Alcohol front-loading among college students: exploring the need for prevention intervention.

Dear Editor:

College life is often described as the most tumultuous yet liberating experience of one's life. With the high prevalence of alcohol use among collegians, it is likely this particular age group will experience the negative consequences of at-risk drinking. Thus, this at-risk population is a critical target for alcohol-related research. Survey research indicates that 80% to 90% of college students consume alcoholic beverages, and almost half of these are high-risk drinkers (Wechsler, Lee, Kuo, & Lee, 2000). Field research has shown that many students drink alcohol with the explicit intent of becoming intoxicated (Glindemann, Geller, & Ludwig, 1996). This is disturbing when paired with the finding that 35% of all traffic fatalities in 2002 occurred in crashes with at least one legally intoxicated driver, resulting in a total of 15,019 fatalities (NHTSA, 2003). Further, the rate of alcohol involvement in fatal crashes is more than three times higher at night than during the day (63% vs. 19%).

Much research has assessed students' overall alcohol consumption (Perkins, 1992), as well as the impact of the context in which this drinking occurs (Glindemann & Geller, 2003; Senchak, Leonard, & Greene, 1998). Such alcohol consumption can be classified into three categories: a) what one drinks while in the party or bar setting, b) what one drinks before arriving in this environment, and c) what one drinks after leaving this environment. The focus of this study was on alcohol consumption before arriving in a downtown bar setting (i.e., front-loading).

Reasons for front-loading range from saving money to not being of legal age to purchase alcohol (i.e., being 21 years old in Virginia). Expanding on the latter, Turrisi, Padilla, and Wiersma (2000) found that traditional Freshmen (those who enter college immediately following high school graduation) are more likely to drink to higher levels of intoxication than non-traditional Freshmen (those who defer entering college by more than one year) or college Seniors. This suggests that age is more predictive of at-risk drinking than class rank. Under-age students may try to match their legal-age counterparts in levels of intoxication. However, because these students cannot legally purchase alcoholic beverages, they do not have as many opportunities as those of legal drinking age to reach higher levels of intoxication. Thus, front-loading serves as one way to compensate for this imbalance.

Glindemann, Geller, Clarke, Chevaillier, and Pettinger (1998) examined the alcohol consumption of 1,590 individuals in a downtown environment over a three-year period and found actual BAC levels to be highest on Thursday evenings and to get higher as a given night progressed. In a field study of designated drivers (DDs), Timmerman, Geller, Glindemann, and Fournier (2003) found that women claiming to be DDs were less intoxicated (M = .022) than men claiming to be DDs (M = .068), although both probably experienced impaired motor skills, typically beginning at a BAC of .05. These field studies demonstrate the utility of studying alcohol consumption and BACs in naturalistic settings beyond the experimental laboratory.

For the current study, it was hypothesized that: 1) those who front-loaded and consumed alcohol in a bar setting would reach higher levels of intoxication than those who engaged only in frontloading or bar drinking, and 2) those participants who were under 21 would front-load significantly more alcoholic beverages than those who were 21 years old (the legal drinking age in Virginia) or older. Other variables explored in this paper included effects of gender and Greek-life status on front-loading behavior.

All data were collected from pedestrians in the downtown area of a rural college town in Blacksburg, VA. A total of 1,528 participants (1,104 men, 424 women) completed a brief survey on their alcohol consumption for the evening and then had their BAC level assessed. The age range of participants was 18 to 59, with a mean age of 21.8 years. The majority of participants (i.e., 85%) were students attending the nearby university.

A downtown drinking questionnaire (DDQ) was used to assess self-reported drinking behaviors in bar settings, as well as alcohol consumption that occurred before arriving in the downtown area (i.e., front-loading). Each participant was asked to report gender, age, student status, Greek-life membership, and class standing. Then, participants were asked how many alcoholic beverages they had consumed (i.e., front-loaded) before coming downtown, as well as how many drinks they consumed so far while downtown. They were also asked to estimate the time they started consuming alcohol, as well as the time they arrived downtown that evening.

On each observation night, three teams of two trained research assistants conducted the assessment procedures from 11 p.m. until the bars closed (i.e., at 2:00 a.m.). Each team was assigned a particular section of the downtown area to recruit participants. They were instructed to approach every third group of one to five people. Groups of six or more were excluded due to time constraints and difficulty in obtaining information independently among large groups of intoxicated individuals (Timmerman et al., 2003). Individuals who chose to take part in the study were asked to sign an informed consent form, verbally administered the DDQ, administered a Breathalyzer test, and confidentially informed of their BAC. If a participant registered a BAC above .05, they were encouraged not to operate a motor vehicle.

Participants' BAC levels were assessed using hand-held Alco-Sensor IV breathalyzers (Intoximeters Inc., St. Louis, MO; accuracy = [+ or -] .005). They first rinsed their mouth with 2 oz of water to remove any residual alcohol before providing a breath sample. Then, each participant blew into a breathalyzer with standardized sampling procedure. The local police department periodically calibrated all breathalyzers. Survey and BAC data were collected across four semesters (Fall 2000 to Spring 2002). Initial analyses indicated no significant differences in intoxication levels or levels of front-loading across the semesters (all p's > .10). Therefore, the data were collapsed across all four semesters for all subsequent analyses.

The independent variables were gender, Greek-life status (Greek-life students vs. non-Greek-life students), age classification (under 21 vs. 21 and over), and alcohol consumption category (front-loading only vs. front-loading and bar drinking vs. bar drinking only). The dependent variables were actual BAC and self-reported number of drinks consumed before leaving one's residence (i.e., degree of front-loading).

For all analyses, only those participants who had consumed some alcohol that evening were included, reducing the sample size to 1,337. Figure 1 shows mean BAC as a function of Gender, Greek-Life Status, and Alcohol Consumption Category. It is noteworthy that Greek-life men reached high levels of BAC no matter what their alcohol consumption category, while non-Greek-life men, Greek-life women, and non-Greek-life women only reached their highest levels of BAC when they front-loaded and consumed alcohol in a bar setting.

[FIGURE 1 OMITTED]

A 2 Gender X 2 Greek-Life Status X 2 Age Status X 3 Alcohol Consumption Category (front-loading only vs. front-loading and bar vs. bar only) analysis of variance (A_NOVA) was performed on BAC. Results indicated a Gender X Greek-Life Status X Alcohol Consumption Category interaction, F (2, 1313) = 3.34,p < .05.

A Gender X Greek-Life Status interaction, F (1, 1313) = 8.02, p < .05, as well as an Age Status X Alcohol Consumption Category interaction, F (2, 1313) = 3.34,p < .05, were also found. Specifically, as shown in Figure 1, male Greek-life students (M = 0.105) were significantly more intoxicated than male non-Greek-life students (M = .078), female Greek-life students (M = .085), and female non-Greek-life students (M = .077). Also, participants under age 21 who front-loaded and consumed alcohol in a bar setting were most intoxicated (M =. 112). Their BACs were significantly higher than those a) under 21 who front-loaded only (M = .057), b) who consumed alcohol in a bar setting only (M = .066), c) 21 and over who frontloaded only (M = .061), d) who front-loaded and consumed alcohol in a bar setting (M = .093), and e) who consumed alcohol in a bar setting only (M = .074).

Main effects were found for Gender, F (1, 1313) = 7.0,p < .05, Greek-Life Status, F (1, 1313) = 7.92, p < .05, and Alcohol Consumption Category, F (2, 1313) = 30.7, p < .05. No other interactions nor main effects reached significance (all p's > .10).

To examine the mean number of drinks front-loaded, only those individuals who engaged in front-loading behavior (i.e., the frontloading only versus front-loading and bar categories) were included in the analysis, reducing the sample size to 922. Figure 2 shows mean number of drinks front-loaded as a function of Greek-life status, age status, and alcohol consumption category. It is noteworthy that Greek-life students who were under age 21 and did not consume alcohol in a bar setting front-loaded more than any other group. A 2 Gender X 2 Greek-Life Status X 2 Age Status X 2 Alcohol Consumption Category (front-loading only vs. front-loading and bar) A_NOVA was performed on the number of drinks front-loaded. A significant Age Status X Greek-Life Status X Alcohol Consumption Category interaction was found, F (1,906) = 5.01,p < .05.

[FIGURE 2 OMITTED]

Main effects were found for both Gender, F (1,906) = 31.2, p < .05, and Alcohol Consumption Category, F (1,906) = 9.2, p < .05. On average, men (M = 4.9) front-loaded significantly more alcoholic beverages than women (M = 3.4). Those in the front-loading only category (M = 5.7) front-loaded significantly more drinks than those who engaged in front-loading and consumed alcohol in a bar setting (M = 4.3). No other interactions or main effects reached significance, all p's > .10.

The results of this exploratory study are informative, and indicate that front-loading of alcohol is a serious issue among university students. In the current sample, 69% of participants reported they had front-loaded before going downtown to drink for the evening, and those participants who front-loaded and consumed alcohol in a bar setting after arriving downtown reached at-risk levels of intoxication. And, the intoxication levels were highest for the under-age drinkers who front-loaded and consumed alcohol in a bar.

The hypothesis that those who front-loaded and consumed alcohol in a bar setting would reach higher levels of intoxication than those who engaged in front-loading behavior only or those who consumed alcohol in the bar setting only was supported. The mean BAC for those who front-loaded and consumed alcohol in the bar setting was 0.096, while the mean BAC of those who only engaged in front-loading behavior and those who consumed alcohol in the bar setting reached 0.060 and 0.073, respectively. Thus, it seems front-loading alone may not be indicative of at-risk alcohol consumption. However, when front-loading was combined with consuming alcohol in a bar setting BAC levels with this population reached their highest levels, regardless of gender, Greek-life status, and age classification. Greek-life men reached the highest levels of intoxication whether they front-loaded only, front-loaded and consumed alcohol in a bar setting, or consumed alcohol only in a bar setting. All other groups (non-Greek-life men, Greek-life women, and non-Greek life women) only reached their highest level of intoxication when they front-loaded and consumed alcohol in a bar setting.

Surprisingly, the hypothesis that those students who were under 21 would front-load significantly more alcoholic beverages than those who were 21 and over was not supported. This hypothesis was based on the premise that those students who were under 21 could not legally acquire alcohol in a bar setting. However, it's possible many under-age students were able to acquire alcohol in a bar setting, either by use of a fake ID or by having their peers of legal drinking age buy their alcoholic drinks and sneak it to them. This finding supported the possibility that there was no significant difference between the BACs of those who were 21 and over (M = .083) and those who were under 21 (M = .086).

While these results are informative and give impetus for more research on front-loading, there are several limitations to this field study. First, front-loading may not be adequately defined. For example, using our definition, drinking a glass of wine or beer with dinner would be considered front-loading, while typically this behavior does not put a person at risk for DUI. Also, even though participants were ensured of confidentiality, those who were under 21 may have been inclined to present themselves as older because of potential legal implications. Finally, some self-selection bias is possible because sober people may have chosen to not take part in the study because they didn't see a need to get their BAC measured. Therefore, sober participants may be under-represented in the sample.

Nevertheless, these results are informative and should be considered when designing future interventions to prevent or reduce incidences of DUI, such as having individuals pledge to use a designated driver (DD). For example, these interventions would do well to target those planning to front-load and consume alcohol in a bar setting, as these participants reached the highest levels of intoxication. Further, given the high levels of front-loading reported here, individuals should select a DD to use when heading to the bar, as well as from the bar at the end of the night.

References

Glindemann, K. E. & Geller, E. S. (2003). A systematic assessment of intoxication at university parties: Effects of the environmental context. Environment & Behavior, 35(5), 655-664.

Glindemann, K. E., Geller, E. S., & Ludwig, T. D. (1996). Behavioral intentions and blood alcohol concentration: A relationship for prevention intervention. Journal of Alcohol and Drug Education, 41 (2), 120-134.

Glindemann, K. E., Geller, E. S., Clarke, S. W., Chevaillier, C. R., & Pettinger, C. B. (1998). A community-based feedback process for disseminating pedestrian BAC levels. Journal of Prevention & Intervention in the Community, 17(1), 55-68.

National Highway Traffic Safety Administration (2003). Traffic Safety Facts. Publication No. HS-809-606, Washington, DC: National Academy Press.

Perkins, H. W. (1992). Gender patterns in consequences of collegiate alcohol abuse: A 10-year study of trends in an undergraduate population. Journal of Studies on Alcohol, 53, 458-462.

Senchak, M., Leonard, K. E., & Greene, B. W. (1998). Alcohol use among college students as a function of their typical social drinking context. Psychology of Addictive Behaviors, 12(1), 62-70.

Timmerman, M. A., Geller, E. S., Glindemann, K. E., & Fournier, A. K. (2003). Do the designated drivers of college students stay sober? Journal of Safety Research, 34, 127-133.

Turrisi, R., Padilla, K. K., & Wiersma, K. A. (2000). College student drinking: An examination of theoretical models of drinking tendencies in freshmen and upperclassmen. Journal of Studies on Alcohol, 61(4), 598-612.

Wechsler, H., Lee, J. E., Kuo, M., & Lee, H. (2000). College binge drinking in the 1990's: A continuing problem. Journal of American College Health, 48(5), 199-210.

Footnote: This research was partially supported by a grant from the National Institute on Alcohol Abuse and Alcoholism (1 R01 AA013400-02) to E. Scott Geller (principal investigator).

Kent E. Glindemann, Ian J. Ehrhart, Melissa L Maynard, & E. Scott Geller

Virginia Polytechnic Institute & State University