OBJECTIVE

Specific events are associated with heavier and riskier substance use behaviors among college students, including holidays like Halloween, which may include several days of themed parties/events ("Halloweekend"). The current study compared drinking, pregaming (i.e., fast-paced drinking before going out for the night), cannabis use, same-day alcohol and cannabis co-use, and negative alcohol-related consequences over Halloweekend compared with two adjacent non-Halloween weekends among a sample of heavy-drinking university students.

METHOD

Participants ( = 228; 65% female) provided 28 days of daily diary data. We used a three-level generalized linear mixed model (GLMM) approach estimating zero-inflated Conway-Maxwell Poisson regressions to assess the effect of weekend and specific weekend day on number of overall drinks, number of pregaming drinks, and negative alcohol-related consequences. Proportions tests assessed for differences in any cannabis use and daily co-use between Halloweekend and non-Halloween weekends.

RESULTS

Zero-inflated portions of the GLMMs indicated that general drinking, pregaming, and negative consequences were most prevalent on Halloweekend and Fridays and Saturdays. Count portions of the models indicated that general drinking quantity was highest during these periods, and participants experienced a greater number of negative consequences on Halloweekend compared with the weekend before; no differences were observed in the quantity of pregaming drinks consumed across weekends or days. No significant differences in cannabis use or co-use were observed between weekends.

CONCLUSIONS

Given risk associated with Halloweekend compared with weekends immediately before and after, interventions targeting alcohol use and pregaming on Halloweekend may be beneficial to reduce related harm for heavy-drinking students.