Energy Drinks and Alcohol: Links to Alcohol Behaviors and Consequences
Across 56 Days
Megan E. Patrick, Ph.D. a,*, and Jennifer L. Maggs, Ph.D. b
a Institute for Social Research, University of Michigan, Ann Arbor, Michigan
bDepartment of Human Development and Family Studies and Prevention Research Center, Pennsylvania State University, University Park, Pennsylvania
Article history: Received June 3, 2013; Accepted September 20, 2013
Keywords: Alcohol; Drinking; Energy drink; Blood alcohol; College; Consequences; Daily
Purpose: To examine short-term consequences associated with consuming alcohol and energy
drinks compared with consuming alcohol without energy drinks.
Methods: A longitudinal measurement-burst design (14-day bursts of daily surveys in four
consecutive college semesters) captured both within-person variation across occasions and
between-person differences across individuals. The analytic sample of late adolescent alcohol users
included 4,203 days with alcohol use across up to four semesters per person from 508 college
Results: Adding energy drink use to a given day with alcohol use was associated with an increase
in number of alcoholic drinks, a trend toward more hours spent drinking, elevated estimated blood
alcohol content (eBAC), a greater likelihood of subjective intoxication, and more negative consequences of drinking that day. After controlling for eBAC, energy drink use no longer predicted
subjective intoxication but was still associated with a greater number of negative consequences.
Conclusions: The consumption of energy drinks may lead to increases in alcohol consumption and,
after controlling for eBAC, negative consequences. Use of energy drinks plus alcohol represents an
emerging threat to public health.
2014 Society for Adolescent Health and Medicine. All rights reserved.
When late adolescents
consume energy drinks
and alcohol, they are more
likely to consume more
alcohol, become more
intoxicated, and experience more negative consequences compared with
when they consume only
alcohol. Prevention programs designed to reduce
the risks associated with
the consumption of energy drinks and alcohol
Consumption of energy drinks has risen markedly in the
United States since the introduction of Red Bull to the market in
1997 [1,2]. There are hundreds of brands on the market targeted
to young people, with names such as Full Throttle, Rockstar Energy, Monster, and Daredevil [2e5]. Caffeine content of energy
drinks can range from 50 mg to more than 500 mg per can or
bottle, compared with a 12-ounce soda that has 34e54 mg and a
6-ounce brewed coffee that has 77e150 mg [2,6]. The number of
emergency department visits resulting from energy drinks
doubled between 2007 and 2011 . Mixing alcohol with energy
drinks is an emerging trend [2,3,5], with mixed drinks such as
vodka Red Bull and JÃ¤ger bombs (i.e., dropping a shot of the liquor JÃ¤germeister into a glass of Red Bull) becoming popular
among youth [4,8]. The public health implications include not
only physiological risks to individuals (e.g., blacking out, alcohol
poisoning), but also exposing the community to dangerous situations resulting from young adults who may be â€œwide awake
drunkâ€ after a night of partying . Although research on the
public health impact of energy drinks and alcohol is emerging,
there is currently very little research using repeated measures
designs or assessing consumption in naturalistic settings (i.e.,
outside of laboratories). The purpose of this study is to examine
the short-term consequences associated with consuming energy
drinks and alcohol compared with consuming only alcohol
among college students.
Disclaimer: The content here is solely the responsibility of the authors and does
not necessarily represent the official views of the sponsors.
* Address correspondence to: Megan E. Patrick, Ph.D., Institute for Social
Research, University of Michigan, 426 Thompson Street, Ann Arbor, MI 48106-
E-mail address: [email protected] (M.E. Patrick).
1054-139X/$ e see front matter 2014 Society for Adolescent Health and Medicine. All rights reserved.
Journal of Adolescent Health 54 (2014) 454e459
Energy drinks are primarily marketed to adolescents and
young adults [5,9]. Among adolescents and young adults, about
30%e50% consume energy drinks , based on regional samples. Among college students, 40%e60% report using energy
drinks in the past month, with 10% classified as high-frequency
users (52Ã¾ days in the prior year) . Mixing energy drinks
with alcohol is common; one study estimated 24% of college
students had done so in the past month . Among energy drink
users, 54% report using them with alcohol while partying, and
consumption of three or more alcoholic energy drinks during an
evening is common . Very little is known, however, about
behaviors and consequences on specific days on which students
consume both alcohol and energy drinks.
Regulation of alcohol and energy drinks
In November 2010, the U.S. Food and Drug Administration 
(FDA) issued warning letters to four manufacturers of caffeinated
alcoholic beverages that collectively sold products including Four
Loko, Core High Gravity, Moonshot, Joose, and Max. The FDA
stated that based on a scientific review, caffeinated alcoholic
beverages presented a public health concern and that adding
caffeine to malt alcohol beverages was an â€œunsafe food additiveâ€
and in violation of the Federal Food, Drug, and Cosmetic Act .
The FDA action effectively prohibited the sale of premixed alcoholic energy drinks . However, such premixed drinks represented only a small portion of those consumed, and the mixing of
alcohol and energy drinks is expected to continue [6,14] as bar
patrons are still free to order alcohol mixed with energy drinks by
bartenders  or by mixing their own. Despite the curtailment of
sales of premixed energy drinks with alcohol, the relative lack of
regulation of energy drinks more generally has led to vigorous
marketing campaigns by producers making unsubstantiated
claims that they enhance performance, increase attention, and
reduce effects of fatigue and alcohol . Sales of high-caffeine
energy drinks without alcohol has continued to rise [16,17].
Effects of energy drinks on behavior and consequences
Caffeine may reduce ability to accurately judge the intoxicating effects of alcohol. Laboratory research shows that participants with the same blood alcohol concentration (BAC) tend
to report subjectively lower intoxication when energy drinks and
alcohol were consumed compared with alcohol alone [1,18,19];
however, there is also evidence that intoxication rates are not
objectively lower in caffeine plus alcohol study conditions .
Laboratory research on this topic has used small and nonrepresentative samples; however, this early experimental work is key
because it provides some of the only available evidence of the
links between energy drinks and consequences. In one study,
participants who received alcohol mixed with an energy drink
reported lower intoxication, including headache, weakness, dry
mouth, and reduced motor coordination, compared with those in
the alcohol-only condition. Importantly, no differences were
observed in actual impairment in motor coordination or reaction
time resulting from intoxication . In a second study, participants consuming alcohol with energy drinks reported lower
subjective intoxication compared with alcohol-only participants,
and caffeine counteracted some cognitive effects of alcohol (e.g.,
response speed) but not others (e.g., response accuracy), showing
the complexity of the drug interaction . A third study
concluded that consuming alcohol with energy drinks led to
more impairment in behavioral inhibition, although response
activation was not as impaired as in the alcohol-only condition
. Finally, consuming caffeine with alcohol did not counteract
the negative effects of alcohol on driving in a simulator [14,21]. In
sum, if drinkers perceive themselves as higher functioning and
less intoxicated when they also consume caffeine, though
caffeine does not have these effects, it follows that risks for
injury, aggression, and impaired decision-making may increase
Indeed, in a between-person event-level study in a naturalistic setting, bar patrons who consumed alcohol and energy
drinks were more likely to leave the bar highly intoxicated (BAC
.08%), intend to drive, leave the bar later, drink for a longer
period, and consume more total alcohol .
Prior survey and interview research regarding alcohol and
energy drink use has typically collected data only once, asking
respondents about their typical or past month/year behavior. The
limited work thus far suggests that college students who
consume energy drinks tend to consume more alcohol and
experience more alcohol-related consequences than students
who do not consume energy drinks [4,11,22,23]. Others have
questioned the causes underlying such findings, in part because
an individualâ€™s personality or sensation seeking may lead to both
heavier energy drink use and heavier alcohol use . Such
studies also typically do not assess whether alcohol and energy
drinks are consumed on the same days, nor do they examine the
consequences associated with adding energy drinks to alcohol.
As a result, little is known about the daily-level consequences of
alcohol plus energy drink use. The present study uses repeated
measures survey data (i.e., up to 56 daily reports per person),
which allows us to compare days an individual consumes energy
drinks and alcohol to days he or she consumes only alcohol. This
strategy is necessary to identify immediate consequences associated with alcohol plus energy drink use, controlling for stable
characteristics of the individual. In addition, we examine the
association between energy drink use and alcohol-related consequences after controlling for alcohol use in order to examine
whether energy drink use has a direct association with alcoholrelated consequences that is independent of the amount of
The present study was designed to examine whether, on
drinking days, the level of alcohol use, subjective intoxication,
and consequences differed as a function of energy drink use.
Specifically, we hypothesized that on days with greater energy
drink use (compared with days of alcohol use alone): (1) alcohol
use would be greater (i.e., students would consume a greater
number of drinks, spend more time drinking, or reach higher
peak levels of intoxication [elevated estimated blood alcohol
content; eBACs]); (2) subjective drunkenness would be less
likely; and (3) negative alcohol consequences would be greater.
In subsequent analyses controlling for eBAC to examine the
unique effect of energy drinks above and beyond alcohol consumption, we hypothesized that on days with greater energy
drink use: (4) subjective drunkenness would be less likely and
(5) negative consequences would be greater.
M.E. Patrick and J.L. Maggs / Journal of Adolescent Health 54 (2014) 454e459 455
The University Life Study (ULS [25,26]) used a measurementburst design to capture within-person variation across multiple
days and between-person differences in change across college
semesters. Each semester for seven semesters, beginning in the
first semester of college, students completed a semester survey
followed by 14 consecutive daily surveys. Data on alcohol and
energy drinks are available from four 14-day bursts of daily
surveys (i.e., 56 days total), from spring of studentsâ€™ second year
(spring 2009) to fall of fourth year (fall 2010). Daily surveys were
programmed so that the day being reported on was clear on each
relevant screen (e.g., â€œOn [Wednesday], did you.â€). Days were
defined as â€œfrom the time you woke up to the time you went to
sleepâ€ to reflect student schedules rather than calendar dates.
Reminder e-mail messages were sent to students who did not
complete the surveys. All procedures were approved by the
institutional review board.
In total, 744 students (65.6%) completed the semester 1 baseline survey and provided informed consent. Mean age was 18.9
years (standard deviation Â¼ .42). A stratified random sampling
procedure was used to achieve a diverse sample with respect to
gender and race/ethnicity. Sampling groups had response rates
from 55.2% to 76.0%. With respect to race/ethnicity, 25.2% of the
ULS sample self-reported as Hispanic/Latino American, 27.3% of
the sample self-reported as white/European American nonHispanic, 23.0% Asian American/Pacific Islander non-Hispanic,
15.6% black/African American non-Hispanic, and 8.7% reported
more than one race non-Hispanic. Of the initial 744 students, 652
completed semester 4 (88%). There were no differences between
those who completed semester 4 and those who did not (n Â¼ 92)
on age, race/ethnicity, age of alcohol onset, or past month alcohol
use at semester 1; however, women were more likely to remain in
the study at semester 4 than men. Completion rates of the daily
surveys were high, with 79%e86% of each semesterâ€™s participants
completing at least 12 of 14 daily surveys. We focus exclusively on
days students reported alcohol use, therefore excluding students
who abstained from alcohol on all sampled days (n Â¼ 134). In
addition, 10 people were excluded because of missing data at
levels 2 or 3. The analytic sample included 4,203 days with alcohol
use (level 1), 1,135 person semesters (level 2), and 508 people
Each day, students were asked, â€œOn [Friday], how many (1)
high energy (caffeinated) drinks like Red Bull, not containing
alcohol did you drink? (2) high-energy drinks with alcohol (e.g.,
Red Bull Ã¾ vodka, or a premixed drink) did you drink?â€ A pulldown menu permitted responses of 0e25Ã¾. Responses from
these two questions were summed and coded as a four-level
ordinal variable: 0, 1, 2, or 3Ã¾ energy drinks consumed that day.
Each day, participants were asked the number of alcoholic
drinks they consumed the prior day and what time they started
their first drink and ended their last drink. Gender, body weight,
and length of time drinking were used to calculate eBAC . In
addition, each day respondents were asked whether they got
drunk (yes/no) as a measure of subjective intoxication. Consequences of alcohol use each day were summed based on responses to the prompt, â€œAs a result of drinking on [Friday], did
you.â€ with a response of yes or no to each of 10 negative consequences (e.g., have a hangover, get in trouble ). Days were
coded as weekends (Thursday to Saturday) or weekdays (Sunday
to Wednesday), based on alcohol patterns among college students [29e32].
Each semester, students were asked whether they lived in
fraternity/sorority housing (yes Â¼ 1, no Â¼ 0).
Plan of analysis
Research questions examined whether days with more energy drink consumption were associated with (1) increases in
alcohol use (i.e., a greater number of drinks, a longer time spent
drinking, or higher eBACs); (2) a lower likelihood of reporting
subjective intoxication; and (3) more negative consequences.
Two additional models predicting (4) subjective drunkenness
and (5) negative alcohol-related consequences controlled for
eBAC at level 1.
Three-level hierarchical linear modeling (HLM ) was used
to separate variance associated with the person (level 3), the
semester (level 2), and the day (level 1). Person-level (level 3)
variables included gender and the personâ€™s mean energy drink
use across all assessed days. Semester-level (level 2) controls
included the linear effect of semester, fraternity/sorority membership that semester, and the mean energy drink use across
days in the semester. Daily-level (level 1) variables included
weekend (vs. weekday) and, the primary predictor of interest,
the number of energy drinks consumed that day. Interactions
between the daily number of energy drinks consumed and
gender were also tested (not shown); none was significant. The
five dependent variables were number of alcohol drinks
consumed, hours spent drinking, eBAC, subjective drunkenness,
Descriptive statistics on days with any drinks
M SD Actual range
Level 3: Person-level constructs
Male gender .46 .50 0e1
Person mean energy drinksa .08 .16 0e1.19
Level 2: Semester-level constructs
Fraternity/sororityb .09 .28 0e1
Semester mean energy drinksc .09 .22 0e1.83
Semester mean eBACd,e .09 .07 0e.50
Level 1: Daily-level constructsf
Energy drinksg .16 .57 0e3
Alcohol drinks, number (all days) .95 2.76 0e26
Alcohol drinks, number (drinking days) 6.07 4.21 1e26
eBACe .10 .085 0e.50
Hours spent drinking alcohol 4.33 3.82 0e22.25
Subjective drunkenness .51 .50 0e1
Negative alcohol consequences 1.01 1.67 0e10
Level 1 N Â¼ 4,128e4,203 person days; level 2 N Â¼ 1,335 person semesters; level 3
N Â¼ 508 people.
eBAC Â¼ estimated blood alcohol content; M Â¼ mean; SD Â¼ standard deviation. a One score per person, averaged across all sampled days and all semesters. b One score per person in each semester, indicating whether a member that
semester. c One score per person in each semester, representing the average energy
drink use across all sampled days that semester.
d One score per person in each semester, representing the average eBAC across
all sampled days that semester.
e When eBAC was used as a dependent variable, it was multiplied by 100 to
facilitate analyses using a Poisson distribution. f All daily-level constructs consist of one score per person per sampled day. g 0 Â¼ none, 1 Â¼ 1, 2 Â¼ 2, and 3 Â¼ 3 or more energy drinks.
456 M.E. Patrick and J.L. Maggs / Journal of Adolescent Health 54 (2014) 454e459
and negative alcohol-related consequences on a given day. HLM
with a Poisson distribution was used for all outcomes except for
the dichotomous variable subjective drunkenness, for which
HLM with a Bernoulli distribution was used.
In the ULS sample, 80.4% of people used alcohol on at least
one of the up to 56 sampled days across four semesters, 51.3% of
people used energy drinks at least one time across sampled days,
and 30.5% of people used energy drinks and alcohol on the same
day at least once across sampled days. On days students used
energy drinks, they also used alcohol on 31.6% of days. On days
students used alcohol, they also used energy drinks on 9.6% of
days. Descriptive statistics for all variables in the analytic sample
are shown in Table 1.
Energy drinks predicting number of alcoholic drinks, hours spent
drinking, and eBAC
Results for the HLM analyses predicting number of alcoholic
drinks, hours spent drinking alcohol, and eBAC are shown in
Table 2. Control variables are described first. At the person-level
(level 3), men consumed a greater number of drinks but also
spent more hours drinking than women and reached lower peak
eBACs. Individuals who in general consumed more energy drinks
(averaged across all days and semesters) also in general
consumed a greater number of alcoholic drinks, and had a trend
(p < .10) toward spending more time drinking alcohol. At the
semester-level (level 2), the number of alcoholic drinks
consumed did not change from second to fourth year of college,
but there were increases across semesters in hours spent
drinking alcohol and decreases in peak eBAC. During semesters
students were living in fraternity/sorority housing, they reported
higher levels of all three drinking indicators. Fluctuations in
energy drink use across semesters (averaged across days within a
semester) did not predict variation in the three alcohol-related
outcome variables. At the daily-level (level 1), all three
indicators of alcohol consumption were higher on weekend days
Consistent with our primary hypotheses for the three indicators
of alcohol use, independent of all the associations mentioned
previously, on days students consumed more energy drinks they
consumed a greater number of alcohol drinks, spent more hours
drinking (at trend [p < .10] level significance), and reached higher
peak eBAC compared with days with only alcohol use.
Energy drinks predicting subjective drunkenness and negative
Models predicting subjective drunkenness and negative
alcohol-related consequences are shown in Table 3. First, models
with the same predictors as those in Table 2 examined the extent
to which energy drink use predicted subjective intoxication and
negative consequences. Control variables indicated that, at the
person level (level 3), men and women were equally likely to
report having gotten drunk, but men reported more negative
consequences. Students who in general consumed more energy
drinks reported more negative alcohol consequences, but had no
greater likelihood of reporting that they got drunk. Across semesters (level 2), getting drunk and negative consequences
decreased from second to fourth year of college. During semesters that students lived with fraternities/sororities, they reported
more subjective drunkenness but no differences in consequences. Students were more likely to report getting drunk in
semesters when their energy drink use was higher (averaged
across days per semester). At the daily level (level 1), drunkenness was more likely and negative consequences greater on
weekends than weekdays.
Addressing our primary hypotheses, students had greater odds
of reporting getting drunk and experienced more negative alcohol
consequences on days they consumed more energy drinks.
Energy drinks predicting subjective drunkenness and negative
consequences, controlling for eBAC
Next, eBAC was added as an additional control at the daily
level to isolate the associations between energy drinks and
Multilevel models predicting alcohol use based on daily-level energy drink use
Hours drinking alcohol
Level 3: Person-level
Intercept 3.22 (2.96e3.51)**** 7.32 (6.63e8.09)**** 9.86 (8.90e10.92)****
Male gender 1.34 (1.23e1.46)**** .82 (.73e.92)*** 1.09 (1.01e1.18)**
Person mean energy drinks 1.49 (1.07e2.08)** 1.41 (.93e2.14) 1.30 (.96e1.76)*
Level 2: Semester-level
Semester (linear) .99 (.97e1.01) .94 (.92e.97)**** 1.03 (1.01e1.06)**
Fraternity/sorority 1.16 (1.05e1.28)*** 1.16 (1.01e1.32)** 1.15 (1.03e1.28)**
Semester mean energy drinks 1.03 (.92e1.15) 1.05 (.90e1.22) .99 (.85e1.16)
Level 1: Daily-level
Weekend 1.42 (1.34e1.51)**** 1.33 (1.24e1.42)**** 1.54 (1.42e1.66)****
Energy drinks 1.11 (1.07e1.15)**** 1.13 (1.07e1.19)**** 1.05 (1.00e1.12)*
Level 1 N Â¼ 4,128e4,203 person days; level 2 N Â¼ 1,335 person semesters; and level 3 N Â¼ 508 people. eBAC and time drinking alcohol values were transformed to
integer values for analysis to comply with assumptions of the Poisson model that dependent variables be integers representing count variables; raw eBAC values were
multiplied by 100 and time spent drinking values were multiplied by 4 (because it was reported in 15-minute increments).
CI Â¼ confidence interval; eBAC Â¼ estimated blood alcohol content; ERR Â¼ event rate ratio (from Poisson hierarchical linear modeling).
* p < .10.
** p < .05.
*** p < .01.
**** p < .001.
M.E. Patrick and J.L. Maggs / Journal of Adolescent Health 54 (2014) 454e459 457
outcomes, independent of the amount of alcohol consumed.
Once eBAC was controlled at the daily level, there was no association between energy drink consumption and subjective
intoxication, although we had hypothesized there would be a
negative association. In support of our hypotheses, students reported experiencing more negative alcohol consequences on
days they consumed more energy drinks, even with eBAC on that
day included in the model.
This study was the first to examine consequences associated
with alcohol and energy drink use with daily-level, within-person data. On days when students consumed energy drinks and
alcohol, compared with days when they drank alcohol but no
energy drinks, they drank more alcoholic drinks, reached a
higher eBAC, and showed a trend toward spending more time
drinking. Once these differential levels of intoxication (eBAC)
were controlled, students did not report feeling more intoxicated
on drinking days that they consumed energy drinks, but they did
report experiencing more negative alcohol-related consequences. That is, consistent with our hypotheses, energy drink
use was associated with greater alcohol use and more negative
consequences but, contrary to our hypothesis, not with greater
subjective intoxication after controlling for actual use.
The intoxicated state resulting from consuming high-caffeine
energy drinks with alcohol has been described as â€œwide-awake
drunkâ€ because caffeine is presumed to attenuate some of alcoholâ€™s sedative effects . The present study provides limited
support for this phenomenon. However, the increased alcohol
consumption on days with energy drink use may also support
the process of alcohol priming, such that energy drink consumption may increase motivations to drink more alcohol .
Policy and intervention efforts that seek to mitigate harms
associated with alcohol and energy drinks may take several approaches. With respect to reducing sales, the United States no
longer permits manufacturers to premix high-caffeine products
with alcohol ; however, energy drinks remain widely
available so individuals, hosts, and bartenders (in the United
States and many but not all other countries) are free to mix these
drinks [15,34]. With respect to packaging, creating common labeling standards for high-caffeine and high-alcohol products
similar to those regarding nutritional information on food and/or
units on alcohol (e.g., Federal Trade Commissionâ€™s new rules for
FourLoko high-alcohol beverages ) would provide information to consumers wishing to minimize harm. With respect to
health education, drinkers could be made aware of the discrepancy between subjective and objective intoxication, and the
increased risk for negative consequences when combining
alcohol with high doses of caffeine.
These results also need to be interpreted in the context of a
number of limitations. First, the sample consisted of a single
cohort of students from a single university in the United States,
limiting generalizability. Second, no physiological measures of
intoxication were available, thus we relied on the calculation of
eBAC based on a formula including gender, weight, number of
drinks, and time spent drinking . Such measures provide
reasonably accurate estimates of breath alcohol concentrations
 and are necessary in studies designed not to interfere with
usual behavior in natural settings, but contain important measurement error [37,38]. Third, the measure of subjective intoxication was dichotomous and thus lacked sensitivity for capturing
information about intensity of drunkenness. Fourth, the validity
of self-reports of energy and alcoholic drinks depends on accurate recall and willingness to report behavior. Recall was facilitated by asking questions daily, but of course also limited by
intoxicationdas in all studiesdon heavy drinking days. Fifth, the
time of day of energy drink consumption was not reported and
other caffeinated products were not assessed; however, these
limitations likely served to attenuate the effects of energy drinks.
In addition, the half-life of caffeine is about 4e5 hours, or longer
for higher caffeine consumption [39,40], making it very likely
that the pharmacological effects of caffeine and alcohol overlapped on days students used both energy drinks and alcohol,
whether or not they were consumed in the same container or
sitting. Finally, although the within-person analyses by design
Multilevel models predicting drunkenness and alcohol consequences based on daily-level energy drink use
Daily-level eBAC not controlled Daily-level eBAC controlled
Level 3: Person-level
Intercept .45 (.35e.60)*** .52 (.43e.62)*** .65 (.48e.88)** .52 (.43e.62)***
Male gender .88 (.67e1.15) 1.25 (1.03e1.52)* .88 (.64e1.20) 1.27 (1.04e1.55)*
Person mean energy drinks 1.97 (.81e4.75) 2.70 (1.47e4.96)** 2.07 (.70e6.08) 2.60 (1.40e4.81)**
Level 2: Semester-level
Semester (linear) .87 (.81e.94)*** .93 (.89e.97)** .84 (.77e.92)*** .93 (.89e.97)**
Fraternity/sorority 2.46 (1.52e3.99)*** 1.07 (.81e1.41) 2.81 (1.57e5.03)** 1.06 (.80e1.42)
Semester mean energy drinks 1.68 (1.07e2.62)* .98 (.74e1.29) 1.68 (.93e2.99) .96 (.71e1.30)
Level 1: Daily-level
Weekend 2.48 (2.00e3.07)*** 1.32 (1.16e1.52)*** 1.82 (1.47e2.25)*** 1.16 (1.02e1.32)*
Energy drinks 1.51 (1.28e1.79)*** 1.27 (1.18e1.38)*** 1.18 (.97e1.45) 1.14 (1.04e1.25)**
eBAC d d 1.24 (1.21e1.27)*** 1.07 (1.06e1.08)***
Level 1 N Â¼ 4,128e4,203; level 2 N Â¼ 1,335 person semesters; and level 3 N Â¼ 508 people. eBAC and time drinking alcohol values were transformed to integer values for
analysis to comply with assumptions of the Poisson model that dependent variables be integers representing count variables; raw eBAC values were multiplied by 100
and time spent drinking values were multiplied by 4 (because it was reported in 15-minute increments).
CI Â¼ confidence interval; eBAC Â¼ estimated blood alcohol content; ERR Â¼ event rate ratio (from Poisson hierarchical linear modeling); OR Â¼ odds ratio (from Bernoulli
hierarchical linear modeling).
* p < .05.
** p < .01.
*** p < .001.
458 M.E. Patrick and J.L. Maggs / Journal of Adolescent Health 54 (2014) 454e459
control for all time-stable confounding factors (such as family
background, precollege drinking history, sensitivity to caffeine),
additional fluctuating factors (such as tiredness, food consumed
before drinking, goals for the night, drinking partners) remain
possible time-varying causes of observed associations.
In a recent editorial, Attwood argued that although the public
health concern on this topic may be justified, â€œwe should be
careful that preoccupation with the caffeine/alcohol debate does
not divert attention from the more pressing issue of harmful
alcohol consumptionâ€ . Investigating the public health
impact of energy drink plus alcohol use, compared with the
public health impact of alcohol alone, will be an important step
for documenting the scope of the problem. Results here suggest
that energy drink consumption is, in fact, a new risk factor for
heavier and harmful alcohol consumption that should be
examined in the context of other person-level and situational
factors that predict dangerous drinking. Future research should
build on the strengths of this studyemost notably the collection
of within-person data comparing different types of drinking
dayseto replicate and extend the results in samples that are
more diverse with respect to age, social role status, drinking
culture, and country of origin. More detailed measurement of
daily drinking motivations, academic and work demands, and
sleep patterns would provide a more nuanced picture of how
individuals strategically aim to use both caffeine and alcohol for
varied social and achievement goals.
The National Institute on Alcohol Abuse and Alcoholism
funded the University Life Study (R01 AA019606 to J.L.M.) and
the preparation of this manuscript (R21 AA021426 to M.E.P.).
 Marczinski CA, Fillmore MT, Bardgett ME, Howard MA. Effects of energy
drinks mixed with alcohol on behavioral control: Risks for college students
consuming trendy cocktails. Alcohol Clin Exp Res 2011;35:1282e92.
 Reissig CJ, Strain EC, Griffiths RR. Caffeinated energy drinksea growing
problem. Drug Alcohol Depend 2009;99:1e10.
 Clauson KA, Shields KM, McQueen CE, Persad N. Safety issues associated
with commercially available energy drinks. J Am Pharm Assoc 2008;48:
e55e63; quiz e64e7.
 Oâ€™Brien MC, McCoy TP, Rhodes SD, et al. Caffeinated cocktails: Energy drink
consumption, high-risk drinking, and alcohol-related consequences among
college students. Acad Emerg Med 2008;15:453e60.
 Simon M, Mosher J. Alcohol, energy drinks, and youth: A dangerous mix.
San Rafael, CA: Marin Institute; 2007.
 Arria AM, Oâ€™Brien MC. The â€œhighâ€ risk of energy drinks. JAMA 2011;305:
 Substance Abuse and Mental Health Services Administration, Center for
Behavioral Health Statistics and Quality. The DAWN Report: Update on
emergency department visits involving energy drinks: A continuing public
health concern. Rockville, MD: SAMHSA; January 10, 2013.
 Thombs DL, Oâ€™Mara RJ, Tsukamoto M, et al. Event-level analyses of energy
drink consumption and alcohol intoxication in bar patrons. Addict Behav
 Babu KM, Church RJ, Lewander W. Energy drinks: The new eye-opener for
adolescents. Clin Pediatr Emerg Med 2008;9:35e42.
 Seifert SM, Schaechter JL, Hershorin ER, Lipshultz SE. Health effects of
energy drinks on children, adolescents, and young adults. Pediatrics 2011;
 Arria AM, Caldeira KM, Kasperski SJ, et al. Energy drink consumption and
increased risk for alcohol dependence. Alcohol Clin Exp Res 2011;35:365e75.
 Malinauskas BM, Aeby VG, Overton RF, et al. A survey of energy drink
consumption patterns among college students. Nutr J 2007;6:35.
 FDA News Release. FDA warning letters issued to four makers of caffeinated alcoholic beverages. Nov 2010. Available at: http://www.fda.gov/
 Howland J, Rohsenow DJ, Arnedt JT, et al. The acute effects of caffeinated
versus non-caffeinated alcoholic beverage on driving performance and
attention/reaction time. Addiction 2011;106:335e41.
 Rossheim ME, Thombs DL. Artificial sweeteners, caffeine, and alcohol
intoxication in bar patrons. Alcohol Clin Exp Res 2011;35:1891e6.
 Energy Drinks and Shots: U.S. Market Trends. Packaged facts, Jan 29, 2013.
Available at: http://www.packagedfacts.com/Energy-Drinks-Shots-7124908/.
 Knowles D. Marketed to teens, energy drink sales poised to nearly double over
the nextfive years, despite healthwarnings from pediatricians. New York Daily
News. Available at: http://www.nydailynews.com/news/national/energydrink-sales-soar-years-article-1.1263491#ixzz2UdG8hSNf; Feb 13, 2013.
 Ferreira SE, de Mello MT, PompÃ©ia S, de Souza-Formigoni ML. Effects of
energy drink ingestion on alcohol intoxication. Alcohol Clin Exp Res 2006;
 Marczinski CA, Fillmore MT. Clubgoers and their trendy cocktails: Implications of mixing caffeine into alcohol on information processing and
subjective reports of intoxication. Exp Clin Psychopharmacol 2006;14:
 Fillmore MT, Roach EL, Rice JT. Does caffeine counteract alcohol-induced
impairment? The ironic effects of expectancy. J Stud Alcohol 2002;63:
 Liguori A, Robinson JH. Caffeine antagonism of alcohol-induced driving
impairment. Drug Alcohol Depend 2001;63:123e9.
 Price SR, Hilchey CA, Darredeau C, Fulton HG, Barrett SP. Energy drink coadministration is associated with increased reported alcohol ingestion.
Drug Alcohol Rev 2010;29:331e3.
 Velazquez CE, Poulos NS, Latimer LA, Pasch KE. Associations between energy drink consumption and alcohol use behaviors among college students.
Drug Alcohol Depend 2012;123:167e72.
 Verster JC, Aufricht C, Alford C. Energy drinks mixed with alcohol:
Misconceptions, myths, and facts. Int J Gen Med 2012;5:187e98.
 Patrick ME, Maggs JL. College studentsâ€™ evaluations of alcohol consequences as positive and negative. Addict Behav 2011;36:1148e53.
 Small M, Bailey-Davis L, Morgan N, Maggs J. Changes in eating and physical
activity behaviors across seven semesters of college: Living on or off
campus matters. Health Educ Behav 2013;53:300e2.
 Matthews DB, Miller WR. Estimating blood alcohol concentration: Two
computer programs and their applications in therapy and research. Addict
 Patrick ME, Maggs JL. Short-term changes in plans to drink and importance
of positive and negative alcohol consequences: Between- and withinperson predictors. J Adolesc 2008;31:307e21.
 Del Boca FK, Darkes J, Greenbaum PE, Goldman MS. Up close and personal:
Temporal variability in the drinking of individual college students during
their first year. J Consult Clin Psychol 2004;72:155e64.
 Hartzler B, Fromme K. Heavy episodic drinking and college entrance. J Drug
 Oâ€™Connor RM, Colder CR. Predicting alcohol patterns in first-year college
students through motivational systems and reasons for drinking. Psychol
Addict Behav 2005;19:10e20.
 Maggs JL, Williams LR, Lee CM. Ups and downs of alcohol use among firstyear college students: Number of drinks, heavy drinking, and stumble and
pass out drinking days. Addict Behav 2011;36:197e202.
 Raudenbush SW, Bryk AS. Hierarchical linear models: Applications and
data analysis methods. 2nd ed. Thousand Oaks, CA: Sage; 2002.
 Attwood AS. Caffeinated alcohol beverages: A public health concern.
Alcohol Alcohol 2012;47:370e1.
 Marczinski CA, Fillmore MT, Henges AL, Ramsey MA, Young CR. Mixing an
energy drink with an alcoholic beverage increases motivation for more
alcohol in college students. Clin Exp Res 2013;37:276e83.
 Tepper R. Four lokoâ€™s new packaging will require â€˜alcohol factsâ€™ label and
resealable containers. Huffington Post. Available at: http://www.huffington
 Hustad JTP, Carey KB. Using calculations to estimate blood alcohol concentrations for naturally occurring drinking episodes: A validity study.
J Stud Alcohol Drugs 2005;66:130e8.
 Clapp JD, Min JW, Trim RS, et al. Predictors of error in estimates of blood
alcohol concentration: A replication. J Stud Alcohol Drugs 2009;70:683e8.
 Kaplan GB, Greenblatt DJ, Ehrenberg BL, et al. Dose-dependent pharmacokinetics and psychomotor effects of caffeine in humans. J Pharmacokinet
 Mandel HG. Update on caffeine consumption, disposition and action. Food
Chem Toxicol 2002;40:1231e4.
M.E. Patrick and J.L. Maggs / Journal of Adolescent Health 54 (2014) 454e459 459