LAUSR

AIMS This study used event-level data to identify the threshold (i.e. number of standard alcoholic drinks) at which specific negative and positive alcohol-related consequences are likely to occur. METHODS Ninety-six college students aged 18-20 reporting weekly heavy episodic drinking (HED) or at least one negative alcohol-related consequence in the past 2 weeks completed ecological momentary assessment over 28 days. Participants reported number of standard drinks consumed and negative (e.g. nausea) and positive (e.g. new friend) alcohol-related consequences on 492 drinking nights. The area under the receiver operating characteristic (AUROC) curve measured the ability of an increasing number of drinks consumed to discriminate between nights with versus without a given consequence. The Youden Index method was used to identify the optimal threshold of drinks for each consequence. Thresholds were examined for each consequence by participants' biological sex and heavy drinking status. RESULTS Across subgroups, most consequences occurred on nights where participants reported higher consumption levels relative to nights where consequences did not occur. Thresholds for negative consequences ranged between four and nine drinks and thresholds for positive consequences were between three and six drinks. CONCLUSIONS Many negative consequences are likely to occur following the traditional HED threshold of 4+/5+ drinks (females/males), with more severe consequences occurring at slightly higher thresholds. Positive consequences are likely to occur at lower thresholds. There may be an optimal number of drinks that maximize positive while minimizing negative consequences for heavy drinking college students.