Predictors of breath alcohol concentrations in college parties.

BACKGROUND: Alcohol use and subsequent consequences are harmful for individual college students. Other students and the university can also be negatively impacted by the consequences of alcohol use.

METHOD: A field-based study was used to assess the alcohol use environment at college parties. Researchers replicated a previous study by driving and walking a route to identify parties primarily on Thursday, Friday, and Saturday evenings between 9:00 PM and 1:00 AM across an academic year. Parties were randomly sampled. Hosts were asked for permission to enter the party at each sampled location. A census of partygoers was attempted at each party. Participants were asked to complete a brief survey and give a breath sample. All participants were recruited into a follow-up survey. Bivariate and multivariate analyses of individual-level and party-level factors associated with intoxication are presented.

RESULTS: The research team identified 29 parties: 16 were approached, and 12 were surveyed. Overall, 112 participants were surveyed for a response rate of approximately 28.7% of partygoers. Controlling for demographic characteristics, consumption of shots of liquor/spirits was significantly associated with a five times greater risk for intoxication. Notably, drinking games were protective of breath alcohol concentration (BrAC) risk in this model. Individuals who reported engaging in drinking games were 74% less likely to report a BrAC above the U.S. legal limit, while controlling for underage drinking in the model. Several party characteristics were identified that increased overall BrAC at the parties, including whether the party was themed, if it was a Greek life party, and whether there were illicit drugs present. Notably, when intoxication is examined by gender and party theme, women are significantly more likely to be intoxicated at themed parties: 75% were above 0.08 at themed parties compared to 35% above 0.08 at non-themed parties.

CONCLUSIONS: Field-based data collection methods can, and should, be modified to conduct needs assessment and evaluation of prevention programs on college campuses. The findings on this campus were different than the originally sampled campus. Prevention programs should target unique risks identified on each campus, and to respond to problematic party behaviors with comprehensive programming rather than policy-level bans.