent brain and on ways in which these neurological features may predispose adolescents to behave in particular ways, including the initiation of drinking behavior. Studies are also needed to explore whether adolescents may show reduced sensitivity to alcohol intoxication, leading in some cases to higher alcohol intake to attain reinforcing effects. Finally, the possibility that adolescents are particularly vulnerable to the long-term effects of alcohol on cognitive development needs to be investigated more thoroughly.

The college context. Since contextual factors are correlated with drinking by young people, studies are needed to:

  • Examine alcohol retail outlet density and alcohol pricing with respect to the specific college context;
  • Investigate whether students “self-select” for high-binge institutions, and how students arrive at their perceptions of a college’s or university’s high drinking rate;
  • Assess the relationship of high-risk drinking in college to the surrounding communities’ tolerance for drinking; students’ perceptions of drinking at their college relative to their perceptions of drinking at other colleges; and students’ individual beliefs about alcohol and about their own drinking patterns; and
  • Elucidate how cultural factors on campus influence high-risk drinking, including protective factors such as social relationships and networks that appear to decrease risk (Weitzman and Kawachi, 2000).


Previous | Back to Table of Contents | Next


Historical document
Last reviewed: 9/23/2005

About Us
Site Map
Privacy Policy
Contact Us
Web site Policies

NIAAA logo HHS logo USA dot gov logo