Alcohol mixed with energy drinks are robustly associated with patterns of problematic alcohol consumption among young adult college students

Addictive Behaviors 41 (2015) 136–141

Contents lists available at ScienceDirect

Addictive Behaviors

Alcohol mixed with energy drinks are robustly associated with patterns of problematic alcohol consumption among young adult college students Daniel J. Snipes ⁎, Amy J. Jeffers, Brooke A. Green, Eric G. Benotsch Virginia Commonwealth University, United States

H I G H L I G H T S • Alcohol mixed with energy drinks (AmEDs) are related to alcohol dependence. • Consumers of AmEDs are not higher in sensation seeking than alcohol only consumers. • AmED use robustly predicts alcohol dependence over sensation seeking and impulsivity.

a r t i c l e

i n f o

Available online 20 October 2014 Keywords: AmED Young adult AUDIT Energy drinks Alcohol

a b s t r a c t Background: Young adults are a population at great risk for problematic health behaviors. Alcohol mixed with energy drink (AmED) consumption is a relatively popular health risk behavior among young adults. AmED consumption continues to illustrate negative outcomes in the research literature, having been linked with other substance use, high-risk sexual behavior, and sexual victimization. Limited research to date has examined associations between AmED consumption and patterns of alcohol dependence. Methods: Undergraduate college students (n = 757) filled out an online survey which assessed their drinking habits in the past week and month, including their consumption of AmED beverages, personality characteristics, substance use, and problematic alcohol consumption via the Alcohol Use Disorders Identification Test (AUDIT). Results: A minority of participants reported AmED consumption in both the past month (11.6%) and past week (9.7%). Compared to their alcohol-only drinking counterparts, AmED consumers scored significantly higher on measures of impulsivity, and lower on anxiety sensitivity when compared to their alcohol-only drinking counterparts. In multivariate analyses, AmED consumption was robustly associated with patterns of alcohol dependence (AUDIT score ≥ 8) among young adult college students, while controlling for energy drink use, alcohol use, personality factors, substance use, and demographic variables. Conclusions: AmED consumption in the past month is robustly associated with problematic alcohol consumption. The present study describes harmful outcomes associated with AmED consumption, and extends the literature on the combined effects of alcohol and energy drinks on young adult risk behaviors. Further research needs to address causal mechanisms for the AmED and problematic alcohol consumption relation. © 2014 Published by Elsevier Ltd.

1. Introduction Energy drinks are highly caffeinated beverages marketed to increase energy, stamina, and wakefulness. Energy drinks are commonly marketed to young adults (Heckman, Sherry, Mejia, & Gonzalez, 2010), which may contribute to greater normativity for consumption. Alcohol mixed with energy drinks (AmED) has become a relatively common trend in alcohol consumption behavior among college students, and research suggests that between 24 and 40% of college students report past month consumption (Arria & O'Brien, 2011; Arria et al., 2011; Brache & Stockwell, 2011; ⁎ Corresponding author at: 806 W. Franklin St., Richmond, VA 23284, United States. Tel.: +1 804 828 1193; fax: +1 804 828 2237. E-mail address: [email protected] (D.J. Snipes).

http://dx.doi.org/10.1016/j.addbeh.2014.10.010 0306-4603/© 2014 Published by Elsevier Ltd.

Velasquez, Poulos, Latimer, & Pasch, 2012; Snipes & Benotsch, 2013). These drinks gained significant media attention when the FDA issued a letter to manufacturers of AmED beverages that caffeine was not a “generally recognized as safe” additive to alcoholic beverages, requesting the removal of caffeine from their alcohol products (FDA, 2010). This decision was predicated on numerous studies, which have linked AmED consumption with negative outcomes. 1.1. AmED and risk outcomes Consumption of AmED beverages has been shown to be relatively risky in terms of health behaviors, having been linked to engaging in high-risk and casual sex (Snipes & Benotsch, 2013; Miller, 2012), intending to drive after drinking (Thombs et al., 2010), and being taken

D.J. Snipes et al. / Addictive Behaviors 41 (2015) 136–141

advantage of sexually (O'Brien, Arria, Howland, James, & Marczinski, 2011; Snipes, Green, Javier, Perrin, & Benotsch, 2014). Consuming AmED has been shown to make an individual more likely to underestimate their impairment from alcohol, consume a greater number of alcoholic drinks, and drink to higher blood alcohol concentrations than alcohol-only drinkers (Ferreira, de Mello, Pompeia, & de SouzaFormigoni, 2006; O'Brien, McCoy, Rhodes, Wagoner, & Wolfson, 2008; Thombs et al., 2010). 1.2. AmED and patterns of alcohol dependence AmED consumption may also be a risk factor for a pattern of alcohol dependence. Consuming AmED beverages has been shown to increase the desire for more alcohol, more so than alcohol-only beverages (Marczinski, Fillmore, Henges, Ramsey, & Young, 2013). Some indirect research on AmED and high-risk drinking has generally supported a relation between AmED consumption and patterns of alcohol dependence. For example, research has shown associations between AmED consumption and binge drinking and other high-risk drinking behaviors (Brache & Stockwell, 2011; O'Brien et al., 2008; Woolsey, Waigandt, & Beck, 2010). However, examinations of specific patterns of alcohol dependence are generally lacking. A study examining work place drinking investigated the role of AmED consumption in alcohol dependence among the Taiwanese working population (Cheng, Cheng, Huang, & Chiou-Jong, 2012). While AmED consumption was not a primary focus in their study, they found that patterns of alcohol dependence (defined as ≥2 on the CAGE questionnaire) had a high prevalence (38.7% for men, 23.3% for women) among those who reported consuming AmED beverages (Cheng et al., 2012). While no other information (including significance tests) was provided in their study, it provides an impetus for suggesting that there may be a connection between AmED consumption and patterns of alcohol dependence. A more recent study by Lau-Barraco, Milletich, and Linden (2013) examined caffeinated alcoholic beverages (CABs) and alcohol severity (defined by Alcohol Use Disorders Identification Test [AUDIT] scores greater than 8). Lau-Barraco et al. (2013) divided participants into low/high groups in terms of alcohol use and CAB use. Their analysis found that high CAB/high alcohol consumers had significantly higher AUDIT scores than their low alcohol/low CAB group. However, measuring CAB instead of AmED drinking behavior is a more liberal estimate of a different type of drinking behavior. CAB beverages may, or may not, include energy drinks, which is critical as energy drinks may contain greater amounts of caffeine than most caffeinated sodas, as well as additional ingredients such as taurine, which has been shown to interact with alcohol (Olive, 2002). Additionally, their analysis was focused on examining different groups of alcohol consumers (high CAB/high alcohol consumers vs. high alcohol/low CAB consumers), and not identifying risk factors for patterns of alcohol dependence. 1.3. Methodological problems and controversy It is worth noting the occasions when AmED beverages do not show detrimental associations. In mostly international studies, there has been evidence that AmED beverages do not reduce subjective effects of intoxication or greatly increase risk propensity (Peacock, Bruno, & Martin, 2012; Peacock, Bruno, Martin, & Carr, 2013). Some of the research linking energy drinks and negative alcohol-related outcomes (e.g., Marczinski et al., 2013) has come under criticism as being relatively meaningless when examining real-world outcomes (Peacock & Bruno, 2013). For example, as Peacock and Bruno note, while Marczinski et al.'s (2013) study shows that AmED consumption primes users to desire more alcohol, the strength of the desire (e.g., 28 on a scale from 0 [the absence of desire] to 100 [very much desire]) may not cause any meaningful change in drinking choices in the real world. Thus, research is needed on both AmED priming and actual subsequent alcohol intake.

137

Peacock and Bruno (2013) argue that some of the variance in the relation between AmED and harmful outcomes may be related to demographic factors, such as the finding that young men who are more likely to drink AmED may also tend to take more risks. Peacock and Bruno (2013) also comment that studies controlling for important extraneous variables, such as sensation seeking, are absent. Research should attempt to control for factors that may be consistent with high-risk drinking (e.g., impulsivity, sensation seeking, race, gender), especially in cross-sectional studies. Thus, Peacock and Bruno (2013) make a viable and testable critique of studies examining links between AmED consumption and risk outcomes when they state that sensation seeking should be controlled for when examining AmED use. Sensation seeking may in fact be an important variable to assess when examining AmED consumption. Zuckerman (2007) stated that personality traits, like sensation seeking, might encourage individuals to participate in a wide array of risk behaviors, including high-risk drinking. While this makes conceptual sense, the evidence for sensation seeking explaining relations between AmED and risk outcomes is lacking. From the few examples that can be found in the literature, there is more evidence that relations between AmED consumption and negative outcomes are not confounded by sensation seeking or risk taking propensity. Arria et al. (2011) showed that links between energy drinks and alcohol dependence were significant, even after controlling for impulsive sensation seeking, conduct problems, and a history of alcohol and drug abuse. Moreover, Brache and Stockwell (2011) found that AmED consumption was associated with a range of high-risk drinking behaviors, a finding that remained significant after controlling for risk taking propensity. There has been limited study of AmED beverages and their relation to constructs such as patterns of alcohol dependence. However, there is a clear need for such evidence in light of the past research which shows clear associations between AmED consumption and high-risk drinking behavior (O'Brien et al., 2008). The goals of the present study were to describe associations among AmED consumption and patterns of alcohol dependence in a sample of young adult college students. Our analysis aimed to measure potentially confounding personality constructs thought to be associated with risk behavior (e.g., sensation seeking, impulsivity), and use those variables as statistical controls. 2. Method Data were collected from a subject pool of 757 undergraduate college students ages 18–25 at a mid-Atlantic university. Participants were students in psychology course and received course credit for their participation. Surveys were not completed during class hours; participants were free to fill out their survey in their free time online at any time before the end of the semester. Participants were free to select our survey from a list of other online studies being performed, based on a brief description of the purpose of each study. Participants were unaware of the main goal of the study. They were given an alternative assignment if they did not wish to participate. Approval was obtained for this study through the university's institutional review board. 2.1. Measures 2.1.1. Demographics Participants were asked to report their age, gender, race/ethnicity, sexual orientation, and relationship status. 2.1.2. Substance use Participants were asked a series of questions about their use of the following substances in the past 3 months: energy drinks, marijuana, ecstasy, methamphetamine, cocaine, ketamine, and “poppers” (amyl or butyl nitrate). Responses were scored on a Likert-type continuum ranging from 1 (none) to 4 (at least every week). Measures similar to these have shown utility in previous research (Benotsch, Snipes, Martin, & Bull, 2013).

138

D.J. Snipes et al. / Addictive Behaviors 41 (2015) 136–141

2.1.3. Personality constructs As a measure of personality constructs, the Substance Use Risk Profile Scale (SURPS; Woicik, Stewart, Pihl, & Conrod, 2009) was administered. The SURPS is a measure of four dimensions of personality/psychological functioning that are historically associated with substance use. These are hopelessness (sample item: “I feel that I'm a failure.”; α = .84), anxiety sensitivity (“I get scared when I'm too nervous.”; α = .68), impulsivity (“I often don't think things through before I speak.”; α = .70), and sensation seeking (“I like doing things that frighten me a little.”; α = .69). Internal reliabilities for these subscales were all sufficient for short scales (α ≥ .68). Participants answered a series of questions for each subscale, with response options ranging from 1 (Strongly Disagree) to 4 (Strongly Agree). 2.1.4. Alcohol and alcohol mixed with energy drinks Participants were asked to report the number of alcoholic drinks consumed in the past month (“In the past month, how many alcoholic drinks did you consume?”). Following that question, participants were asked to report the number of alcoholic drinks consumed in the past month that contained energy drinks (“Of those drinks, how many of them were alcohol mixed with energy drinks?”). These questions allowed us to compute a difference score to estimate the amount of non-AmED alcohol consumption, as well as an estimate of AmED consumption. This was done by subtracting the number of alcoholic drinks from the number of AmED drinks, resulting in an estimate of non-AmED alcohol. Similar measures of AmED consumption and non-AmED alcohol consumption have been used in previous research (Snipes & Benotsch, 2013). Past studies have shown the adequacy of alcohol measures which ask participants to reflect on the amount of alcohol consumed up to 12 months post-consumption (Greenfield, Nayak, Bond, Ye, & Midanik, 2006; National Council on Alcohol Abuse, Alcoholism—Task Force on Recommended Alcohol Questions, 2003). In the same way, participants were asked about the number of alcohol and AmED beverages consumed in the past week, as well as the number of days participants consumed alcoholic beverages in the past month. 2.1.5. Patterns of alcohol dependence Participants completed the 10-item Alcohol Use Disorders Identification Test (AUDIT; Babor, Higgins-Biddle, Saunders, & Monteiro, 2001). This measure is used to detect probable alcohol dependence in clinical samples. The recommended cut-off score for suggesting alcohol dependence is 8 (Babor et al., 2001); this was the cut-off used in this

study. We used a score of 8 to indicate a pattern of alcohol dependence, rather than a diagnosis of alcohol dependence, as this was a self-report survey and no clinical assessment was performed. 2.2. Data quality assurance & analysis The data collected were relatively complete, with less than 5% missing data overall. Certain variables were highly skewed (e.g., substance use, AUDIT scores, & alcohol consumption variables), and non-parametric tests were used to correct for this skewness. As there was some degree of missing data, some analyses have slightly different n's resulting in slightly different estimates for each analysis. Logistic regressions were checked for multicollinearity via VIF and tolerance values, with regression-based analyses showing no signs of multicollinearity (defined here as VIF N3 and tolerance values b .20). An alpha level of .05 was used, and marginal effects (p b .10) are reported, but not discussed. 3. Results The average age of the sample was 18.90 years (SD = 1.51). The sample was mostly female (68.8%), and mostly White (51.1%). The second most populous racial group was African Americans (22.5%), followed by Asian Americans (12.0%), “Other” (7.5%), Hispanic (6.5%), and Native American (0.4%). Participants were also reported being mostly heterosexual (94.3%), followed by bisexual (3.2%), gay/homosexual (1.8%), and “queer” (0.7%). A large number (55.9%) of our sample reported consuming alcohol in the past month (43.8% in the past week). Participants who reported consuming at least one AmED beverage in the past month comprised 11.6% (n = 87) of our sample, while 9.7% (n = 72) of participants reported AmED consumption in the past week. 3.1. Patterns of alcohol dependence, drug use, and personality constructs Tables 1 and 2 show differences between non-alcohol consumers, alcohol-only consumers, and AmED consumers for both past month and past week AmED consumption. As our drug use variables were skewed, they were dichotomized and chi-square analyses were performed. t-Tests were performed for the SURPS subscales. These results show that those who consumed AmED in the past month (as well as the past week) were significantly more likely to report AUDIT scores greater than or equal to 8 and general energy drink consumption, in

Table 1 Alcohol in the past month and substance use. No alcohol in the past month (n = 331) AUDIT score (≥8) Marijuana Ecstasy Methamphetamine Cocaine Ketamine Poppers Energy drinks

Hopelessness Anxiety sensitivity Impulsiveness Sensation seeking Total AUDIT score

3.9%a,b 10.6%a,b 1.2%a,b 0.9% 1.2% 0.3%m1 1.8% 27.8%a,m

Alcohol consumed in the past month (no AmED reported) (n = 333) 33.9%a,c 41.1%a,m 7.8%a 0.6% 1.5% 0.6%m2 2.1% 34.5%b,m

AmED consumed in the past month (n = 87) 67.8%b,c 55.2%b,m 10.3%b 3.4% 4.6% m1,m2 4.6% 3.4% 81.6%a,b

χ2 179.34⁎⁎⁎ 106.70⁎⁎⁎ 19.99⁎⁎⁎ 5.45m 4.84m 14.48⁎⁎ 0.89 87.16⁎⁎⁎

Mean (SD)

Mean (SD)

Mean (SD)

F

12.00 (3.45) 13.81 (2.48)a,b, 10.13 (2.64)a,b 14.59 (3.51)a,b 1.64 (3.67)a,b

11.82 (3.21) 13.33 (2.60)a,c 10.59 (2.21)a,c 16.29 (3.23)a 6.58 (4.77)a,c

12.40 (3.40) 12.57 (2.64)b,c 11.37 (2.47)b,c 17.10 (3.29)b 10.25 (5.83)b,c

1.09 8.75⁎⁎⁎ 9.48⁎⁎⁎ 30.60⁎⁎⁎ 174.39⁎⁎⁎

Means and percentages with shared superscript letters and/or numbers are significantly different from each other (using Tukey post-hoc tests for ANOVA and Bonferroni corrections for chi-square tests). m1 and m2 Descriptors. ⁎⁎ p b .01. ⁎⁎⁎ p b .001. m Refers to marginally significant (p's b .10) differences after Bonferroni correction.

D.J. Snipes et al. / Addictive Behaviors 41 (2015) 136–141

139

Table 2 Alcohol in the past week and substance use. No alcohol in the past week (n = 415) AUDIT score (≥8) Marijuana Ecstasy Methamphetamine Cocaine Ketamine Poppers Energy drinks

5.1%a,b 14.0%a,b 1.4%a,b 0.7% 1.2% 0.5%m1 1.4% 28.2%a,m

Hopelessness Anxiety sensitivity Impulsiveness Sensation seeking Total AUDIT score

Alcohol consumed in the past week (no AmED reported) (n = 252) 48.0%a 48.0%a 10.3%a 1.2% 2.0% 0.4%m2 3.2% 38.1%b,m

χ2

AmED consumed in the past week (n = 72)

202.41⁎⁎⁎ 108.68⁎⁎⁎ 27.84⁎⁎⁎ 2.46 3.23 18.07⁎⁎⁎

58.3%b 52.8%b 9.7%b 2.8% 4.2% m1,m2 5.6% 2.8% 80.6%a,b

2.35 72.77⁎⁎⁎

Mean (SD)

Mean (SD)

Mean (SD)

F

11.91 (3.37) 13.73 (2.50)a,b 10.14 (2.60)a,b 14.69 (3.47)a,b 2.00 (3.59)a,b

11.93 (3.33) 13.22 (2.62)a 10.74 (2.13)a,c 16.75 (3.09)a 8.06 (4.77)a,c

12.35 (3.47) 12.61 (2.65)b 11.61 (2.39)b,c 17.21 (3.39)b 9.65 (6.21)b,c

0.54 7.49⁎⁎ 13.33⁎⁎⁎ 38.73⁎⁎⁎ 203.33⁎⁎⁎

Means and percentages with shared superscript letters and/or numbers are significantly different from each other (using Tukey post-hoc tests for ANOVA and Bonferroni corrections for chi-square tests). m1 and m2 Descriptors. ⁎⁎ p b .01. ⁎⁎⁎ p b .001. m Refers to marginally significant (p's b .10) differences after Bonferroni correction.

To further examine the robustness of the relation between AmED consumption and patterns of alcohol dependence, a hierarchical logistic regression was performed. The logistic regression analysis (Table 3) predicted one of two outcomes: those who had AUDIT scores less than 8 (n = 566), and those with AUDIT scores 8 or greater (n = 185). The first step of the model included demographic factors, which provided a significant improvement in model fit over that of a constant-only

model, χ2 = 43.54, df = 5, Nagelkerke R2 = .084, p b .001. Being single, older, and White were all significant risk factors for patterns of alcohol dependence. The second step of the model added substance use, which significantly improved model fit, χ2 = 115.04, df = 6, Nagelkerke R2 = .283, p b .001. The only substance that significantly predicted patterns of alcohol dependence was marijuana. The next step of the model included psychological factors (hopelessness, anxiety-sensitivity, sensation seeking, and impulsivity), which significantly improved model fit, χ2 = 37.36, df = 4, Nagelkerke R2 = .341, p b .001. Both sensation seeking and impulsiveness were significant predictors of patterns of alcohol dependence. The next step of the model included both energy drinks and alcohol, which significantly improved model fit, χ 2 = 162.56, df = 2, Nagelkerke R2 = .564, p b .001. Both energy drinks and past month alcohol consumption were significant predictors of patterns of alcohol dependence. The final step of the model included past month AmED consumption, which again improved model fit, χ2 = 22.23, df = 1, Nagelkerke R2 = .591, p b .001. AmED consumption was a significant predictor of patterns of alcohol dependence over and above demographics, drug use, personality factors, energy drink use,

Table 3 Sequential logistic regression analysis predicting problematic alcohol consumption using past month measure of AmED (18–25 years old) ≤7 vs. ≥8.

Table 4 Sequential logistic regression analysis predicting problematic alcohol consumption using past week measure of AmED (18–25 years old) ≤7 vs. ≥8.

addition to lower anxiety sensitivity and greater impulsivity when compared to alcohol-only consumers. Differences were found in personality constructs when comparing alcohol-only and AmED consumers in both the past month and past week. Results suggest that AmED consumers had significantly lower anxiety-sensitivity than did alcohol-only consumers, but had greater impulsivity scores for both past month and past week reports. Interestingly, there were no significant differences in sensation seeking between AmED consumers and alcohol only consumers (p's N .05). 3.2. AmED consumption and problematic alcohol consumption

Variable and step

OR

CI

B

S.E.

p

1.

1.11 0.76 1.70 1.53 2.53 2.29 1.94 3.87 0.70 2.12 1.29 1.02 0.93 1.20 1.09 1.13 1.38 1.38

(1.00, 1.24) (0.53, 1.09) (0.86, 3.37) (1.08, 2.17) (1.80, 3.60) (1.87, 2.80) (0.96, 3.91) (0.45, 33.33) (0.17, 2.79) (0.24, 18.40) (0.66, 2.54) (0.96, 1.08) (0.86, 1.01) (1.09, 1.31) (1.02, 1.16) (1.10, 1.15) (1.06, 1.80) (1.17, 1.62)

.108 −.277 .530 .425 .926 .826 .662 1.35 −.361 .752 .257 .015 −.075 .178 .085 .118 .320 .321

.054 .185 .349 .179 .182 .103 .358 1.10 .708 1.10 .345 .031 .042 .045 .033 .013 .135 .082

b.05 ns ns b.05 b.001 b.001 ns ns ns ns ns ns ns b.001 b.05 b.001 b.05 b.001

2.

3.

4. 5.

Age Female Non-heterosexual Single White Marijuana Ecstasy Meth Cocaine Ketamine Poppers Hopelessness Anxiety sensitivity Impulsiveness Sensation seeking Alcohol (month) Energy drinks AmED use (month)

Note. N = 751. ns = not significant. a Nagelkerke R2.

R2a

Variable and step

OR

CI

B

S.E.

p

1.

1.11 0.76 1.73 1.53 2.52 2.28 1.96 3.87 0.69 2.01 1.29 1.01 0.93 1.20 1.09 1.13 1.38 1.56

(1.00, 1.24) (0.53, 1.09) (0.87, 3.44) (1.08, 2.17) (1.76, 3.60) (1.86, 2.79) (0.97, 3.99) (0.45, 33.31) (0.17, 2.78) (0.24, 18.21) (0.66, 2.54) (0.96, 1.08) (0.85, 1.01) (1.09, 1.31) (1.02, 1.16) (1.10, 1.15) (1.06, 1.79) (1.02, 2.39)

.108 −.275 .549 .423 .923 .824 .675 1.35 −.368 .740 .254 .014 −.076 .178 .085 .118 .319 .444

.054 .185 .351 .179 .182 .104 .361 1.10 .709 1.10 .346 .031 .042 .045 .033 .013 .135 .217

b.05 ns ns b.05 b.001 b.001 ns ns ns ns ns ns ns b.001 b.05 b.001 b.05 b.05

.084 2.

.283 3.

.341 4. .564 .591

5.

Age Female Non-heterosexual Single White Marijuana Ecstasy Meth Cocaine Ketamine Poppers Hopelessness Anxiety sensitivity Impulsiveness Sensation seeking Alcohol (week) Energy drinks AmED use (week)

Note. N = 754. ns = not significant. a Nagelkerke R2.

R2a

.083

.283

.342 .564 .570

140

D.J. Snipes et al. / Addictive Behaviors 41 (2015) 136–141

and alcohol use. Other predictors that remained significant in the final model include: Female gender (OR = 1.86, p = .03), marijuana (OR = 1.70, p b .001), impulsiveness (OR = 1.18, p = .004), and past month alcohol consumption (OR = 1.12, p b .001). The above analyses were replicated, but with our past week alcohol use measure. These results were similar and can be viewed in Table 4. Other predictors that remained significant in the final model include: Female gender (OR = 1.79, p = .04), marijuana (OR = 1.66, p b .001), impulsiveness (OR = 1.18, p = .003), and past week alcohol consumption (OR = 1.12, p b .001). 4. Discussion The results from the present study suggest the possibility of an increased risk of substance use and patterns of alcohol dependence among young adult college students who consume AmED beverages. Using two different measures of alcohol and AmED consumption (past week and past month), there were both strong and marginal univariate differences observed. Past month and past week AmED consumers were more likely than alcohol only consumers to report general energy drink consumption and patterns of alcohol dependence. Some marginally significant (after Bonferroni correction) differences between alcohol only consumers and AmED consumers were also only found. Marijuana use was marginally higher among past month AmED consumers than among alcohol only consumers, while ketamine was marginally different between these groups for both past week and past month reports. Significant differences were observed on some personality measures, with AmED consumers scoring higher on impulsivity, and lower anxiety sensitivity than alcohol only consumers. This makes conceptual sense, as anxiety sensitivity was associated with less alcohol consumption when the SURPS was being created (Woicik et al., 2009). Furthermore, individuals who are high in anxiety sensitivity may avoid the combination of energy drinks with alcohol, as the stimulant properties of energy drinks may exacerbate anxiety. These results suggest that AmED consumption appears to be a robust risk factor for patterns of alcohol dependence in young adults. AmED consumption may be more harmful than alcohol only consumption, as past research has shown that AmED consumption is associated with high-risk drinking habits (Brache & Stockwell, 2011; O'Brien et al., 2008; Woolsey et al., 2010) and increases the desire for more alcohol after consumption (Marczinski et al., 2013). With an increased desire to consume alcohol, individuals consuming AmED may be more likely to consume a greater number of alcoholic beverages in one sitting, and be more likely to underestimate their impairment (Ferreira et al., 2006; O'Brien et al., 2008; Thombs et al., 2010), which may lead to patterns of alcohol dependence. Higher reports of substance use between AmED consumers when compared to other categories of alcohol consumers are in line with previous work (Snipes & Benotsch, 2013). Individuals with alcohol problems may be at a heightened risk for other substance use/misuse (Falk, Yi, & Hiller-Sturmhöfel, 2008), which illustrates an additional risk for AmED consumers with patterns of alcohol dependence. Surprisingly, sensation seeking was not associated with AmED consumption in univariate analyses, but was associated with patterns of alcohol dependence in multivariate analyses. Sensation seeking typically serves as the mediator in many analyses, explaining why certain individuals are prone to engaging in risky behaviors (e.g., Walther, Cheong, Molina, & Pelham, 2012). Sensation seekers may have a proclivity for risky behaviors as a result of their personality disposition to seek out stimulation or thrill. However, the present study found that AmED consumption was not linked with sensation seeking (Tables 1 & 2), and that AmED consumption was associated with patterns of alcohol dependence above and beyond the influence of sensation seeking. This suggests that sensation seeking may not fully account for associations among AmED consumption and negative outcomes, which is an important finding given the criticism that links between AmED and risk outcomes are

potentially due to sensation seeking (Peacock & Bruno, 2013). Notably, AmED consumers also scored higher on impulsivity than did alcoholonly consumers. Little research has sought to examine impulsivity in the context of AmED consumption, but research has indicated that individuals with alcohol use disorders have higher levels of impulsivity than do healthy controls (Soloff, Lynch, & Moss, 2000). Energy drinks, as shown initially in a study by Arria et al. (2011), were also associated with patterns of alcohol dependence. However, the individual effect of energy drink use on alcohol problems did not eclipse the predictive power of AmED consumption in our model. As AmED is a mixture of energy drinks and alcohol, it might be expected for both alcohol and energy drinks to eclipse the predictive variance in AmED consumption (as AmED is a combination of both of those factors). However, the present study found that AmED consumption powerfully predicted patterns of alcohol dependence above and beyond the sum of its parts (alcohol and energy drinks), as well as personality traits like impulsiveness and sensation seeking. As this present study highlights, preventing AmED use in young adults is concerning given the associated risks and consequences. Although the U.S. Food and Drug Administration (FDA) prepackaged alcohol mixed with energy drinks in 2010, attempts to prevent caffeinated alcohol use will need to go beyond the traditional approach of regulating the sale and marketing of AmED as individuals tend to create their own concoction (Kponee, Siegal, & Jernigan, 2014). Young adults should be made aware of the risks, both physiological and behavioral, associated with AmED use. Given the concerning prevalence of AmED use, and its likely continued incline, stronger regulation of energy drinks is needed (Velasquez et al., 2012). 4.1. Limitations and implications for future research The present study utilized a convenience sample of college students from one university, thus limiting the generalizability of the results. All measures were based on self-report, which may result in over- or under-reporting of behaviors. Additionally, the cross-sectional design limits interpretation of causality. Although AmED consumption was found to be a risk factor for patterns of alcohol dependence, an alternative explanation could be that individuals with alcohol dependence problems seek out various alcoholic beverages that are trendy and popular. Longitudinal research is needed to more closely examine this relation. While this research adds to the literature in terms of understanding risks for alcohol dependence in the context of AmED beverage consumption, our results cannot infer as to why AmED beverages are a risk factor for such a hazardous outcome. Future research should examine the mechanisms by which energy drinks can influence patterns of alcohol dependence, and further understand the interactions between alcohol and energy drinks in the context of alcohol use disorders. Further, while research has demonstrated that many individuals consume AmED while partying (Malinauskas, Aeby, Overton, Carpenter-Aeby, & Barber-Heidal, 2007), additional research is needed to more closely examine the variety of situations in which individuals consume AmED beverages. 4.2. Conclusion Despite the aforementioned limitations, the present study is among the first to examine associations between AmED consumption and patterns of alcohol dependence. Results suggest that AmED consumers are more likely to engage in substance use and are likely to have patterns of alcohol dependence. This study adds to the literature illustrating problematic drinking behavior among consumers of AmED or energy drink beverages (Arria et al., 2011; Brache & Stockwell, 2011; O'Brien et al., 2008; Woolsey et al., 2010). Researchers, clinicians, and interventionists with interest in patterns of alcohol dependence may benefit from additional examination into consumption of AmED beverages.

D.J. Snipes et al. / Addictive Behaviors 41 (2015) 136–141 Role of funding sources This material is based upon work supported by the National Science Foundation Graduate Research Fellowship under Grant No. (NSF DGE-1147383). Any opinion, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The funders played no role in the design, collection, analysis, or interpretation of the data.

Contributors Daniel Snipes and Eric Benotsch designed the study and collected data. Daniel Snipes performed data analyses. Daniel Snipes, Brooke Green, and Amy Jeffers completed all writing in the manuscript. All authors approved the final version of this manuscript. Conflict of interest The authors on this paper state no conflicts of interest.

References Arria, A.M., Caldeira, K.M., Kasperski, S.J., Vincent, K.B., Griffiths, R.R., & O'Grady, K.E. (2011). Energy drink consumption and increased risk for alcohol dependence. Alcoholism: Clinical and Experimental Research, 35(2), 365–375. Arria, A.M., & O'Brien, M. (2011). The “high” risk of energy drinks. Journal of the American Medical Association, 305(6), 600–601. Babor, T.F., Higgins-Biddle, J.C., Saunders, J.B., & Monteiro, M.G. (2001). AUDIT: The alcohol use disorders identification test, guidelines for use in primary care (2nd ed.). Geneva: World Health Organization. Benotsch, E.G., Snipes, D.J., Martin, A.M., & Bull, S.S. (2013). Sexting, substance use, and sexual risk behavior in young adults. Journal of Adolescent Health, 52(3), 307–313. Brache, K., & Stockwell, T. (2011). Drinking patterns and risk behaviors associated with combined alcohol and energy drink consumption in college drinkers. Addictive Behaviors, 36(12), 1133–1140. Cheng, W. -J., Cheng, Y., Huang, M. -C., & Chiou-Jong, C. (2012). Alcohol dependence, consumption of alcoholic energy drinks and associated work characteristics in the Taiwan working population. Alcohol and Alcoholism, 47(4), 372–379. Falk, D., Yi, H., & Hiller-Sturmhöfel, S. (2008). An epidemiologic analysis of co-occurring alcohol and drug use and disorders. Alcohol Research and Health, 31(2), 100–110. Ferreira, S.E., de Mello, M.T., Pompeia, S., & de Souza-Formigoni, M.L.O. (2006). Effects of energy drink ingestion on alcohol intoxication. Alcoholism: Clinical and Experimental Research, 30, 598–605. Food and Drug Administration (2010). Caffeinated alcoholic beverages. Retrieved July 5, 2013 from. http://www.fda.gov/Food/IngredientsPackagingLabeling/ FoodAdditivesIngredients/ucm190366.htm Greenfield, T.K., Nayak, M.B., Bond, J., Ye, Y., & Midanik, L.T. (2006). Maximum quantity consumed and alcohol-related problems: Assessing the most alcohol drunk with two measures. Alcoholism: Clinical and Experimental Research, 30(9), 1576–1582. Heckman, M.A., Sherry, K., Mejia, D., & Gonzalez, E. (2010). Energy drinks: An assessment of their market size, consumer demographics, ingredient profile, functionality, and regulations in the United States. Comprehensive Reviews in Food Science and Food Safety, 9(3), 303–317. Lau-Barraco, C., Milletich, R.J., & Linden, A.N. (2013). Caffeinated alcohol consumption profiles and associations with use severity and outcome expectancies. Addictive Behaviors, 39(1), 308–315. Kponee, K. Z., Siegel, M., Jernigan, D. H., & Carpenter-Aeby, T. (2014). The use of caffeinated alcoholic beverages among underage drinkers: Results of a national survey. Addictive Behaviors, 29, 253–258.

141

Malinauskas, B.M., Aeby, V.G., Overton, R.F., Carpenter-Aeby, T., & Barber-Heidal, K. (2007). A survey of energy drink consumption patterns among college students. Nutrition Journal, 6(1), 35–41. Marczinski, C.A., Fillmore, M.T., Henges, A.L., Ramsey, M.A., & Young, C.R. (2013). Mixing an energy drink with an alcoholic beverage increases motivation for more alcohol in college students. Alcoholism: Clinical and Experimental Research, 37(2), 276–283. Miller, K.E. (2012). Alcohol mixed with energy drink use and sexual risk-taking: Casual, intoxicated, and unprotected sex. Journal of Caffeine Research, 2(2), 62–69. National Council on Alcohol Abuse, Alcoholism—Task Force on Recommended Alcohol Questions (October 15–16) (2003). Recommended sets of alcohol consumption questions. Retrieved December 30, 2011 from. http://www.niaaa.nih.gov/Resources/ ResearchResources/TaskForce.htm O'Brien, M.C., Arria, A.M., Howland, J., James, J.E., & Marczinski, C.A. (2011). Caffeine, alcohol, and youth: A toxic mix. Journal of Caffeine Research, 1(1), 15–21. O'Brien, M., McCoy, T., Rhodes, S., Wagoner, A., & Wolfson, M. (2008). Caffeinated cocktails: Energy drink consumption, high-risk drinking, and alcohol-related consequences among college students. Academic Emergency Medicine, 15(5), 453–460. Olive, M.F. (2002). Interactions between taurine and ethanol in the central nervous system. Amino Acids, 23(4), 345–357. Peacock, A., & Bruno, R. (2013). “High” motivation for alcohol: What are the practical effects of energy drinks on alcohol priming? Alcoholism: Clinical and Experimental Research, 37(2), 185–187. Peacock, A., Bruno, R., & Martin, F.H. (2012). The subjective physiological, psychological, and behavioral risk‐taking consequences of alcohol and energy drink co‐ingestion. Alcoholism: Clinical and Experimental Research, 36(11), 2008–2015. Peacock, A., Bruno, R., Martin, F.H., & Carr, A. (2013). The impact of alcohol and energy drink consumption on intoxication and risk-taking behavior. Alcoholism: Clinical and Experimental Research, 37(7), 1234–1242. Snipes, D.J., & Benotsch, E.G. (2013). High-risk cocktails and high-risk sex: Examining the relation between alcohol mixed with energy drink consumption, sexual behavior, and drug use among college students. Addictive Behaviors, 38(1), 1418–1423. Snipes, D.J., Green, B.A., Javier, S.J., Perrin, P.B., & Benotsch, E.G. (2014). The use of alcohol mixed with energy drinks and experiences of sexual victimization among male and female college students. Addictive Behaviors, 39(1), 259–264. Soloff, P.H., Lynch, K.G., & Moss, H.B. (2000). Serotonin, impulsivity, and alcohol use disorders in the older adolescent: A psychobiological study. Alcoholism: Clinical and Experimental Research, 24(11), 1609–1619. Thombs, D.L., O'Mara, R.J., Tsukamoto, M., Rossheim, M.E., Weiler, R.M., …, et al. (2010). Event-level analysis of energy drink consumption and alcohol intoxication in bar patrons. Addictive Behaviors, 35(4), 325–330. Velasquez, C.E., Poulos, N.S., Latimer, L.A., & Pasch, K.E. (2012). Associations between energy drink consumption and alcohol use behaviors among college students. Drug and Alcohol Dependence, 123(1–3), 167–172. Walther, C.A.P., Cheong, J., Molina, B.S.G., & Pelham, W.E. (2012). Impulsivity and sensation seeking as mediators of the relation between ADHD and engagement in and frequency of alcohol use. Alcoholism: Clinical and Experimental Research, 36(SI-1), 51-A. Woicik, P.A., Stewart, S.H., Pihl, R.O., & Conrod, P.J. (2009). The substance use risk profile scale: A scale measuring traits linked to reinforcement-specific substance use profiles. Addictive Behaviors, 34(12), 1042–1055. Woolsey, C., Waigandt, A., & Beck, N.C. (2010). Athletes and energy drinks: Reported risktaking and consequences from the combined use of alcohol and energy drinks. Journal of Applied Sport Psychology, 22(1), 65–71. Zuckerman, M. (2007). Sensation seeking and risky behavior. American Psychological Association.