A mixed-methods approach to improve the measurement of alcohol-induced blackouts: ABOM-2

Abstract

Background: Alcohol-induced blackouts describe memory loss resulting from alcohol consumption. Approximately half of college students report experiencing a blackout in their lifetime. Blackouts are associated with an increased risk for negative consequences, including serious injury. Research has documented two types of blackouts, en bloc (EB) and fragmentary (FB). However, research is limited by the lack of a validated measure that differentiates between these two forms of blackout. This study used a mixed-methods approach to improve the assessment of FB and EB among young adults. Specifically, we sought to improve the existing Alcohol-Induced Blackout Measure (ABOM), which was derived from a relatively small pool of items that did not distinguish FB from EB.

Methods: Study 1 used three rounds of cognitive interviewing with U.S. college students (N = 31) to refine existing assessment items. Nineteen refined blackout items were retained for Study 2. Study 2 used face validity, factor analysis, item response theory, and external validation analyses to test the two-factor blackout model among U.S. heavy-drinking college students (N = 474) and to develop and validate a new blackout measure (ABOM-2).

Results: Iterative factor analyses demonstrated that the items were well represented by correlated EB and FB factors, consistent with our hypothesis. External validation analyses demonstrated convergent and discriminant validity. These analyses also provided preliminary evidence for the two factors having differential predictive validity (e.g., FB correlated with enhancement drinking motives, while EB correlated with coping and conformity motives).

Conclusions: The Alcohol-Induced Blackout Measure-2 (ABOM-2) improves the measurement of blackout experiences among college students. Its use could facilitate the examination of EB and FB as differential predictors of alcohol-related outcomes in future studies.

Keywords: assessment; blackout; measurement; memory; young adult.

Figure 1.. Graded Response Model Item Threshold Estimates.

Difficulty thresholds represent the threshold between two response options such that “Difficulty Threshold 1” is equal to the difficulty threshold between a response of 0 (never) and 1 (1 time), “Difficulty Threshold 2” represents the difficulty threshold between a response of 1 (1 time) and 2 (2 times), etc. There are four total thresholds for each item because there are 5 possible response options ranging from 0 (Never in the past 30 days) to 4 (4+ times in the past 30 days). Fragmentary 1=unable to remember a few minutes, Fragmentary 2=reminded about things you had previously forgotten, Fragmentary 3=fuzzy memories of events, Fragmentary 4=memories that became clear only after reminded, Fragmentary 5=remember a few minutes after being reminded, Fragmentary 6=remember a small part of the day after being reminded, En Bloc 1=wake up with no idea where you had been, En Bloc 2=unable to remember hours at a time, En Bloc 3=lost several hours of your life, En Bloc 4=suddenly realized you have no memory, En Bloc 5=unable to remember what happened the night before.