Sleep is essential for the optimal health of children and adolescents(Reference Panel, Watson and Badr1). In 2015, a study recommended that adolescents (aged 14–17 years) must have a minimum of 8–10 h of sleep a night(Reference Hirshkowitz, Whiton and Albert2). The nightly sleep duration of adolescents in Korea is 4·9–6·5 h of sleep, which is less than the recommended sleep duration(Reference Noh, Kim and Cheon3).
Poor sleep is associated with disturbances in cognitive and psychomotor functions, including mood, thinking, concentration, memory, learning, vigilance and reaction time(Reference Altevogt and Colten4–Reference Stutts, Wilkins and Osberg6). Chronic sleep deprivation poses a serious threat to the academic success, health and safety of adolescents(Reference Owens, Group and Adolescence7). Poor sleep in adolescence is associated with negative outcomes in several areas of health and functioning, including obesity, depression, school performance, quality of life and risk-taking behaviours(Reference Shochat, Cohen-Zion and Tzischinsky8). Poor sleep is caused by inconsistent class schedules that vary from day to day, early or late obligations, use of technology and stimulants, and consumption of caffeine and energy drinks(Reference Hershner and Chervin9).
In particular, the consumption of energy drinks increases sleep latency, the effects of which can persist for up to 8 h(Reference Hershner and Chervin9). Energy drinks contain high levels of caffeine in combination with ingredients such as amino acids, sugars, sweeteners, guarana, taurine, ginseng, L-carnitine, herbal supplements and vitamin B(Reference Galimov, Hanewinkel and Hansen10). Energy drink consumption has become increasingly widespread among adolescents owing to targeted marketing and the marked effects of providing a quick boost in energy, increasing alertness and ease of availability(Reference Masengo, Sampasa-Kanyinga and Chaput11).
An American study that analysed data from 2003 to 2016 found that energy drink consumption significantly increased among adolescents over that period and that boys consumed energy drinks at a significantly higher level than girls across all age groups(Reference Vercammen, Koma and Bleich12). According to previous studies, energy drinks pose several challenges. First, they are associated with an increase in heart rate and arterial blood pressure. Second, high-energy drink consumption may increase the risk of obesity and type 2 diabetes mellitus. Third, energy drink consumption leads to symptoms of anxiety, insomnia, gastrointestinal upset, muscle twitching, restlessness and periods of inexhaustibility(Reference Al-Shaar, Vercammen and Lu13,Reference Alsunni14) .
Despite their harmful effects, as energy drinks counteract the effects of insufficient sleep, adolescents consume excessive amounts of these drinks(Reference Leal and Jackson15,Reference Grandner, Knutson and Troxel16) . In Korea, adolescent energy drink consumption increased from 2015 to 2020(Reference Kim, Baek and Kim17). The reasons for energy drink consumption were revealed to be taste, followed by the energy boost provided by these drinks. Awakeness has also been reported as a reason(Reference Ruiz and Scherr18). Adolescents who sleep late consume more energy drinks to stay awake(Reference Lebacq, Desnouck and Dujeu19).
Almost previous studies have reported that consumption of energy drinks is associated with poor sleep(Reference Faris, Jahrami and Al-Hilali20,Reference Patrick, Griffin and Huntley21) . While other studies have investigated the association between sleep duration and energy drink consumption or sleep satisfaction and energy drink consumption, our research stands out in that it examines the association between sleep and energy drink consumption using both of sleep duration and sleep satisfaction(Reference Chaput, Tremblay and Katzmarzyk22,Reference Riera-Sampol, Rodas and Martínez23) . Thus, we hypothesise that poor sleep is associated with higher amount of energy drinks consumption.
This study investigated the association between sleep (sleep duration and sleep satisfaction) and energy drink consumption in adolescents using representative data from the 15th Korea Youth Risk Behavior Web-Based Survey (KYRBS) of Korean adolescents.
Methods
Study population and data collection
This study used data from the 15th KYRBS conducted in 2019 by the Korea Disease Control and Prevention Agency (KDCA). The KYRBS is an ongoing national cross-sectional, self-administered, structured questionnaire with a complex research design that includes multistage cluster sampling. The purpose of the KYRBS is to identify health behaviours under the present conditions, report health indicators to plan and appraise health promotion for adolescents, and compare the health indicators of adolescents from different countries. The target participants consisted of middle and high school students from seventeen Korean provinces, including public and private institutions. The stratified cluster random sampling method was used for sampling; individual schools were chosen as the primary sample units (PSU) and selected as the permanent random sample for each cluster. The classroom was considered the secondary sample, and one classroom was randomly extracted for each grade from the selected sample school.
KYRBS uses an online survey system that does not allow respondents to proceed to the next section of the questionnaire unless all questions in the current section are answered. Responses with logical errors or outliers are processed as missing values. The questionnaire contained approximately 120 items across fifteen categories, including demographic characteristics and health-associated behaviours. Middle school students (aged 14–16 years) and high school students (aged 17–19 years) were the target populations in our study. This survey included 57 303 adolescents, of which 50 455 participants were selected for the final analysis after excluding those with missing data.
Ethics approval for KYRBS was waived by the KDCA’s Institutional Review Board in accordance with the Bioethics and Safety Act of 2015. All KYRBS participants provided informed consent. KYRBS fully complied with the rules outlined by the Declaration of Helsinki.
Measurements
Dependent variable
Energy drink consumption was the dependent variable. To measure the same, the question ‘During the last 7 d, how often have you consumed energy drinks?’ was scored on the following scale: haven’t drunk in the last 7 d, 1–2 times a week, 3–4 times a week, 5–6 times a week, 1 time a day, 2 times a day and more than 3 times a day. In this study, participants who did not drink in the last 7 d were reclassified as ‘no’ and the others were considered ‘yes’.
Independent variables
Variables of interest
Sleep characteristics included sleep duration and satisfaction. Sleep duration was answered by the question ‘During the last 7 d, what time did you usually wake up and go to bed?’ for the following two periods: weekdays (Monday to Friday) and weekends (Saturday to Sunday). In this study, sleep duration was calculated by subtracting bedtime on weekdays from waking time and was divided into five categories: less than 5 h, 5–6 h, 6–7 h, 7–8 h, and 8 h or more. Sleep satisfaction was answered by the question ‘During the last 7 d, how satisfied have you felt with your hours of sleep in terms of fatigue recovery?’ The answers were very sufficient, sufficient, just enough, not enough or not enough at all. In this study, we reclassified sufficient and sufficient as ‘sufficient’, just enough as ‘just enough’, and not enough and not enough at all as ‘not enough’.
Covariates
Factors related to energy drink consumption were classified into general and health-related characteristics. General characteristics included the sex, school year, subjective academic performance and subjective economic status. The school year was divided into ‘middle school (14–16 years old)’ and ‘high school (17–19 years old)’. Subjective academic performance was scored based on academic performance in the last 12 months as high, mid-high, medium, mid-low or low. Earlier studies on health behaviour and perceived stress of high school students were classified by reference(Reference Ruiz and Scherr18). The detailed categories were reclassified as ‘medium and above’ and ‘mid-low and lower’, where high, mid-high, and medium were categorised as ‘medium and above’, and mid-low and low were categorised as ‘mid-low and lower’. Subjective economic status was rated as high, mid-high, medium, mid-low or low. In this study, high and mid-high were reclassified as ‘high’, medium as ‘medium’, and mid-low and low as ‘low’.
Health-related characteristics included subjective health status, perceived stress, BMI, asthma history and physical activity. Subjective health status was determined by the question ‘How do you usually rate your health?’ Answers such as ‘very healthy’ and ‘healthy’ were classified as ‘good’, ‘normal’ as ‘normal’, and ‘unhealthy’ and ‘very unhealthy’ as ‘poor’. The level of perceived stress was categorised as: ‘very much’, ‘much’, and ‘a little’ as ‘more stressed’ and ‘not much’ and ‘not at all’ as ‘less stressed’. According to the 2017 Korean Pediatric Society’s standard growth charts for Korean children and adolescents, four categories were created: underweight (≤ 15th quartile), normal weight (16th–84th quartile), overweight (85th–95th quartile) and obesity (≥ 95th quartile or BMI ≥ 25)(Reference Kim, Yun and Hwang24). Asthma history was defined by answering ‘yes’ to the following question: ‘Have you been diagnosed with asthma by a doctor?’ Physical activity was defined by answering the question ‘During the last 7 d, on how many days did your heart rate increase from normal, or did you have a total of 60 min or more per day (regardless of type) of more than normal physical activity?’ The answers were not in the last 7 d, 1 d a week, 2 d a week, 3 d a week, 4 d a week, 5 d a week, 6 d a week or every day. In this study, not in the last 7 d was reclassified as ‘not in the last 7 d’, 1 d a week, 2 d a week, and 3 d a week as ‘1–3 times a week’, 4 d a week and 5 d a week as ‘4–5 times a week’, and 6 d a week and every day as ‘6–7 times a week’.
Statistical analyses
KYRBS data were surveyed via systematic sampling and contained weighted values; hence, statistical analysis was performed by applying weighted values(Reference Choi, Park and Kim25). Independent variables were compared using the chi-square test to determine their association with energy drink consumption. We performed multivariate logistic regression analysis to evaluate the association between poor sleep and higher amount of energy drink consumption as one model for all categories, and the results were reported as adjusted OR (aOR) and 95 % CI. Furthermore, sensitivity analysis, that is, multinomial logistic regression analysis(Reference Shih, Wu and Pan26), was conducted to investigate the association between sleep and consumption amount of the energy drinks after classifying by adjusting the selected covariates. The differences were considered statistically significant level of P-value (P < 0·05). All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc.).
Results
Participant characteristics
In total, 50 445 adolescents from the original sample were included in this study. Participants’ characteristics are presented in Table 1. Sleep pattern characteristics: the majority of respondents are reported sleeping duration about 6–7 h (23·8 %), followed by 5–6 h (22·4 %), 7–8 h (19·9 %), less than 5 h (17·1 %) and more than 8 h (16·7 %). While the recommended sleep duration for adolescents is more than 8 h, the recommended sleep duration satisfaction rate for Korean adolescents was only 16·7 %. Over the past 7 d, almost half of the adolescents, 46·5 %, reported insufficient sleep satisfaction. This was followed by just enough sleep satisfaction reported by 32·6 % and sufficient sleep satisfaction reported by 21·0 %. Regarding energy drink consumption, 30·4 % Korean adolescents reported that they consumed energy drinks more than once a week.
Summary statistics (n 50 455)
* Days when the heart rate increases from normal or is more than 60 min total per d with more physical activity than normal (regardless of type).
In general, most of the participants were boys (51·5 %) and high school students (52·4 %). They had mid-low or lower subjective academic performance (68·9 %) and a medium subjective economic status (48·1 %). Regarding health-related behaviours, most of the participants had a good subjective health status (70·6 %), perceived more stress (80·9 %), underweight (72·3 %), no experience of asthma (93·0 %) and reported 1–3 d a week of physical activity (43·4 %) (Table 1).
Differences in energy drink consumption according to participant characteristics
Regarding the relationship between sleep characteristics and energy drink consumption, energy drink consumption was reported by 42·0 % of those with less than 5 h of sleep, 33·2 % of those with 5–6 h, 29·4 % with 6–7 h, 24·9 % with 7–8 h and 22·5 % with 8 h or more (P < 0·001). Regarding sleep satisfaction among those who reported consuming energy drinks, 25·1 % were satisfied, 28·9 % said they had just enough sleep and 33·8 % reported insufficient sleep and consuming energy drinks (P < 0·001).
Regarding general characteristics, 32·1 % of boys and 28·6 % of girls reported energy drink consumption (P < 0·001); by school year, 26·6 % were in middle school (14–16 years old) and 33·8 % were in high school (17–19 years old); and by subjective academic performance, 29·5 % had mid-low and lower performance records and 32·3 % had mid-level and higher records (P < 0·001). Subjective economic status was high in 30·9 %, medium in 29·6 % and low in 32·8 % of participants (P = 0·0013).
Regarding health-related behaviours, among adolescents consuming energy drink was reported as follows: based on subjective health status, 29·3 % reported good, 30·4 % reported normal and 35·0 % reported bad (P < 0·001); based on perceived stress, 31·7 % reported being more stressed and 24·9 % reported being less stressed (P < 0·001); by BMI, 29·9 % were normal weight, 28·4 % were underweight and 32·7 % were overweight or obese (P < 0·001); as for asthma history, 30·2 % were in the ‘No’ group and 33·2 % were in the ‘Yes’ group (P = 0·0008). Regarding physical activity, energy drink consumption was reported as follows: 28·2 % reported ‘not in the last 7 d’, 31·4 % reported ‘1–3 d a week’, 30·9 % reported ‘4–5 d a week’ and 33·4 % reported ‘6–7 d a week’ (P < 0·001; Table 2).
General characteristics of the participants by energy drinks consumption
Values are presented as numbers (%).
Relationship between participant characteristics and energy drink consumption
Regarding sleep duration characteristics, shorter sleep duration was related to energy drink consumption. It was found that adolescents with fewer than 5 h of sleep (OR, 2·36; 95 % CI (2·14, 2·60)) consumed more energy drinks than adolescents with 7–8 h of sleep (OR, 1·12; 95 % CI (1·03, 1·23)), 6–7 h of sleep (OR, 1·38; 95 % CI (1·26, 1·51)) and 5–6 h of sleep (OR, 1·62; 95 % CI (1·48, 1·78)). Those with lower sleep satisfaction consumed more energy drinks than those with sufficient sleep satisfaction (aOR, 1·07; 95 % CI (1·01, 1·14) for just enough and aOR, 1·14; 95 % CI (1·07, 1·21) for not enough). Therefore, sleep (sleep duration and sleep satisfaction) has a negative relationship with energy drink consumption.
The sex-based energy drink consumption showed that boys had higher consumption than girls (aOR, 0·76; 95 % CI (0·72, 0·80)), whereas high school students (aOR, 1·07; 95 % CI (1·01, 1·14)) consumed energy drinks more often than middle school students. Regarding subjective academic performance (grades), we found that mid-low and lower-grade students (aOR, 1·12; 95 % CI (1·06, 1·18)) and those with a high economic status (aOR, 1·10; 95 % CI (1·03, 1·19)) consumed energy drinks more than their selected reference group. However, there was no difference between middle subjective economic status and energy drink consumption (aOR, 0·99; 95 % CI (0·93, 1·06)).
For subjective health status, energy drink consumption was higher for those with poorer health (aOR, 1·11; 95 % CI (1·05, 1·17) for normal health; aOR, 1·14; 95 % CI (1·05, 1·23) for poor health). The proportion of respondents who were more stressed and reported energy drink consumption was higher than those who were less stressed (aOR, 1·25; 95 % CI (1·18, 1·33)). For BMI, the proportions of those who were overweight and obese who reported energy drink consumption were higher than those who belonged to the normal weight category (aOR, 1·07; 95 % CI (1·02, 1·13)). However, there was no difference between being underweight and energy drink consumption (aOR, 0·94; 95 % CI (0·86, 1·03)). In addition, we did not observe any difference between asthma history and energy drink consumption. Finally, for physical activity, the proportion reporting energy drink consumption among those who engaged in physical activity was higher than that among those who did not engage in physical activity (aOR, 1·18; 95 % CI (1·12, 1·24) for 1–3 times a week; aOR, 1·20; 95 % CI (1·11, 1·29) for 4–5 times a week; aOR, 1·35; 95 % CI (1·24, 1·47) for 6–7 times a week) (Table 3).
Logistic regression on energy drink consumption
Relationship between characteristics of sleep and amount of energy drink consumption
Table 4 and Figure 1 illustrate the multinomial logistic regression results of the sensitivity analysis. The relationship between sleep duration and amount of energy drink consumption showed that adolescents with less than 5 h of sleep (aOR, 6·37; 95 % CI (4·72, 8·61)) had the highest relationship with energy drink consumption (i.e. at least once a day), followed by 5–6 times per week (aOR, 4·61; 95 % CI (3·36, 6·34)) and 3–4 times per week (aOR, 3·29; 95 % CI (2·76, 3·94)). The results showed a significant difference (P < 0·05). In terms of sleep satisfaction, the adolescents who had not enough sleep satisfaction (aOR, 1·44, 95 % CI (1·16, 1·78)) were likely to consume more drinks about 5–6 times per week. However, except for at least once per d (P = 0·489), the other categories showed partially significant differences.
Sensitivity analysis: multinomial analysis of the association between sleep and energy drink consumption
aOR, adjusted OR.
Result of sensitivity analysis of multinomial logistic regression between sleep duration and amount of energy drink (ED) consumption
Discussion
This study used a large sample of 2019 KYRBS data from a national survey based on a sample of Korean adolescents. As hypothesised, poor sleep was associated with a higher amount of consumption energy drink consumption. Our results showed that adolescents who did not sleep long enough or had lower sleep satisfaction consumed energy drinks at higher frequencies. In addition, our study found associations between energy drink consumption and boys, high school students, high subjective academic status, bad subjective health status, more perceived stress, and more physical activity. However, mixed results were found for associations between energy drink consumption and subjective economic status and BMI.
Our findings are consistent with those of previous studies(Reference Riera-Sampol, Rodas and Martínez23,Reference Kim, Yun and Hwang24,Reference Sanchez, Martinez and Oriol27,Reference Kaldenbach, Leonhardt and Lien28) . Sampasa-Kanyinga et al. found that shorter sleep duration was associated with higher energy drink consumption(Reference Sampasa-Kanyinga, Hamilton and Chaput29). Lohsoonthorn et al. found that lower sleep satisfaction was related to energy drink consumption in adolescents(Reference Lohsoonthorn, Khidir and Casillas30). However, previous studies have reported that the consumption of energy drinks is associated with poor sleep, which can impact the onset of sleep and reduce sleep time, efficiency, and satisfaction levels(Reference Faris, Jahrami and Al-Hilali20,Reference Patrick, Griffin and Huntley21) . Conversely, we found that poor sleep was associated with the consumption of energy drinks. Several studies have investigated the association between sleep and energy drink consumption, but this study was the first to examine the association between sleep (sleep duration and sleep satisfaction) and energy drink consumption(Reference Chaput, Tremblay and Katzmarzyk22,Reference Riera-Sampol, Rodas and Martínez23,Reference Kaldenbach, Leonhardt and Lien28) . Moreover, this study adds to the growing body of literature on the relationship between sleep and unhealthy beverages. For example, adolescents who slept for a short duration and had low satisfaction with this level of sleep had higher odds of consuming sugar-sweetened beverages(Reference Grummon, Sokol and Lytle31,Reference Ward, Jospe and Morrison32) . To the best of our knowledge, this is the first study in Korea to collectively investigate sleep duration and sleep satisfaction to determine their association with energy drink consumption.
Furthermore, the covariates were consistent with those of previous studies. Boys were more likely to consume energy drinks than girls, and high school students were more likely to consume energy drinks than middle school students. Our results are consistent with the finding that the consumption of energy drinks more than once a week was more frequent among boys than girls(Reference Leal and Jackson15,Reference Grandner, Knutson and Troxel16,Reference Lebacq, Desnouck and Dujeu19) , which is consistent with previous studies showing that high school students consume more energy drinks than middle school students(Reference Chaput, Tremblay and Katzmarzyk22,Reference Choi, Park and Kim25,Reference Park, Lee and Lee33) . Adolescents who engaged in physical activity in our study consumed more energy drinks than those who did not. Park et al. determined that the consumption of energy drinks was higher among those who engaged in physical activity 7 or more times a week compared to those who did so twice a week or less(Reference Park, Onufrak and Blanck34). Some studies have found that physical activity is associated with improved sleep quality(Reference Shih, Wu and Pan26–Reference Sampasa-Kanyinga, Hamilton and Chaput29) and higher energy drink consumption(Reference Faris, Jahrami and Al-Hilali20–Reference Kim, Yun and Hwang24). However, we observed no evidence of an association between overall sleep quality and physical activity in our study. Instead, the more stressed groups and those with mid-low or lower academic performance records consumed more energy drinks.
Many prior studies have suggested that adolescents consume energy drinks to stay awake(Reference Lebacq, Desnouck and Dujeu19,Reference Ishak, Ugochukwu and Bagot35) . Korean adolescents’ academic performance is vital for their psychosocial development and preparation to enter college and adulthood. High school students in Korea attend classes between 08:00 and 17:00. Nearly, all Korean high school students attend private after-school institutions or stay at school for evening studies that may continue until 22:00 h(Reference Park and Kim36). Under the strong sociocultural and psychological influence of college entrance examinations, most Korean adolescents further curtail their sleep duration(Reference Bae and Wickrama37). Sleep dissatisfaction causes tiredness/fatigue(Reference Al-Shaar, Vercammen and Lu13). Furthermore, a previous study showed that shorter sleep duration is related to daytime sleepiness and poor daytime functioning(Reference Owens, Group and Adolescence7). Consequently, daytime sleepiness occurs owing to a lack of sleep, which leads to energy drink consumption to stay awake(Reference Do, Kang and Kim38). However, our results showed that the mid-low and lower academic groups consumed more energy drinks than the medium and above groups. In fact, people believe that consuming energy drinks helps them with their homework and study(Reference Champlin, Pasch and Perry39). Many students currently believe that using energy drinks will facilitate their study and work on school projects(Reference Malinauskas, Aeby and Overton40). However, the effects of high caffeine consumption on stress and sleep can lead to inefficient learning(Reference Park and Kim36,Reference Pettit and DeBarr41) .
It would be better for adolescents to get sufficient sleep; however, this may not be feasible because of their many classes and studies(Reference Yang, Kim and Patel42). Poor sleep and energy drink consumption are not beneficial to adolescent health, and public health attention and management are needed. Sleep hygiene education improves sleep quality by providing comprehensive education regarding the harmful effects of caffeine, tobacco and alcohol. It encourages adolescents to build exercise routines, reduce stress, and manage their sleep and napping times(Reference Hirshkowitz, Whiton and Albert2). Thus, sleep hygiene education may be considered a method to address the issue of energy drink consumption.
Limitations
This study has some limitations. First, although results showed the association that poor sleep was associated with higher amount of energy drinks consumption, the data were obtained from a cross-sectional survey; thus, this study can only indicate statistical associations and has a problem of reverse causality. Second, the data were based on self-reported data provided by the respondents online. The survey respondents may have understood or overstated their responses. Third, despite the significant emotional understanding of Korean characteristics of increasing adolescents’ anxiety under the strong sociocultural and psychological influence of college entrance examinations, there were data limitations such as the lack of test anxiety in the variables. Therefore, future studies should address the association between sleep and energy drink consumption, including psychological variables such as academic anxiety and stress caused by academic demands. Fourth, this study used data from 2019, which may have affected its interpretation. The data may have been amplified or attenuated during the pandemic period.
Conclusion
In conclusion, this study showed that poor sleep is associated with higher amount of energy drinks consumption among Korean adolescents. As the number of adolescents who consume energy drinks is increasing, efforts should be made to provide sleep hygiene education to enable adolescents to better understand sleep hygiene, improve sleep duration and satisfaction, and reduce the consumption of energy drinks.
Acknowledgements
None.
Financial support
This research did not receive any specific grants from funding agencies in the public, commercial or not-for-profit sectors.
Conflicts of interest
There are no conflicts of interest.
Authorship
D.H.K. and T.H.K. contributed to the study conception and design. D.H.K., B.K. and T.H.K. performed the statistical analyses. D.H.K., B.G.K., T.H.K. and S.G.L. wrote the first draft of this manuscript. T.H.K. and S.G.L. contributed to manuscript revision and read and approved the submitted version.
Ethics of human subject participation
This study was conducted in accordance with the guidelines laid down in the Declaration of Helsinki, and the ethics approval for KYRBS was waived in 2015 by the Korea Centers for Disease Control and Prevention IRB under the Bioethics & Safety Act and opened to the public for academic use. All KYRBS participants provided informed consent.
Open access
