Alcohol Mixed with Energy Drinks (AmED) Use among University Students: A Systematic Review and Meta-Analysis

In the last decades, there has been a huge increase in the consumption of both Energy Drinks (EDs) and alcohol and, concurrently, these two trends generated the additional practice of mixing ED with alcohol, known as Alcohol mixed with Energy Drink (AmED). One of the most important group of AmED consumers is represented by young. Indeed, the study population of the researches in this field are mainly represented by college students and the results evidence a great range of negative consequences for health. The purpose of the systematic review was to explore the prevalence of AmED consumption among undergraduate students, together with motivations to their use, adverse effects and health-related behaviors associated to AmEDs use. The review was conducted according to the PRISMA Statement and PubMed, Scopus and Web of Science were interrogated. 42 articles, published from 2008 to 2021, were included in the review. An overall prevalence rate of 37% was estimated for AmEDs use in undergraduates, with geographical differences. Although a decrease in consumption was observed throughout the studied period, a continuous monitoring on this phenomenon is needed for identifying those risk groups that could develop risky behaviors related to AmEDs consumption and provide them targeted educational interventions.


Introduction
The consumption of Energy Drinks (EDs) is very popular because of their advertised effects such as the increase of energy, concentration, athletic performance, metabolism and stimulation of mental activity and alertness, which represent the main reasons to use these beverages, other than liking the taste [1][2][3][4]. One of the most relevant group of consumers is represented by students, which use EDs to overcome the high levels of stress related to study commitments [5] and to get benefits in cognitive performance, concentration and mood [6] as well as to stay awake and increase energy levels. After more than 20 years since the debut of the first ED company, there has been a huge increase in their consumption, complemented by an overall increase in alcohol intake [7], and concurrently these two trends generated the additional practice of mixing ED with alcohol, which is known as the consumption of Alcohol mixed with Energy Drink (AmED). Compared with alcohol alone, consuming AmED increases total alcohol consumption; one hypothesis which justifies this increase is that the stimulant effects of caffeine, one of the main ingredients of EDs, may counteract the depressant effects of alcohol [4]. Indeed, most of the research comparing AmED consumers with alcohol only consumers reported higher levels of alcohol in AmED consumers respect to alcohol users. Usually, the target of these studies is Table 1. Eligibility criteria used in the article selection process based on the PICOS framework.

Population
University students, independently by their gender and age

Intervention Consumption of AmEDs in everyday life
Comparison Age-, gender-and condition-matched control group (if present)

Outcomes
To explore the prevalence of AmED consumption; adverse effects and health-related behaviors associated to AmEDs consumption and motivations to their use were also analyzed Study design Cross-sectional studies Three electronic databases (PubMed, Scopus and Web of Science) were questioned using the following query string: "energy drink" AND "alcohol" OR "AmED" AND ("university students" OR "college students" OR "undergraduate*"). Table S1 shows the detailed search strategy.
The search was performed from 20 to 31 May 2022 and was carried out by title, abstract, and MeSH terms on PubMed or keywords on Scopus and Web of Science. Table 2 shows the inclusion and exclusion criteria used in the selection process. Table 2. Inclusion and exclusion criteria used in the article selection process.

Inclusion Criteria Exclusion Criteria
Articles reporting data about AmED specific consumption of university students, independently by their gender and age Articles presenting studies that included individuals who were not university students or studies which regarded the consumption of other dietary supplements than AmEDs or only EDs or alcohol alone or other substances Articles presenting cross-sectional studies Articles presenting clinical trials, experimental studies, reviews, meta-analysis, case studies, proceedings, qualitative studies, editorials, commentary studies and any other type Articles published in English language, from the inception to 31 May 2022 Articles published in languages other than English Titles and abstracts obtained from the databases were transmitted to the reference software Zotero systematic review manager for the process of assessment. The subsequent step was screening by title and abstract the potentially eligible studies, following the inclusion criteria; the screening was performed by five authors (F.G., C.P., F.V., A.D.G.) independently. Then, full texts were read independently by the consensus team (F.G., C.P., F.V., A.D.G.) and disagreements about their inclusion were achieved by consensus among the authors.

Data Extraction Process and Quality Assessment
A specific set of categories were chosen as the extracted data following consensus of all authors: bibliographic information like author, year, country, sample size, study subject/population with age and gender; and AmED consumption estimate, associated healthrelated behaviors, reason and adverse effects due to AmED consumption as outcomes.
The Newcastle-Ottawa Quality Assessment Scale-NOS-adapted for cross-sectional studies was used for quality assessment. An overall quality rating was assigned to each eligible article according to the number of criteria met as follows: Good Quality (all criteria met, low risk of bias); Fair Quality (1 criteria not met or 2 criteria unclear, moderate risk of bias); Poor Quality (2 or more criteria not met, high risk of bias). Five authors (F.G., C.P., F.V., A.D.G.) independently assigned a score to each study, and disagreements were settled by consensus among all the authors.

Statistical Analysis
Comprehensive Meta Analysis 4.0 (Biostat, Englewood, NJ, USA) was used for meta-analysis and statistical elaborations. The prevalence of AmED consumption and 95% CI were extracted for each study. The pooled estimation of prevalence was calculated using random-effects model for higher external validity of findings because are included studies with different populations. According to previous studies [16,17], the formula Logit = Ln(p/(1 − p)), where p represents the prevalence rate and Ln the natural logarithm, was used to transform prevalence rate in its logarithmic form, and V(Logit) = 1/np + 1/n(1 − p), where V represents the variance, was used to transform samples' variance. The conversion is based on the formula p = eLogit/(eLogit + 1), with e being the base of the natural logarithm.
In order to evaluate the heterogeneity of the selected studies, the I 2 test and the classical measure of heterogeneity Cochran's Q (Hedges Q statistic) were used. The following thresholds of I 2 were employed: <25% = low heterogeneity; <50% = moderate heterogeneity; and >75% = high heterogeneity [18]. To assess the publication bias, the Egger's test and Funnel plot were employed [19]. Meta-regression and subgroup analyses were performed to evaluate the sources of heterogeneity [20][21][22]. For meta-regression analysis, sample size, gender and age of participants, WHO Regions location (European Region, African Region, Region of the Americas, South-East Asia Region, Eastern Mediterranean Region, Western Pacific Region), years since publication and methodological quality of the studies were considered as possible sources of heterogeneity; the time considered to assess AmED consumption was also included into the meta-regression analysis by considering four subgroups (Past week, Past 30-days, Past 60-days, Past 90-days and Past year). Three articles [23][24][25] were excluded from the meta-analysis and meta-regression analyzes because they did not report the prevalence rate of AmED consumption. Two other papers were excluded because they involved the same cohort of participants [26]. Figure 1 shows the steps of the article selection process used for the systematic review following the PRISMA statement [15].  On a total of 714 records found, 557 were screened for inclusion and 51 assessed for eligibility. Five articles were excluded because not specific for AmED, one article because the study population did not include only university students, two articles because they were not pertinent, and one article because it was not in English. Finally, 42 articles met the inclusion criteria and were included in the analysis (Tables 3 and 4). On a total of 714 records found, 557 were screened for inclusion and 51 assessed for eligibility. Five articles were excluded because not specific for AmED, one article because the study population did not include only university students, two articles because they were not pertinent, and one article because it was not in English. Finally, 42 articles met the inclusion criteria and were included in the analysis (Tables 3 and 4).     13.4% were current smokers, problematic drug use was low and average scores on the AUDIT-C were situated at proposed cut-offs for hazardous drinking.

NR
Brache and Stockwell 2011 Canada [33] Significant associations between the consumption of AmED and any stimulant drug use (cocaine, crack-cocaine, amphetamines, and crystal meth) NR   AmED users reported higher average days drinking per week, average number of drinks per occasion, total binge drinking episodes in past year, greatest number of drinks on one occasion, and total number of drinks than alcohol-only consumers To act aggressively, to be more alert, to feel stronger, to feel sober up quicker and to drive a vehicle were more commonly reported by AmED than by alcohol-only consumers Woolsey et al., 2015 USA [63] AmED users are more likely to drive after drinking, drive while knowingly over the blood alcohol content driving limit and to ride with an intoxicated driver 36% reported feeling more confident, 45% felt they could drink more alcohol, 25% felt energy drinks reduce the negative effects of alcohol, 20% felt energy drinks sober them up quicker, and 13% felt that they were more capable to drive AmED users reported more driving despite knowing they had too much alcohol to drink, driving over the blood alcohol content driving limit, more alcohol average drinks, days drinking (30 day), days drunk (30 day

AmED Consumption Prevalence and Related Aspects
Among the selected studies, the prevalence of AmED users ranges from 25.6% [37] to 84.4% [32]. Five articles considered the differences of consumption according to gender [36,41,53,57,63], showing no significant differences with the exception of the study by Woolsey et al. [64], which found a major intake in males.

Meta-Analysis and Meta-Regression Results
With regards to the meta-analysis results, the estimated overall prevalence of AmED consumption (Figure 2) was 37% in undergraduate students (95% CI, the range of prevalence is 13-99%), with significant heterogeneity among studies (Q test: p < 0.001; I 2 = 99.4%). Sensitive analysis did not substantially change the pooled prevalence of AmED consumption, which resulted equal to 38% (95% CI: 13-99%) with the inclusion of Linden-Carmichael et al., 2017, 2018 [43,44], suggesting that no one single study had a disproportional impact on overall prevalence. Sensitive analysis did not substantially change the pooled prevalence of AmED consumption, which resulted equal to 38% (95% CI: 13-99%) with the inclusion of Linden-Carmichael et al. 2017, 2018 [43,44], suggesting that no one single study had a disproportional impact on overall prevalence. A visual inspection of funnel plot suggested no publication biases in the present study (p = 0.001) ( Figure S1).
To investigate the role of the study characteristics in influencing the heterogeneity of the global prevalence of AmED consumption, a meta-regression analysis was performed considering sample size, gender, age, publication year, time to which AmED consumption was referred to, methodological quality of the study and WHO region in which the study was performed. The results showed that the prevalence of AmED consumption was independent by the amount of females in the sample (p = 0.750), mean age (p = 0.140), time of AmED consumption (p = 0.240), or methodological quality (p = 0.250). Instead, the prevalence of AmED consumption in the world slightly showed a decrease with the increase of the year of publication (p < 0.05). According to the meta-regression analysis, the study location showed a slight action of moderation (Q = 3.19, df = 2, p < 0.05). The prevalence of AmED consumption was 73% in AMR Region, 19% in EUR, 8% in WPR.

Discussion
Mixing Alcohol with Energy Drinks is a common practice among young people and a great concern for public health because it can be associated with several adverse effects and other risky habits, such as binge drinking and alcohol dependence [4,7,8,28,65]. To explore the prevalence of AmED consumption and motivations to their use among undergraduate students, the available literature from 2008 to 2021 was analyzed in this review. The results showed that the estimated overall prevalence of AmED consumption was 37% A visual inspection of funnel plot suggested no publication biases in the present study (p = 0.001) ( Figure S1).
To investigate the role of the study characteristics in influencing the heterogeneity of the global prevalence of AmED consumption, a meta-regression analysis was performed considering sample size, gender, age, publication year, time to which AmED consumption was referred to, methodological quality of the study and WHO region in which the study was performed. The results showed that the prevalence of AmED consumption was independent by the amount of females in the sample (p = 0.750), mean age (p = 0.140), time of AmED consumption (p = 0.240), or methodological quality (p = 0.250). Instead, the prevalence of AmED consumption in the world slightly showed a decrease with the increase of the year of publication (p < 0.05). According to the meta-regression analysis, the study location showed a slight action of moderation (Q = 3.19, df = 2, p < 0.05). The prevalence of AmED consumption was 73% in AMR Region, 19% in EUR, 8% in WPR.

Discussion
Mixing Alcohol with Energy Drinks is a common practice among young people and a great concern for public health because it can be associated with several adverse effects and other risky habits, such as binge drinking and alcohol dependence [4,7,8,28,65]. To explore the prevalence of AmED consumption and motivations to their use among undergraduate students, the available literature from 2008 to 2021 was analyzed in this review. The results showed that the estimated overall prevalence of AmED consumption was 37% in the populations studied, with a great variability and a significant heterogeneity among the studies.
Among the possible factors influencing the heterogeneity of the global prevalence of AmED consumption arose the study location, with AMR and EUR regions showing the major prevalence. However, it should be noted that the data reported by the National Institute on Drug Abuse at The National Institutes of Health in USA showed a substantial decline in lifetime alcohol use among youths from 1991 to 2017 [64]. These data are in agreement with the observation that the year of publication is negatively correlated with the prevalence of AmED consumption, and recent studies reported a lower prevalence of AmED among the undergraduate students compared to older ones. Probably, the campaigns implemented to contrast addictions had influence also on the consumption of AmED [64]. However, the overall prevalence of AmED consumption worldwide is still not negligible, especially considering that the studied consumers are young. Besides, the AmED consumption can determine the increase of other risk factors, as hazardous drinking, defined by WHO as a pattern of alcohol consumption that increases the risk of harmful consequences for the user or others [65], and heavier alcohol intake and consequently drunk driving [25][26][27][28]32,[53][54][55][56]58,[61][62][63][64], sexual risk-taking behavior, smoking and drug consumption [24,30,33,46,[49][50][51][52]58], which can be threatening for health and life.
In addition, participants have reported several adverse effects such as high propensity for sensation seeking, impulsiveness, interpersonal consequences and having arguments or fights and headache or dizziness. Other frequently reported adverse effects were difficulty to limit the alcohol quantity or binge drinking. Particularly, binge drinking is defined as consuming 5 or more drinks on an occasion for men or 4 or more drinks on an occasion for women [66]. It is a harmful risk behavior related with serious injuries and several diseases and with an increased risk of alcohol use disorder [67].
Moreover, it should be considered that the quarantines occurred during the COVID-19 pandemic had significantly changed alcohol use in many countries, shifting places of consumption from bars and restaurants to home, mainly thanks to a widely increasing ecommerce. As reported by the Organisation for Economic Co-operation and Development, alcohol sales increased by 3% to 5% in Germany, the United Kingdom and the United States in 2020 compared to 2019 [68]. Even though many individuals reported a decrease or no change in their consumption behaviors during the lockdowns, there has been an increase in frequency and quantity of alcohol use, especially among women, parents of young children, people with higher income and those with anxiety and depressive symptoms in many countries. Some of the problems associated with harmful alcohol consumption and risky behavior such as binge drinking were intensified by the pandemic, even though the long-term impacts of COVID-19 on alcohol consumption are not completely known so far [68,69]. Therefore, it is possible that the pandemic and related control measures have exacerbated even the use of AmEDs, especially among specific population groups, leading to different prevalence rates.
For these reasons, it is essential to understand the consumption motivations and also the possible consequences of AmEDs use, in order to structure new health programs to counteract this phenomenon. AmEDs consumption seems to be related with neuroendocrineindependent brain stress systems, that influence drinking behavior in a dynamic and complex manner [70,71]. Thus, just like the excessive alcohol use as a common response to stress is caused by several triggers, including psychological motivations, as boredom, disruption to routines, distress, also the AmED consumption can be triggered by these reasons [72,73]. However, also physiological causes, such as nutritional deficiencies, dehydration, hormonal changes or the activation of reward-related brain areas, can determine their consumption [74]. Among the examined studies only thirteen investigated the reasons behind the AmED consumption, revealing several common threads beyond to get drunk: to reduce negative effects of alcohol, to relax or to enjoy at parties [23,30,42,57,63] to hide the alcohol's flavor or reduce sedation of alcohol alone [35,48] to treat hangover [34,62,63]. A non-negligible percentage of university students consume AmEDs to hide the flavor of alcohol, to drink more and feel less drunk and this data is in line with the literature [48,52]. Indeed, energy drinks may alter the effects of alcohol through the inhibition of dopamine transmission [75,76]. This aspect is a concern and underlines the tight association between AmEDs and alcohol use.
There are several limitations to our analysis. First of all, the data in each selected study were obtained by self-report survey and recall bias may have interfered with obtaining reliable information. In addition, the heterogeneity of recall period and measures of consumption frequency may represent other biases. Above all, the quality of the majority of the studies was not good, which affect the validity of our findings. However, this review offers a systematic picture of AmEDs consumption and related aspects worldwide, together with the strength of a meta-analytic analysis. In order to reduce the effects of articles' heterogeneity, the meta-regression analysis was controlled for sample size, geographical location, year of publication, time chosen for reporting AmEDs consumption and methodological quality of the studies, together with gender and age of participants. Notwithstanding the reported variability, it seems that only the time of publication and the study location had a slight effect on the results.

Conclusions
Our findings show that a global prevalence rate of 37% is estimated for AmEDs use in undergraduates, with geographical differences. Although a decrease in consumption was observed throughout the period in which the selected studies were published, a continuous monitoring on this phenomenon is needed in order to identify those risk groups that could develop risky behaviors related to AmEDs consumption and provide them targeted educational interventions. The creation of ad hoc surveillance systems could help healthcare systems in controlling the risks possibly related with AmEDs use.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/nu14234985/s1, Figure S1: Funnel Plot Results related to the global prevalence of AmED consumption among undergraduates; Table S1: Detailed search strategy.