Mixing alcohol and energy drinks in daily life: A Commentary on

Consuming alcohol mixed with energy drinks (AmEDs) is linked with heavy alcohol use and a wide range of negative consequences from drinking such as risk for injury, driving while under the influence, and sexual risk behavior (e.g., Linden & Lau-Barraco, 2014; Marczinski & Fillmore, 2014; Roemer & Stockwell, 2017). The majority of the research on AmED use has been restricted to between-person comparisons, limiting our understanding of whether drinking AmEDs increases the risk for alcohol-related harm versus riskier drinkers are more likely to consume AmEDs. Verster, Benson, Johnson, Alford, Godefroy, and Scholey (2018) conducted a meta-analysis of between- and within-person studies examining AmED and alcohol-only use. They concluded that, on average, AmED consumers drink more than alcohol-only consumers (a between-person difference), but within AmED consumers, individuals drink equal amounts and experience fewer consequences on AmED occasions relative to alcohol-only occasions (a within-person difference). Within-person comparisons are essential for understanding the unique risks associated with drinking AmEDs; however, the studies from which these findings are based have important methodological limitations that should be illuminated. In light of recent daily diary evidence published after Verster and colleagues completed their literature review (Linden-Carmichael & Lau-Barraco, 2017a), we strongly caution readers from prematurely concluding that, among AmED consumers, AmED occasions are equally safe or safer than occasions involving alcohol alone.

The within-person studies cited in Verster and colleagues offered an important, preliminary examination of whether participants generally report drinking more on AmED occasions than drinking occasions without energy drinks. Drinking behaviors in these prior studies were assessed by asking participants to retrospectively estimate how many alcoholic beverages they typically consumed over the course of a wide range of time, such as past 30 days (Brache & Stockwell, 2011; de Haan, de Haan, van der Palen, Olivier, & Verster, 2012) to 6 months (Peacock, Bruno, & Martin, 2012). Some found that AmED occasions were linked with heavier alcohol use (e.g., Brache & Stockwell, 2011; Price, Hilchey, Darredeau, Fulton, & Barrett, 2010) whereas others found equal or less use during AmED occasions (e.g., de de Haan, de Haan, van der Palen, Olivier, & Verster, 2012; Woolsey, Waigandt, & Beck, 2010). Asking participants to retrospectively report on the amount they consumed during a typical occasion over a large time window can produce clear recall biases (Gmel & Daeppen, 2007; Gmel & Rehm, 2004). The recall bias could lead to potential inaccurate estimates of the amount consumed during an average AmED session versus an average alcohol-only drinking session. One way to reduce this recall bias and examine within-person effects is to use a daily diary design. Such designs assess behaviors daily over some time frame (e.g., reporting alcohol use each day for 14 days) rather than aggregated (e.g., “how much alcohol do you typically consume?”) or retrospectively recalled (e.g., timeline follow back to recall daily use on each of the past 30 days) by the respondent. In light of evidence that recall of alcohol intake is diminished after 2 to 3 days (Ekholm, 2004), daily diary studies have a clear advantage in asking participants to report on their drinking behavior last night. These designs also can be used to estimate the odds of experiencing negative consequences of a drinking occasion after controlling for the amount of alcohol consumed, thereby isolating the within-person effect of drinking AmEDs relative to alcohol alone.

Recently, Linden-Carmichael and Lau-Barraco (2017a) conducted a 14-day diary study of 18–25-year-old heavy-drinking college students who reported recent (past week) caffeinated alcohol use (e.g., AmEDs). Participants reported their drinking behavior each day for 14 consecutive days, including the type and amount of alcohol consumed, harms experienced, and drinking-related psychosocial factors (e.g., motives and social context). Multilevel modeling, an analytic technique that takes into account within- and between-person effects, compared drinking occasions involving (a) alcohol mixed with a caffeinated, nonenergy drink mixer (e.g., rum and soda), (b) alcohol mixed with an energy drink, and (c) alcohol without a caffeinated mixer. Trait impulsivity was included as a person-level covariate to adjust for any differences in typical impulsive behavior. Results indicated that participants consumed significantly more alcohol on occasions in which they consumed alcohol with caffeine (regardless of whether the mixer was an energy drink) relative to drinking occasions involving alcohol alone. In addition, after controlling for the amount of alcohol consumed, AmED occasions were associated with more alcohol-related harms than drinking occasions involving alcohol alone. No differences in negative consequences were observed when comparing alcohol-only and alcohol with a cola-caffeinated mixer. The most commonly reported consequences on days in which individuals mixed alcohol with caffeine pertained to blacking out or self-care concerns. Blacking out from drinking is particularly concerning given that this subtype of alcohol-related problem has been shown to be among the most predictive of problematic drinking behavior in college students (Read, Beattie, Chamberlain, & Merrill, 2008). It should be noted that no significant differences were found for engagement in sex after drinking and driving after drinking, which may have been related to low frequency of reporting these behaviors during the 2-week period. Further studies using these data revealed that, in comparison with alcohol-only days, AmED days were associated with greater odds of drinking in potentially risky social environments (e.g., bars or clubs, while pregaming; Linden-Carmichael & Lau-Barraco, 2017b) and for unique reasons (i.e., drinking to fit in; Linden-Carmichael & Lau-Barraco, 2018), which may contribute to the harms reported.

These more recent daily diary findings refute the conclusions reached by Verster and colleagues. On the basis of data gathered through the use of a stronger methodological approach than prior studies, we conclude not only that AmED users are a riskier subpopulation than alcohol-only users but also that AmED users are at greater risk for harms and drink more alcohol when they drink AmEDs. In concert with the majority of studies demonstrating a variety of risks posed by AmED use in cross-sectional (e.g., Brache & Stockwell, 2011; Snipes, Jeffers, Green, & Benotsch, 2015) and experimental research (e.g., Marczinski, Fillmore, Henges, Ramsey, & Young, 2012), we assert that consumption of AmEDs continues to be a public health concern. It is our hope that this daily diary evidence serves as a basis for future research teasing apart the between- and within-person effects of AmED use in real-world settings, including a comparison of other factors noted in Verster and colleagues (2018), such as objective versus subjective intoxication and risk-taking behavior. Some field research omitted from this review has found bar patrons who consumed AmED beverages to be more likely to leave bars intoxicated and state intentions to drive (Thombs et al., 2010). There appears to be a dose-dependent relationship between the level of caffeine use (via energy drinks or cola beverages) and the magnitude of alcohol intoxication in bar patrons (Thombs et al., 2011). Furthermore, as the daily diary data we noted were collected from U.S. college students, it will be important for future research to consider patterns of use from other populations, including noncollege young adults and participants outside the U.S. In consideration of all current evidence on AmED use, it would be premature and erroneous to conclude that AmED use is safe. From a public health perspective, it is important that researchers continue to understand the negative outcomes associated with high-risk alcohol use behaviors, including AmED use.