Substitution, addition or stepping‐stone? Exploring the temporal associations between the consumption of alcoholic and alcohol‐free beverages among Dutch university students

Abstract Background and aim Stimulating the use of alcohol‐free beverages is portrayed as one strategy to reduce excessive alcohol use and alcohol‐related harm among students in higher education. This strategy relies on the assumption that students drink alcohol‐free to substitute alcohol (i.e. ‘substitution effect’), while other possibilities like drinking alcohol‐free in addition to alcohol (i.e. ‘addition effect’) or alcohol‐free acting as a step‐up to consume (more) alcohol (i.e. ‘stepping‐stone’ effect) have largely been neglected. Consequently, this pre‐registered study aimed to examine bi‐directional temporal associations between the quantity of alcohol and alcohol‐free use among university students (aim 1). Next, these bi‐directional temporal associations were examined in subgroups: females vs. males (aim 2) and members vs. non‐members of student associations (aim 3). Design Observational study, utilizing wave 2, wave 3 and wave 4 of Radboud University's Healthy Student Life study. Data were collected May–July 2022, 2023 and 2024. Setting The Netherlands. Participants 9951 university students, of which 6790 students had only cross‐sectional data, while 2383 and 778 students participated in 2 and 3 surveys, respectively. At wave 2, participants' mean age was 23 (standard deviation = 4.0). 70% were female, 17% were members of a student association, 13% were international students and 57% were enrolled in a bachelor programme. Measurements At all waves, participants self‐reported the frequency and quantity of alcohol and alcohol‐free beverage consumption, using two (adapted) items of the Alcohol Use Disorders Identification Test – Consumption (AUDIT‐C). Additionally, participants self‐reported on: gender, student association membership, living situation, relationship status and other substance use. Last, information on biological sex, age, educational programme and (inter)national status was received from the university administrative records. Random Intercept Cross‐Lagged Panel Models were used to study reciprocal effects between the quantity of alcohol and alcohol‐free use, by decomposing observed scores in between‐person differences and within‐person dynamics. Results None of the analyses, in the total sample (aim 1; b between 0.02 and 0.03, p between 0.414 and 0.779) or in subgroups (aims 2–3; b between −0.001 and 0.12, p between 0.192 and 0.988), provided support for within‐person cross‐lagged effects between the quantity of alcohol and alcohol‐free use. Summary Among Dutch university students, no evidence was found for substitution or stepping‐stone effects of alcohol‐free beverages, which (indirectly) suggests that students generally drink alcohol‐free as an additive to alcohol, rather than as a substitute or stepping‐stone.