Principal findings
Spurred by the steady rise in ED use and the targeted marketing of these drinks to young adults, much of the existing research regarding caffeine use by college students has focused exclusively on EDs and their associations with a variety of risky health behaviors [
9]. Studies have shown, however, that other sources of caffeine such as coffee and soft drinks are more frequently used by college students than EDs, and these sources of caffeine should be considered when evaluating associations between caffeine and other substance use and problem behaviors [
25]. The present study is among the first to look concurrently at coffee and ED use in college students and to evaluate associations between their use and alcohol, tobacco and other drug use; alcohol use problems; and parental substance abuse and mental health problems. Analyses found students who consumed EDs (with or without concurrent coffee use) were most likely to report other substance use, alcohol-related problem behaviors, and peer/family risk factors for substance use followed by students who consumed coffee only, and finally, students who reported using neither EDs nor coffee. The data are particularly noteworthy for the consistent response pattern observed across almost all domains assessed.
In most previous research, the relationship between other sources of caffeine and adverse health behaviors was either not considered
,14,17,30 or used as a covariate in the data analysis [
11]. The focus on EDs as the singular source of caffeine in these studies started with the compelling data from O’Brien and colleagues (2008) who reported an association between the use of EDs mixed with alcohol and both risky drinking and alcohol-associated adverse health behaviors [
8]. Subsequent researchers continued to focus on EDs and risky health behaviors, in part because of the intense marketing efforts ED makers directed at college-age students and the relatively higher amounts of caffeine in EDs compared to traditional sources of caffeine (e.g., 40 mg caffeine in a 12-oz can of Coca-Cola vs 80 mg in a 12-oz can of Red Bull). More recently, however, many specialty coffee drinks (150 mg in a 12-oz cappuccino) and even soft drinks (110 mg caffeine in a 12-oz can of Coke Energy) contain caffeine in amounts like those found in EDs. Another reason researchers focused singularly on EDs was because these beverages often are consumed more rapidly than hot caffeinated beverages like coffee. Many thought the relatively rapid rate of consumption of EDs may lead to higher caffeine levels and thus, greater association with risky health behaviors, compared to caffeinated drinks that are typically consumed more slowly, like hot coffee drinks. White et al. (2016) however, recently showed there was no clinically significant difference in caffeine exposure (i.e., T
max, MRT, MAT or AUC
0–∞) regardless of the rapidity with which caffeine was consumed [
30].
Much of the early research in college students who mixed EDs with alcohol showed that these students consumed alcohol more frequently, in higher amounts, and with more episodes of binge and problem drinking than students consuming alcohol without ED mixers [
8,
30,
31]. Not surprisingly, AmED users also were more likely than non-AmED users to engage in other risky health behaviors including risky sexual behavior, dangerous driving behavior, and physical altercations [
8,
32,
33]. Both clinical and laboratory research suggest students who consume AmED have altered perceptions of their levels of intoxication, with these students not recognizing their levels of impairment [
8,
34]. Early research also consistently found college students who used ED, independent of concomitant alcohol use, were more likely to report alcohol use; meet criteria for alcohol dependence; use tobacco, marijuana, and nonmedical prescription drugs; and engage in risky sexual and physical behaviors [
10,
14,
21].
A few significant exceptions to the early ED-only and AmED-only focused research in college students showed that other sources of caffeine also were associated with risky health behaviors. Thombs and colleagues (2011) compared the effects of AmEDs to alcohol mixed with cola and alcohol alone on alcohol use in college students [
35]. The researchers found a dose-dependent relationship between the estimated amount of caffeine consumed from both EDs and soft drinks and risky alcohol use. Using data from a group of Icelandic college students, Kristjansson et al. (2015) showed that daily consumption of coffee, soft drinks, and EDs, but not tea, was positively associated with drinking AmEDs [
36]. In addition, Anderson and Juliano (2012) showed that estimated mean weekly caffeine consumption, regardless of the source, was positively correlated with the amount of alcohol consumed by college students [
37]. These cross-sectional studies suggest the amount of caffeine consumed is more important than the source of caffeine with regard to the likelihood that college students will engage in adverse health behaviors. More recently, Dillon and colleagues (2019) investigated the relationship between all sources of caffeine and adverse health behaviors in college freshmen [
38]. They found that students who consumed caffeine daily from any source were more likely to report alcohol, cigarette, and nonmedical drug use and problem drinking than those who did not consume caffeine.
In the present study, we elected to focus on coffee and ED consumption in college students for three reasons. First, these beverages typically have the highest caffeine content and, over time, they have come to represent a greater proportion of caffeine intake in US children and adolescents [
39]. Second, coffee is used frequently by college students [
24,
40], with one recent convenience sample survey of college students finding coffee to be their primary source of caffeine intake (72%), followed by soft drinks (69%), tea (61%), and EDs (36%) [
24]. Third, research done by our group prior to the surge in popularity of EDs underscored the importance of considering coffee when evaluating the effects of caffeine on substance use. Our research showed that college women who drank coffee daily were more likely to report heavier drinking and alcohol-related problems than non-daily coffee drinkers [
25].
Like previous work, this research found an association between caffeine and risky health behaviors. This relationship was more robust for students in the ED + Co group compared with those who drank coffee only and those consuming neither beverage. While the cross-sectional nature of the work limits our ability to establish a causal relationship between caffeine, other substance use, and alcohol use problems, the associations are likely a result of a combination of genetic, psychobiological, and environmental factors.
Our study is among the first to look at familial factors associated with caffeine use. We found participants reporting ED ± Co use were more likely to report maternal alcohol problems and depression/anxiety symptoms as well as paternal alcohol and drug problems and depression/anxiety. Such familial clustering may occur because of both a shared environment and genetic factors. In fact, Kendler, Myers, and Gardner (2006) [
23], in a study in adult twins, found that a link between caffeine use and the development of substance use and psychiatric disorders was due primarily to familial factors, including genetic factors. With the compelling and consistent association between EDs and risk-taking behaviors, most frequently other substance use, researchers have linked sensation-seeking personality traits and ED use. College students who scored higher on measures of sensation-seeking were more likely to consume ED and AmED [
10,
14,
41]. This may be due to caffeine’s potentiation of the psychostimulant effect of other drugs of abuse through its effects on the adenosine and dopamine pathways. In addition, when combined with alcohol, caffeine blunts the depressant effects and enhances the stimulant effects of alcohol, which alone is associated with risk-taking, by affecting the same pathways [
42]. The increased stimulant effect, decreased depressant effects, and propensity for risk-taking may lead to increased sensation-seeking behavior, including ED use.
Environmental factors likely impact the association between caffeine use and risky health behaviors as well. Almost all college students use caffeine regularly [
43]. At the same time, most college students are in the age range, late teens and early 20s, at highest risk for the onset of many substance use disorders [
26]. The temporal intersection between high frequency caffeine use and increased prevalence of substance use may explain the association between caffeine and risky health behaviors. Patterns between ED use and other drug use have also been found in younger age groups (8th, 10th, and 12th graders) [
3]. In addition, alcohol and other substances like marijuana and tobacco are often part of the college milieu, and students may use caffeine to affect the pharmacodynamic effects of these other substances. For instance, students may concurrently consume caffeine to offset the depressant effects of alcohol or marijuana while socializing or use caffeine to increase their energy when they have school obligations after a night of heavy drinking. Finally, there is evidence that peer influence increases adolescents’ substance use, and this may contribute to the risky behaviors reported by our sample [
44]. Indeed, we found that students who used EDs and/or coffee (ED ± Co and Co groups) were more likely to report smoking and having friends who smoked than the NoCE group.
Limitations
There are several limitations to the present study. First, we relied on retrospective self-report data to address our research question. Second, participants were surveyed about recent (past 30 days) caffeine consumption, which did not allow us to examine use patterns over longer periods. Nonetheless, a 30-day timeframe focused on recent caffeine use appeared to be an appropriate starting point for examining substance use/problems associated with cross-beverage caffeine consumption. Third, the low number of ED only (no coffee) users (N = 84) prevented statistical power for a 4-group comparison. Instead, present study analyses included an ED + Co group in which three-fourths of the sample reported use of both ED and coffee (76%) and one-fourth reported ED use but no coffee. One advantage of the ED + Co group is that it is similar to much of the published research in which ED use was defined without attention to concurrent coffee use, and this allows our data to be compared to the extant literature. Fourth, only frequency of caffeine use was assessed, with no quantity of use data. The survey used for this research was originally designed to assess alcohol use in college students, with limited caffeine use questions, and future research should collect more detailed quantitative data about quantity and frequency of caffeine use. Fifth, caffeine use was restricted to only coffee and ED use; other sources of caffeine intake (e.g., tea, sodas) were not included.