The project described in this report was conducted at the University of North Carolina at Chapel Hill (UNC-CH). Like many university communities, UNC-CH has experienced alcohol-related deaths and injuries during the past several years and much attention has been focused on alcohol issues. This project examined the nature of student drinking at UNC-CH, then used that information to develop and evaluate a program designed to dispel the common misperceptions about drinking by students. A unique feature of this project was the collection of voluntary breath measurements from students returning to their residences late at night. These data provided the core information for a program designed to help students realize that drinking less common than many think. While the overriding concern alcohol poses to highway safety professionals is alcohol-related motor vehicle crashes, strategies that address the problem of alcohol impairment in general can reduce the tragic consequences of drinking and driving while also reducing other alcohol-related injuries and deaths.
The program that was developed and implemented is not intended to stand alone. It is designed to serve as an important component in a comprehensive campus alcohol program. It is hoped that, in conjunction with the variety of other programs on campus that support healthy behaviors and wise choices, this program will help to prevent future tragedies.
A 1997 national survey of students from 130 U.S. colleges & universities found, based on self-report, that 43 percent of students had consumed either five drinks (males) or four drinks (females) on a single occasion during the past two weeks. These persons are typically labeled as ‘heavy episodic’ or ‘binge’ drinkers (Wechsler et al., 1998). Twenty-one percent were frequent ‘heavy episodic’ drinkers (five/four drinks on an occasion three or more times in the past two weeks). This four-year follow-up of a similar 1993 survey (Wechsler et al., 1994) found few changes in student drinking, but did detect increases in consequences associated with drinking and motivations for drinking. A repeat of this survey in 1999 covering 119 campuses detected both encouraging and discouraging changes, with a greater proportion of abstainers, but also more of the heaviest drinkers (Wechsler, et al., 2000).
Excessive consumption of alcohol by college students exposes them to a variety of risks. In addition to the risks of driving after drinking (Sleet, Wagenaar, & Waller, 1989; National Highway Traffic Saftey Administration, 1999) , students also experience increased risk of injury as pedestrians and bicyclists, and as the result of falls and fires (Hingson & Howland, 1993). Heavy drinking is associated with greater probabilities of sexual assault, health problems, unsafe and unplanned sexual activity, sexual harassment, impaired sleep and study time, and interpersonal problems (Presley, Meilman & Lyerla, 1997, Roizen, 1997; Harrington & Leitenberg, 1994). Alcohol use is commonly involved in a variety of unintentional injuries, including drowning (Smith et al., 1999), suicide, homicide (Goodman et al., 1991) and injuries resulting from interpersonal violence (Martin, 1992). Because of the prevalence on college campuses of old multi-story buildings, especially residence halls, falls from windows, off buildings and down stairwells are a common cause of death and injury for college students. Excessive drinking is often a contributing factor in fall-related death and injury (Hingson & Howland, 1993). UNC-CH in recent years has experienced the death of an impaired student who ran into the side of a moving vehicle, and student deaths from alcohol-related falls and fires, tragedies that galvanized the resolve to address student drinking. Clearly, alcohol is a serious problem at UNC-CH and college campuses everywhere. As campuses attempt to address this problem, it is important to better understand actual alcohol use by students.
A number of recent studies have examined self-reported drinking behavior among college students, but these studies present an incomplete picture of the phenomenon. Although reports of having five drinks on an occasion -- a commonly used measure -- clearly indicate atypically heavy drinking, which is associated with a variety of problems, this is an imprecise measure. For example, consumption of four drinks by a small inexperienced female drinker, during a one-hour period would produce a substantially higher BAC and greater impairment than that same amount consumed over the course of an entire evening by a large, experienced male drinker. Despite this shortcoming, almost no data have been collected that reveal degree of impairment or blood alcohol concentration (BAC) levels reached when students drink.
With the exception of experimental work, collection of actual impairment (BAC) data has been limited almost exclusively to roadside surveys of drivers (Foss & Beirness, 1996; Voas et al., 1998). To our knowledge only two other research teams have obtained BAC data in surveys of college students. Both of these involved non-random samples obtained near drinking establishments. In 1988 Werch et al. obtained BAC measurements from individuals outside bars adjacent to a university campus, although respondents’ student status was not determined. More recently Glindeman et al. (1998) measured BACs for nearly 1,600 individuals along a commercial strip adjacent to a large university (83% of whom were students) over a three year period. This was an intervention designed to provide BAC information to pedestrians, rather than an attempt to obtain information from a representative sample. Nonetheless, the data did allow some objective examination of the nature of student drinking. Among the noteworthy findings was that results from objective BAC measurement were somewhat at variance with findings from self-report surveys. This further emphasizes the need for improved measurement of student alcohol use.
The present study was conducted to develop a more complete understanding of student alcohol use on one campus, to support development of a program to reduce excessive or dangerous alcohol use. The several shortcomings of simple retrospective accounts of drinking in response to closed-end survey questions led us to seek an alternative measurement. Concerns about whether individuals actually count (and remember) the number of drinks they have, the decreased likelihood of accurate counting and/or recall as one becomes impaired, difficulties with the notion of a ‘standard’ drink especially for individuals who often consume beer from containers of decidedly non-standard size (combined with varying alcohol content in different drinks even given equivalent volume) all point to the need for an additional approach to measuring student drinking. Direct BAC measurement provides a highly desirable adjunct to self-reports of drinking, especially when assessing the effects of interventions. In the present study, which we believe is the first of its kind, we combined responses to typical survey questions, with a direct measurement of BAC.
The goal of this project is to examine the nature of student drinking behavior at UNC-Chapel Hill then to develop and evaluate a comprehensive program designed to ameliorate the problems of excessive drinking and driving (or walking or riding a bicycle) while impaired among this population. As a distinguishing feature of this project, voluntary breath measurements obtained from a representative sample of students were used to enhance our understanding of alcohol use on this campus. This information, in conjunction with responses to a brief interview, provided useful information and insights about the nature of college student drinking behavior as we developed a comprehensive program. This unique information also became an integral part of the program.
From the beginning, this project reflected a "Safe Communities" approach, wherein various groups and organizations within and external to the community form partnerships, bringing their unique resources together, to address a problem. The traditional means of identification of the alcohol problem (self-reported survey data, crash and alcohol-related incident data) was augmented with the voluntary breath measurement data (collected at night as students returned to their places of residences). Since analysis of the data indicated a large discrepancy between perceived and actual consumption of alcohol by students, a year-long information program using the social norms approach was developed. This program included incentive campaigns through which students were rewarded for knowing or displaying the accurate alcohol facts. The primary evaluation of this program was a repeat of the breath measurement survey.
Having several partners was particularly valuable to this project’s ability to perform a variety of tasks. Each brought the kinds of resources that they are uniquely suited to provide. The North Carolina Governor’s Highway Safety Program provided the initial support to explore the feasibility of doing a project in which a college campus was treated as a community, and provided the funding for the collection of the survey data. As the scope of the project grew into one that held promise to become a national model, using a unique approach to data collection on a large campus with a national reputation, the National Highway Traffic Safety Administration provided the funding needed to expand the project. The University provided the funds for the incentive programs and the placement of advertisements in the campus newspaper.
The implementation of the program was accomplished through several units and constituencies of the University working together with the Highway Safety Research Center (HSRC). Among the various players from the University were the Vice Chancellor for Student Affairs, the Dean of Students, individuals from the Center for Healthy Student Behaviors, the University housing office, including the Director and residence hall staff, representatives of student government including several student body presidents, representatives of the Greek community including the Director of Greek Affairs and presidents of the Interfraternity and Panhellenic Councils, the University Parents Council (who provided financial support for parts of the program), and the University Office of News Services. Without any one of these parties, the project could not have been done.