Caffeine Awareness in Children: Insights from a Pilot Study

INTRODUCTION

Caffeine is the most commonly ingested psychoactive substance and its use is ubiquitous.^1–4 It is naturally found in coffee, tea, and cocoa but it is also a common ingredient in a host of commercial beverages that are routinely consumed by children.⁵ The consumption of caffeine by children and adolescents is increasing.⁶

The pattern of caffeine intake is different between adults and children, with children preferring soft drinks as their primary source of caffeine.⁷ Soft drinks may contain 30–55 mg of caffeine per 12 oz.⁸ Although the US Food and Drug Administration has not established limits for pediatric caffeine consumption, Canadian guidelines recommend no more than 45 mg/day of caffeine in children aged 4 to 6 y, 62.5 mg/day for ages 7–9 y, and 85 mg/day for ages 10–12 y.⁹ Although caffeine's effects on adults have been studied, much less is known about its effects on children and adolescents. Recently, Temple et al.¹⁰ showed that acute caffeine consumption in 12–17 year old children has significant physiological effects on heart rate and diastolic blood pressure in a dose-dependent manner. Caffeine has also been linked to anger in adolescents,¹¹ as well as even violent behavior and conduct disorder.⁸

In terms of sleep consequences, a cross-sectional study of 4,243 school-aged children found a twofold increased risk in sleep disturbances in school-aged and adolescent children who drank either coffee or soft drinks.¹² Further, research shows that although adolescents require between 8.5 and 9.5 h of sleep per night, they frequently get much less sleep than this.¹³ Adolescents also commonly report variable weekday/weekend bedtimes, and daytime sleepiness.¹⁴ In a 2011 National Sleep Foundation (NSF) poll, more than one-fourth (28%) of Generation Z'ers reported sleeping 2 h longer on weekends than week-days as a way to “catch up” on their sleep.¹⁵ Daytime sleepiness itself may trigger increased caffeine use,¹⁶ which can in turn lead to delayed sleep onset.¹⁷

Given that caffeine has an elimination half-life as long as 10 h,¹⁸ the effects of caffeine can be long lasting. Various measures have been suggested in relation to the use of caffeine in pediatric populations with sleep disturbances. These range from avoiding caffeine several hours before bedtime to restricting its use after 14:00.¹⁹

Although some pediatric literature describes the effects of caffeine on sleep and daytime functioning,^14,20–23 there is little or no literature describing caffeine awareness in children. Although previous polls such as the ones conducted by the NSF in 2006 and 2011 asked whether the children consumed caffeinated beverages such as tea, soda, coffee, and energy drinks, they did not ask about specific kinds or brands of beverages consumed by the children.^15,24 Hence, the conclusions of such polls were a priori dependent on responders being able to reliably distinguish between caffeinated and noncaffeinated beverages.

We report preliminary data from a survey of seventh and eighth graders at a single public school in Cuyahoga County, OH, using the pilot Caffeine Literacy and Sleep Study (CLASS) questionnaire in conjunction with the Cleveland Adolescent Sleepiness Questionnaire (CASQ).²⁵ The objective of this study was to evaluate if children are aware of the caffeine content in drinks and sodas commonly found on grocery store shelves. We hypothesized that: (1) a majority of adolescents are unaware of the caffeine content of common beverages, and (2) adolescents may wrongly perceive that light-colored or clear drinks do not contain caffeine.

METHODS

School principals of several public and private schools in the Cleveland metropolitan area were approached by initial telephone contact to identify potential study sites. The first school to agree to participate in this study was chosen as the target school. According to information obtained from the US census website, this school is located in a community of about 40,000 people with 7% of the population in the 10- to 14-y age group and 7.5% in the 15- to 19-y age group.²⁶ The community is predominantly African American (73.7%) with 20.7% Whites and 1.5% Hispanic or Latino population. The median family household income is $30,505 with 20.4% of the families living below the poverty level. There are 85.8% of adults who have at least a high school degree, and 32.5% have a bachelor's degree or higher educational status.

The CLASS and CASQ survey instruments were used after approval of the protocol by the Cleveland Clinic Institutional Review Board. The questionnaires were designed to be anonymous and brief. The CLASS survey is a 15-question, non-validated pilot instrument designed by the study authors to evaluate children's recall of caffeinated beverage intake by type, quantity, and timing, as well as prior-week sleep habits by recall (Figure 1). The survey assesses awareness of presence or absence of caffeine in common beverages. The CLASS survey instrument was intentionally restricted to 15 items in order to make it easier to complete in the classroom.

The CASQ is a validated instrument used to measure excessive daytime sleepiness in adolescents.²⁵ It has good internal consistency (Cronbach alpha, 0.89). The CASQ has been shown to be highly positively correlated with sleepiness scores as assessed by the School Sleep Habits Survey (SSHS) and the Pediatric Daytime Sleepiness Scale (PDSS).²⁵ CASQ scores are also significantly correlated with age and sleep parameters such as sleep debt and sleep duration on school days and weekends. In addition, CASQ has been shown to have a significant correlation with the apnea- hypopnea index (AHI) in children with sleep disordered breathing.

Prior to administering the questionnaires, a letter was sent through the school offices addressed to the parents of prospective survey participants to inform them of the survey and give them an opportunity to opt out. The students were given another opportunity to opt out, immediately prior to the distribution of the questionnaire in a predetermined class at the participating school. Following this, both questionnaires were distributed and collected during the specified class period. All seventh- and eighth-grade school students who attended school that day and did not opt out were included. Statistical analyses were performed using SPSS software (version 22.0. Armonk, NY: IBM Corp.). Between-group comparisons were made using t test. Proportions were compared using Chi-square test. Significance was set at a level of 0.05. Data are presented as mean ± standard deviation unless otherwise indicated.

RESULTS

Of the 635 seventh- and eighth-grade students who attended school on the day of the study, 555 (87%) participated (55.5% male, 57.3% in seventh grade, median age 12 and 13 y for seventh and eighth grade, respectively). Of the 80 students who opted out, eight were due to parental refusal. The remaining 72 students opted out at the time the questionnaires were distributed in the school. Table 1 shows the characteristics of students with regards to their sleep habits, self-reported caffeine intake, and ability to identify the drinks with the most and least caffeine as well as the presence and absence of caffeine in their favorite drink. Only about 14% reported no caffeinated beverage use. However, nearly 29% of the children were unaware that their favorite drink contained caffeine and more than 50% were unable to correctly identify the drinks with the most caffeine content.

Table 1 Caffeine intake and sleep habits of seventh and eighth graders.

A low percentage of students correctly identified light-colored sodas lacking caffeine: 7-Up (24.1%), Sierra Mist (38.9%), ginger ale (39.8%), Sprite (39.8%), and Fresca (53.7%) (Figure 2A). Boys were incorrect about ginger ale more often than girls (60.7% versus 38.3%, p < 0.02, data not shown). The percentages of students correctly identifying caffeinated light-colored beverages were: Arizona Green Tea (43.5%), Mello Yellow (50.9%), A&W cream soda (67.6%), and Mountain Dew (93.5%) (Figure 2B).

In terms of sleep habits, over 77% reported sleeping around or before 22:00 on weekdays but less than 19% did so on weekends (Figure 3A). CASQ scores differed significantly between those sleeping before versus after midnight on weekends (p < 0.0001; Figure 3B) and those sleeping around or before 22:00 versus later on weeknights (p < 0.0001; Figure 3C).

DISCUSSION

Sodas are a common source of caffeine among adolescents⁷ and are associated with daytime sleepiness, insufficient sleep, and poorer sleep quality.¹² Although ours is a pilot study with limited questions, it had a high response rate without parental bias due to in-school administration. The main aim of our study was to assess if children were aware if specific drinks, particularly light-colored beverages, contained caffeine. We found significant lack of knowledge in children about the caffeine content in common drinks.

Several previous studies and national polls have attempted to assess caffeine consumption in adolescents.^12,15,22,24 For example, the 2006 NSF Sleep in America Poll of adolescents ages 11 to 17 y asked how many cups or cans of caffeinated beverages (such as soda, energy drinks, coffee, tea, iced coffee, or iced tea) they typically drink each day.²⁴ The brand names of these drinks were not queried in the 2006 NSF poll. Similarly, the 2011 NSF poll did not query the brand names of caffeinated drinks consumed by children.¹⁵

Although the aforementioned literature discusses caffeine intake and sleepiness, the methods did not address children's awareness of particular drinks having or lacking caffeine. Our study suggests that children have low awareness of caffeine content in common beverages, particularly light-colored beverages. This lack of knowledge may have significant implications because it may potentially confound the results of population surveys reporting on caffeine use in children. Given the adverse effects of higher doses of caffeine consumption, our study identifies a potential target for educating adolescents about the dangers of inadvertently consuming excessive caffeine due to their lack of knowledge about the caffeine content in common beverages.

In keeping with prior published data, our data too show significant differences between weekday and weekend sleep times and sleepiness scores of teens. However, these previously reported data may be biased because our study found significant lack of knowledge about caffeine content in common beverages in this population.

Our study suffers from several limitations. First, this was a pilot study with a limited sample size. As such, the study may not have been adequately powered to detect small differences between groups. Second, the data have been obtained from a single Cleveland-area middle school. As such, the generalizability of the results to all adolescents or even to non-urban schoolchildren is limited. Third, little is known about the characteristics of the children who opted not to participate in the study or of those who did not attend school that day. Whether the results would have been different had those children agreed to participate is open to speculation. Fourth, we selected drinks commonly found on grocery store shelves, and may have missed some of the “fad” drinks—the caffeinated drinks that students are actually choosing on their own. Such drinks may include “energy drinks” which are specifically marketed to youth and contain higher amounts of caffeine than soft drinks (70–130 mg/12 oz.).^8,27 Future studies should incorporate the assessment of the knowledge about an expanded selection of popular drinks (especially the energy drinks) and also assess current consumption choices. Fifth, because we were primarily interested in caffeine awareness and because the time allotted for in-classroom survey administration was limited, we were forced to keep the questionnaire brief. As a result, some potentially important factors were not assessed. These include, among others, the children's awareness of the long half-life of caffeine and their demographic details. Sixth, the CLASS is a nonvalidated instrument, which planned future studies will address. Seventh, the lack of reliability of caffeine intake evident in our study precluded any analysis with respect to sleep-wake times.

Despite these limitations, our study adds new knowledge to the field by identifying that adolescents have low awareness of the caffeine content of common beverages and paves the way for future studies to target caffeine literacy and educational efforts in this population of students.

DISCLOSURE STATEMENT

This was not an industry supported study. The authors have indicated no financial conflicts of interest. The work was performed at the Cleveland Clinic Sleep Disorders Center, Cleveland, OH.

REFERENCES