The Relationship Between Electronic Nicotine Delivery System Use and Alcohol Consumption: A Neurocognitive and Behavioral Investigation

American English
Embargo Lift Date
Committee Chair
Degree Year
Department of Psychology
Purdue University
Journal Title
Journal ISSN
Volume Title
Found At

Increasing research shows that the use of electronic nicotine delivery systems (ENDS) is associated with higher rates and quantity of alcohol consumption; however, no research to date experimentally examines the relationship between ENDS use and alcohol use. The present study uses a two-session within-subjects design to examine 1) the relationship between ENDS use prime and attentional bias for alcohol related cues and 2) the relationship between ENDS use and laboratory ad libitum alcohol consumption. A total of N = 31 (mean age = 28.71, SD = 11.17; 45.2% women; 54.8% White/Caucasian) healthy users of ENDS who endorsed liking beer completed the present study, which included 1) a dot-probe and eye-tracking task that assessed attentional bias (reaction time, initial orientation, and delayed disengagement) to alcohol images following ENDS prime or no prime and 2) an ad libitum beer consumption task that assessed mL of beer consumed by the participants when concurrent use of ENDS was allowed or not allowed. All analyses controlled for age, race, and gender. Results of repeated measure ANCOVA’s indicate that attentional bias for alcohol does not differ between the ENDS prime or control conditions (F’s 0.01 to 0.12, ηp2’s 0.001 to 0.01). There is a large interactive effect of self-reported days of concurrent use of ENDS and alcohol over the last 14 days (ηp2’s 0.35 to 0.85), small to medium effects of alcohol craving preceding eye-tracking (ηp2’s 0.02 to 0.09), and small to medium effects of ENDS craving preceding eye-tracking (ηp2’s 0.06 to 0.13), all of which show increases in attentional bias following the ENDS prime; however, these results were limited by data quality issues that preclude strong support of these effects. Results of repeated measure ANCOVA’s demonstrate that amount of beer consumed does not differ by ENDS condition, F (4, 26) = 0.03, p = .86, ηp2 = 0.001. Results of a hierarchical linear regression show that amount of ENDS weight change (g) is not significantly related to mL of beer consumed in the ENDS session (b = -86.48, t = -0.90, p = 0.38, ∆R2 = 0.03). Results of linear mixed modeling testing the effect of ENDS puffs on alcohol sips temporally across the ad lib task show puffs are significantly related to sips (estimate = 0.23, SE = 0.07, p = .002) and number ENDS puffs account for some variability in slope of participant sips across participants. Results of repeated measure ANCOVA’s do not demonstrate significant interactions between mL of beer consumed by session and concurrent self-reported ENDS use over the past two weeks (ηp2 = 0.45), alcohol craving, or ENDS craving (ηp2’s = 0.002). Overall, results indicate that increased frequency of ENDS use is related to an increased frequency of beer consumption in real time. Since ENDS is related to alcohol use in time and place, individuals at risk for alcohol use problems should take care in their ENDS use. This study suggests that research should more fully measure and compare event-level and meta-level data on ENDS and alcohol use and that patterns based in the cigarette literature may not always generalize to ENDS.

Indiana University-Purdue University Indianapolis (IUPUI)
Cite As
Alternative Title
Conference Dates
Conference Host
Conference Location
Conference Name
Conference Panel
Conference Secretariat Location
Full Text Available at
This item is under embargo {{howLong}}