Energy drink consumption and substance use risk in middle school students.

OBJECTIVES: Energy drink (ED) sales have increased greatly in recent years and ED is now a common topic in health behavior research. Most studies work with samples of college students and/or young adults and to a lesser degree with high school students. Research is lacking on ED consumption in younger users. The purpose of this study was to fill this gap and assess the prevalence of ED consumption in a sample of middle school students as well to analyze the relationships between ED use and illicit and licit drug use in girls and boys of this age.
METHOD: We use cross-sectional school-survey data from 6-8th grade students in three US mid-Atlantic schools conducted in September to November 2014 (N = 1152, response rate: 82.4%).
RESULTS: Approximately 20% of participants had consumed ED and 10% had smoked cigarettes. Almost 14% had used alcohol at least once in their lifetime and 5.5% marijuana. Boys were more likely than girls to have used ED but no gender difference was observed in the prevalence of illicit substances. However, ED use was positively related to smoking and alcohol use among both genders, but also to several forms of illicit drug use among girls.
CONCLUSION: ED consuming girls are particularly prone to also use illicit substances. This is the first survey-type study which reports a positive relationship between ED consumption with both licit and illicit drug use in middle school-aged girls and boys.

Introduction

The sales of energy drinks (EDs) have increased greatly in recent years (Pomeranz, 2012). A signal of this increase is a surge in hospital emergency department cases of young people consuming excessive amounts of energy drinks (Cotter et al., 2013). ED contains caffeine in concentrations similar to that of regular coffee or between 70 and 130 mg of caffeine per 12 oz (Heckman et al., 2010). Studies have shown that around 70% of high-school aged adolescents consume ED regularly and as many as 20% use them every day or most days (James et al., 2011, Kristjansson et al., 2013a). In extant adolescent health literature on ED, most attention has been devoted to the relationship between ED and alcohol use, particularly the consumption of alcohol mixed with ED (AmED) (e.g., Azagba et al., 2014, Azagba and Sharaf, 2014). This line of research has shown a strong relationship between AmED, heavy drinking and increased odds of alcohol dependence (Marczinski et al., 2013). Additionally, adolescent ED literature has reported associations between ED use and increased odds of sexual risk behaviors (Snipes and Benotsch, 2013), injuries (Hamilton et al., 2013), poorer academic outcomes (James et al., 2011), violent behaviors and conduct disorders (Kristjansson et al., 2013a), and licit and illicit drug use (Arria et al., 2010), to name a few.

However, in the current adolescent ED paradigm studies have been conducted almost exclusively with samples of young adults (e.g., Snipes and Benotsch, 2013), college students (e.g., Woolsey et al., 2015), and less commonly with high-school aged youth (e.g., Kristjansson et al., 2015). Although review studies on caffeine have highlighted ED's potential for harm in younger populations (e.g, Temple, 2009, Schneider et al., 2011), very little research has been conducted with younger participants including those of middle school age. Notable exceptions are Kristjansson et al. (2014) which found a positive relationship between consumption of ED and other caffeine containing soft drinks and several forms of physical complaints in 10–12 years old children, Luebbe and Bell (2009) which discovered a relationship between caffeine use in 5th graders and increased risks for withdrawal-related anxiety and depressive affect, and Miyake and Marmorstein (2015) that found a positive longitudinal relationship between ED use and alcohol use 16 months later using a small sample of 7th and 8th grade students from a single school. Although a relationship has been established between ED consumption and substance use in older adolescents, no large-scale study has identified such a relationship in middle school-aged children to date.

Further, recent evidence has highlighted some important gender differences with regard to patterns of consumption of caffeinated products and its relations to behaviors among children and youth (Temple et al., 2010, James et al., 2015). Some of this literature has shown that girls may be more prone to conduct problems in combination with their caffeine consumption from ED (Kristjansson et al., 2011, Kristjansson et al., 2013a). Nevertheless, as with other published studies in the area of ED and caffeine use among youth, this literature has operated with samples of high school students, college students, and/or young adults. Thus, the two objectives of this study were to 1) assess the relationship between ED use and alcohol, tobacco and other drug use (ATOD) in middle school children, and 2) test for gender differences in both the prevalence of ED use as well as the relationships between ED use and ATOD while employing a large sample of middle school girls and boys.

Method

Sample and procedures

Data for this study were collected as part of the Integrated Community Engagement (ICE) Collaborative in West Virginia. ICE is an ongoing health promotion effort that uses school-survey data to inform stakeholders about risk and protective factors for substance use. Upon passive parental consent (all children were included in the study unless their parents/caregivers opted them out), survey data was collected by teachers under the supervision of a school contact agent that operated as a liaison to the research team. All accessible students in grades 6–8 in three mid-Atlantic middle schools participated in the study (N = 1152. Response rate 82.4%). The study was reviewed and approved by West Virginia University's IRB (protocol # 1406345394). For additional methodological details, this data collection approach and process has been described previously by Kristjansson et al. (2013b).

Measures

The question on ED use was created by the research team for this study but based on Kristjansson et al., 2013a, Kristjansson et al., 2013b, Kristjansson et al., 2015. The substance use questions in the ICE project have been largely adopted from national and international studies, specifically the Monitoring the Future survey in the US (Johnston et al., 2014) and the European School Project on Alcohol and other Drugs (Hibell et al., 2012).

Energy drinks (EDs)

ED use was assessed with the question “During the past 7 days how many times did you drink a can, bottle, or glass of caffeinated energy drinks such as Monster, Red Bull, XL, Full Throttle, or Rock Star?” Response categories ranged from 1 = “I did not drink caffeinated energy drinks during the past 7 days” to 7 = “4 or more times per day”. For the present analyses this variable was dichotomized with 0 = “not used ED in past 7 days” and 1 = “used ED 1 + times in past 7 days”.

Smoking

Tobacco smoking was assessed with the following two questions: 1) “Have you ever tried cigarette smoking, even just one or two puffs? Response categories were 1 = “yes” and 2 = “no”. 2) “During the past 12 months, how often have you smoked cigarettes?” Response categories ranged from 1 = “never”, to 7 = “40 times or more often”. This variable was also dichotomized with 0 = “never” and 1 = “ever”.

Alcohol use

Alcohol consumption was assessed with the following three questions: 1) “Have you ever had a drink of alcohol, other than a few sips?”, 2) “Have you ever been drunk from drinking alcohol?” Response categories for 1 and 2 were 1 = “yes” and 2 = “no”. 3) “During the last 12 months, how often have you had a drink of alcohol (including; beer, wine, wine coolers, and liquor such as rum, gin, vodka, or whiskey) other than a few sips?” Response categories were the same as for 12 month cigarette smoking and were dichotomized with 0 = “never” and 1 = “ever”.

Other drug use

The use of other drugs was assessed with 5 questions that were headed with “In your lifetime, have you ever used the following drugs or substances?” and the questions were 1) “Marijuana (also called weed or pot) or Hashish (also called hash or hash oil)?”, 2) “Sniffed glue, breathed the content of spray cans, or inhaled any paints or sprays to get high?”, 3) “Taken a prescription drug (such as OxyContin, Percocet, Vicodin, Codeine, Adderall, Ritalin, or Xanax) without a doctor's prescription?”, 4) “Synthetic marijuana (also known as K2 or spice) to get high?”, and 5) “‘Bath Salts’ (synthetic stimulants) to get high?” Response categories were coded with 1 = “yes” and 2 = “no”.

Control variables

Four control variables were used in the present analyses. 1) “Family structure”, coded 1 = “lives with both parents” and 2 = “other forms”. 2) “Race/ethnicity”, coded 1 = “white” and 2 = “other”. 3) Separate questions were asked for the highest educational levels of mothers and fathers. Response categories ranged from 1 = “elementary or middle level education or less” to “a graduate degree”.

Analyses

Bivariate analyses for this report were conducted with Chi-Square tests. Multivariable analyses were done using logistic regression for categorical data with separate models run for boys and girls for each outcome variable. Within each outcome variable gender difference in the relationship between ED and the outcome was tested with a mean-centered ED*Gender variable while also including gender in the respective model.

Results

Table 1 shows the prevalence rates for ED and other drugs use in the study population. Approximately 20% of participants reported having ever used ED, around 10% smoked a cigarette at least once in their lifetime, and almost 14% had ever used alcohol. Overall, few gender differences were observed for the prevalence rates of study variables. Boys were significantly more likely than girls to have used ED and ever smoked a cigarette. Other substance use was less prevalent for both genders. Table 2 shows the relationship between the ED use and 10 forms of substance use for boys and girls. All smoking and alcohol measures were positively predicted by ED for both genders. Marijuana use, the use of inhalants, prescription drugs, synthetic marijuana and bath salts were also predicted by ED use among girls but not boys. In six out of ten gender comparisons, the relationship between ED use and the outcome was significantly stronger for girls than boys, and in two instances the relationship between ED use and the outcome was stronger for boys.

Table 1

Frequencies and descriptive statistics for all study variables (95% CI) (N = 1130).

Variables	Boys %	Girls %	Total %	P value for gender diff.
Energy drinks	25.2 (21.5–29.3)	15.6 (12.7–19.0)	20.3 (17.9–22.9)	.001
Ever smoking	12.0 (9.5–15.2)	7.8 (5.8–10.4)	9.8 (8.2–11.8)	.023
Smoking in last 12 months	5.4 (3.7–7.7)	3.3 (2.1–5.2)	4.3 (3.2–5.7)	.110
Ever alcohol	15.8 (12.9–19.3)	12.1 (9.6–15.1)	13.9 (11.9–16.1)	.080
Alcohol in last 12 months	12.0 (9.5–15.2)	8.6 (6.5–11.2)	10.2 (8.5–12.2)	.067
Ever drunkenness	3.0 (1.9–5.0)	3.2 (2.0–5.0)	3.1 (2.2–4.4)	.892
Ever marijuana	6.5 (4.6–9.0)	4.67 (3.1–6.7)	5.5 (4.2–7.1)	.191
Ever inhalants	3.9 (2.5–6.1)	4.8 (3.3–7.0)	4.4 (3.3–5.8)	.516
Ever prescription medication	3.7 (2.4–5.8)	3.6 (2.3–5.6)	3.7 (2.7–5.0)	.918
Ever synthetic marijuana	3.4 (2.1–5.4)	3.1 (1.9–4.9)	3.2 (2.3–4.5)	.787
Ever bath salts	2.5 (1.4–4.3)	2.5 (1.5–4.2)	2.5 (1.7–3.7)	.985
Lives with both parents	64.6 (60.5–68.4)	57.9 (53.8–61.9)	61.2 (58.3–64.0)	.022
White	79.2 (75.6–82.4)	82.0 (78.7–85.0)	80.7 (78.2–82.9)	.237
Mothers education (X-, SD)	2.56 (2.20)	2.94 (2.40)	2.75 (2.31)	.009
Fathers education (X-, SD)	2.79 (2.45)	2.92 (2.56)	2.86 (2.50)	.390

Table 2

Energy drink consumption predicting alcohol, tobacco and other drug use: odds ratios (OR) from logistic regression models, controlling for race/ethnicity, family structure and parental education.

Dependent variables	Boys OR (95% CI)	Girls OR (95% CI)	P value for gender differences
Ever smoking	2.55 (1.33–4.89)	5.99 (2.77–12.93)	.002
Smoking in last 12 months	11.66 (3.69–36.84)	8.17 (2.59–25.71)	.008
Ever alcohol	2.99 (1.70–5.24)	7.17 (3.71–13.88)	.007
Alcohol in last 12 months	3.49 (1.88–6.48)	8.58 (4.11–17.92)	.003
Ever drunkenness	8.23 (2.03–33.38)	3.84 (1.15–12.86)	.005
Ever marijuana	2.19 (.91–5.25)	4.51 (1.70–12.00)	.026
Ever inhalants	2.39 (.80–7.14)	5.44 (2.12–13.98)	.001
Ever prescription medication	1.43 (.43–4.80)	4.04 (1.34–12.17)	.016
Ever synthetic marijuana	2.16 (.66–7.05)	3.61 (1.08–12.09)	.056
Ever bath salts	2.11 (.49–9.13)	2.19 (.51–9.47)	.959

Discussion

This is the first study to use a large-scale school survey to measure ED use and its relationship to other substance use in middle school students in the US. Employing a sample of over 1100 middle school students, our findings show that around one in five middle school students have used ED and that ED use is significantly related to smoking and alcohol use for both girls and boys but also to other forms of substance use among girls. Our findings also indicate that the relationship between ED and several forms of substance use is stronger for girls than boys. These results echo previous findings (James et al., 2015, Temple et al., 2010) which have highlighted gender differences in health and behavioral outcomes associated with caffeine use in children and adolescents. However, the nature of this gender difference is yet to be understood. One hypothesis that has been proposed (James et al., 2015) is that ED and caffeine use are often considered a mild form of delinquency and as such are more common and more socially accepted for boys than girls. The apparent consequence is that more boys than girls use ED on a consistent basis, which is in keeping with our findings, while girls who use ED represent a higher risk group than non-ED using girls. This is also in line with findings by Kristjansson et al., 2013a, Kristjansson et al., 2013b which showed that the caffeine consumption among adolescent high school-aged girls is significantly more strongly related to both violent behaviors and conduct disorders than for boys of the same age category. Future research should aim to explore these gender differences further.

Four limitations are worth noting about our study. First, our data is based solely on self-reports. We are therefore unable to rule out responses without foundation. Second, although we received a high response rate to our survey we rely on data from only three schools in the mid-Atlantic area of the US. Generalizability is therefore at best limited. Third, the cross-sectional nature of our data analyses precludes us from making assumptions regarding cause and effect. Fourth, our measurement for cigarette smoking is somewhat unusual as we do not ask for 30 day levels of smoking in this study population. Although it is not sensible to conclude from our analyses that ED use may lead to smoking, alcohol use, or other substance use in middle school-aged children, our findings should be interpreted as probable to that end. Unfortunately, in the current research paradigm studies on ED and other forms of caffeine use have been scarce, especially in this age group. More research is needed to establish an understanding of the long-term implications of ED use in middle school-age students, preferably using prospective research designs. Researchers should strive to include ED and other forms of caffeine in longitudinal studies while also measuring substance use more broadly.

Conflict of interest declaration

None of the authors declares any conflict of interest.