at www.jneb.org

Podcast available online Research Article

Adolescent Consumption of Sports and Energy Drinks:Linkages to Higher Physical Activity, Unhealthy BeveragePatterns, Cigarette Smoking, and Screen Media UseNicole Larson, PhD, MPH, RDN1; Jessica DeWolfe, MPH1; Mary Story, PhD, RD1,2;Dianne Neumark-Sztainer, PhD, RD1

1DivisioMinneso2GlobalAddressand ComSecond SE-mail: l�2014 Shttp://dx

Journal

ABSTRACT

Objective: To examine patterns of adolescent sports and energy drink (SED) consumption and identifybehavioral correlates.Design: Data were drawn from Eating and Activity in Teens, a population-based study.Setting: Adolescents from 20middle and high schools inMinneapolis/St Paul, MN completed classroom-administered surveys.Participants: A total of 2,793 adolescents (53.2% girls) in grades 6–12.Variables Measured: Beverage patterns; breakfast frequency; moderate to vigorous physical activity(MVPA); media use; sleep; and cigarette smoking.Analysis: Linear and logistic regression models were used to estimate associations between health behav-iors and SED consumption, adjusting for demographics.Results: Over a third of adolescents consumed sports drinks and 14.7%consumed energydrinks at least oncea week. Among boys and girls, both sports and energy drink consumptionwere related to higher video gameuse; sugar-sweetened beverage and fruit juice intake; and smoking (P < .05). Sports drink consumption wasalso significantly related to higher MVPA and organized sport participation for both genders (P < .01).Conclusions and Implications: Although sports drink consumptionwas associated with higherMVPA,adolescents should be reminded of recommendations to consume these beverages only after vigorous,prolonged activity. There is also a need for future interventions designed to reduce SED consumption,to address the clustering of unhealthy behaviors.Key Words: adolescents, sports drinks, energy drinks, dietary intake, physical activity, sleep patterns(J Nutr Educ Behav. 2014;46:181-187.)

Accepted February 17, 2014.

INTRODUCTION

Adolescent consumption of sugar-sweetened beverages is of concernbecause consumption is associatedwith increased risk for dental caries,excess weight gain, and poor dietquality.1-4 Although recent evidenceshowed a decline from 1999 to 2008in the prevalence of soft drink andfruit drink consumption among USadolescents, sports and energy drinkconsumption tripled.5 Sports and

n of Epidemiology and Community Heta, Minneapolis, MNHealth and Community and Family Mfor correspondence: Nicole Larson, PhDmunity Health, School of Public Het, Suite 300, Minneapolis, MN 55454;arsonn@umn.eduOCIETY FOR NUTRITION EDUC.doi.org/10.1016/j.jneb.2014.02.008

of Nutrition Education and Behav

energy drinks are considered to besugar-sweetened beverages alongwith soft drinks and flavored juicedrinks, but their ingredients and pur-ported functions differ. Sports drinksare noncarbonated, flavored drinksthat contain added sugars, minerals,and electrolytes to help replenish thebody during vigorous exercise.6 En-ergy drinks contain high amounts ofcaffeine, often coupled with othernatural stimulants that enhancecaffeine's effects, and may also

alth, School of Public Health, University of

edicine, Duke University, Durham, NC, MPH, RDN, Division of Epidemiology

alth, University of Minnesota, 1300 SouthPhone: (612) 625-5881; Fax: (612) 626-7103;

ATION AND BEHAVIOR

ior � Volume 46, Number 3, 2014

contain vitamins, minerals, protein,and added sugars. Beverages marketedas energy drinks are purported toincrease mental concentration aswell as physical performance.6

The US Food and Drug Administra-tion considers energy drinks to bedietary supplements not subject tofood marketing or ingredient regula-tions.7,8 Therefore, whereas theaverage caffeine content per fluidounce for the 4 top-selling soft drinksin the US was 3.5 mg from 2002 to2006, the average caffeine contentfor the 4 top-selling energy drinkswas 9.6 mg.8 Little information isavailable on the effects of the supple-ments or stimulants in energydrinks, especially among childrenand adolescents. The high caffeinecontent of energy drinks, as well asthe high sugar and calorie content ofmany sports and energy drinks, hasdrawn much concern from health

181

182 Larson et al Journal of Nutrition Education and Behavior � Volume 46, Number 3, 2014

professionals as their consumptionamong adolescents has increased.9

In June, 2011, the American Acad-emy of Pediatrics (AAP) issued a reportexamining the marketing, ingredients,and possible negative effects of sportsand energy drinks.9 They concludedthat small amounts of sports drinkscould be appropriate for young peopleparticipating in vigorous physical ac-tivity inhot, humidweather. However,for the average young athlete, sportsdrinks are unnecessary and couldcontribute to negative health out-comes, such as excess weight gain andtooth decay.9 With regard to energydrinks, the AAP concluded that thereis no benefit provided, and becausethese drinks create a risk for overstimu-lation of the nervous system, theyshould not be consumed by adoles-cents. Consumption of energy drinksby young people has resulted in casesof seizure, myocardial arrhythmia,and even death.7-10

Although soft drink and overallsugar-sweetened beverage consump-tion among adolescents has beenwell studied, few studies in adolescentpopulations have examined the con-sumption of sports and energy drinksor factors associated with theirconsumption. One study found thatsports and fruit drink consumptionwere unrelated to weight status andwere associated with higher levels ofphysical activity as well as consump-tion of milk, fruit, and vegetables,but were also associated with negativehealth behaviors, such as higher con-sumption of energy-dense foods andsedentary behavior.11 In addition, arecent study found that sports andenergy drink consumption were loweramong black and low-income adoles-cents.5 To the best of the currentauthors' knowledge, no study has spe-cifically examined other factors thatmay be associated with sports and en-ergy drink consumption among ado-lescents. To address this gap in theliterature, and to inform public healthefforts targeting consumption, thecurrent study used data from apopulation-based survey to describepatterns of sports and energy drinkconsumption among adolescents,and to examine associations withdietary factors, physical activity,media use, hours of sleep, andcigarette smoking. These factors wereselected for consideration based on

their potential for modification toimprove overall health if addressedas part of interventions targetingsports and energy drink consumptionalong with other risk behaviors.

METHODSStudy Design and Population

Data were drawn from Eating andActivity in Teens (EAT 2010), apopulation-based study designedto examine dietary intake, physicalactivity, weight control behaviors,weight status, and factors asso-ciated with these outcomes inadolescents.12,13 Surveys and anthro-pometric measures were completedby 2,793 adolescents during the2009–2010 academic year. The studypopulation included adolescents from20 public middle schools and highschools in the Minneapolis/St Paulmetropolitan area of Minnesota.Consideration was given to enroll-ment size and diversity as well asinvolvement in other research studieswhen recruiting school districts.Personnel at the participating schoolsreported that no sports or energydrinks were available for purchase onschool grounds. The mean age of thestudy population was 14.4 years (SD,2.0 years); 46.1% were in middleschool (sixth through eighth grades)and 53.9% were in high school(ninth through 12th grades).Participants were equally divided bygender (46.8% boys and 53.2% girls).The racial/ethnic composition of thestudy population was diverse, with81.1% of participants reporting abackground other than non-Hispanicwhite. With regard to socioeconomicstatus (SES), participants were distrib-uted across 5 strata based primarilyon parental educational attainment:38.4% low SES, 21.3% low-middleSES, 16.9% middle SES, 12.4% upper-middle SES, and 7.3% high SES (3.7%did not report SES).

Trained research staff administeredsurveys and measured adolescents'height and weight during selectedhealth, physical education, andscience classes. Measurements werecompleted in a private area andsurveys were administered following astandard protocol during 2 50-minuteclass periods. After survey completion,participants were given a $10 gift card.

All study procedures were approved bythe University of Minnesota Institu-tional Review Board Human SubjectsCommittee and by the research boardsof the participating school districts.Adolescents were given the opportu-nity to assent only if their parent orguardian did not refuse by returning asigned letter that explained the studypurpose and procedures. Among ado-lescents at school on survey administra-tion days, 96.3% had parental consentand chose to participate.

Survey Development andMeasures

The EAT2010 survey is a 235-item, self-report instrument designed to assess arange of factors of potential relevanceto weight status and weight-relatedbehaviors among adolescents. Surveydevelopment was guided by expert re-view and extensive pilot testing withadolescents.14 The estimates of mea-sure test-retest reliability reportedbelow were determined for a 1-weekperiod in a diverse sample of 129 mid-dle school and high school students.

Sports and energy drink intake. Theresearchers assessed usual past-yearintake of sports and energy drinkson the EAT 2010 survey using 2 ques-tions, which separately asked aboutthe frequency of consuming eachtype of drink (response categories:never or < 1/mo, 1–3/mo, 1/wk,2–4/wk, 5–6/wk, 1/d, and$ 2/d). Pop-ular brand name examples were givenfor energy drinks (Red Bull, FullThrottle, and Rockstar) and sportsdrinks (Gatorade and Powerade).Regular consumption of sports andenergy drinks was defined as at least1 drink/wk (test-retest agreement:81.0% for sports drinks and 83.3%for energy drinks) based on the distri-bution of intake in the sample.

Physical activity and sport participa-tion. Moderate to vigorous physicalactivity (MVPA) was measured byseparately asking participants howmany hours they spent doing stren-uous and moderate activity in atypical week, with several examplesof each level of activity provided.15

Response categories were: none,< 0.5 hour, 0.5–2 hours, 2.5–4 hours,4.5–6 hours, and$ 6 hours (test-retest

Journal of Nutrition Education and Behavior � Volume 46, Number 3, 2014 Larson et al 183

reliability for total MVPA: r ¼ 0.73).Sports participation was assessed byasking the number of sports teams inwhich the adolescent was involvedin the past year (test-retest reliability:r¼ 0.86).16 This measure was dichoto-mized to represent none vs any sportsteam participation in the past year.

Media use. Participants indicated theaverage number of leisure-time hoursthey spent playing video games(Xbox/PlayStation/other electronicgames played when sitting) andwatching television (TV), DVDs, orvideos separately for both weekdaysand weekend days.17,18 Responsecategories for each question were:0 hours, 0.5 hour, 1 hour, 2 hours, 3hours, 4 hours, and $ 5 hours; theseresponses were scored 0, 0.5, 1, 2, 3,4, and 6 hours, respectively, to createcontinuous variables representingtotal weekly hours of use (test-retestreliability: r ¼ 0.84 for video games;r ¼ 0.67 for TV/DVDs/videos).

Sleep. The researchers assessed sleephabits using measures drawn from apreviously validated questionnaire.19

Participants were asked when theyusually went to bed and got out ofbed, separately for weekdays andweekend days. Average weekdayand weekend day sleep durationswere calculated from the timesprovided.20 Overall average dailysleep duration was calculated as(weekday duration � 5 / 7) þ (week-end day duration � 2 / 7).

Cigarette smoking. Smoking statuswas determined from reported ciga-rette use during the past year(nonsmoker ¼ never used; ever-smoker ¼ used at least a few times).Test-retest agreement for this measurewas high (92.9%).

Breakfast frequency and otherbeverage intake. To assess breakfastconsumption, adolescents were asked,‘‘During the past week, how manydays did you eat breakfast?’’ Fiveresponse options ranged from ‘‘never’’to ‘‘every day’’ (test-retest reliability,r ¼ 0.76). Dietary intake of sugar-sweetened soft drinks, fruit drinks,100% fruit juice, and milk wereassessed with the semiquantitative,

149-item Youth and Adolescent FoodFrequency Questionnaire. A beverageserving was defined as the equivalentof 1 glass or can. The validity andreliability of this tool have beenpreviously examined and found tobe within acceptable ranges for die-tary assessment.21,22

Weight status. Height and weightmeasurements were completed usingstandardized procedures. Body massindex was calculated and sex- andage-specific percentiles were deter-mined using reference data from theCenters for Disease Control and Pre-vention growth tables.23 Percentileswere used to classify respondents asoverweight (85th to < 95th) or obese($ 95th).

Demographics. Grade level, race/ethnicity, and SES were assessed byself-report. Classification tree method-ology was used to generate 5categories of SES.24 The prime determi-nant of SES was the higher educationlevel of either parent. Subsidiaryvariables were student eligibility forfree/reduced-price school meals, fam-ily public assistance receipt, and parentemployment status.

Statistical Analysis

All analyses were conducted using theStatistical Analysis System (version9.3, SAS Institute Inc, Cary, NC,2011) and stratified by adolescentgender. Descriptive statistics were firstcalculated to examine patterns ofsports and energy drink consumption.The chi-square statistic was used toidentify differences in consumptionaccording to demographic character-istics and weight status. The authorsused linear and logistic regressionmodels that included a random effectto account for clustering withinschools to estimate associations be-tween health behaviors and regularsports and energy drink consumption,while adjusting for grade level, SES,and race/ethnicity. One set of modelsincluded frequency of sports drinkconsumption as the independentvariable and each health behavior ofinterest (ie, MVPA, sport participa-tion, media use, dietary intake, break-fast frequency, sleep, and cigaretteuse) was included in a separate model

as the dependent variable. A secondset of models likewise includedfrequency of energy drink consump-tion as the independent variable toexamine associations with eachdependent health behavior variable.The overall F value for each modelexamined was statistically significantat P < .05. To estimate the total vari-ance in health behaviors explainedby consumption of sports and energydrinks along with all of the covariatesconsidered here, nonhierarchicalmodels were further examined thatincluded school identifier as a covari-ate rather than as a random effect.Nonhierarchical models with sportsdrink consumption as the indepen-dent variable explained 4% to 19%of variance in health behaviors, andthe nonhierarchical models withenergy drink consumption as the in-dependent variable explained 4% to18% of variance in health behaviors.A 95% confidence level was used tointerpret the statistical significanceof probability tests. Whenever thedependent variable exhibited positiveskewness, such testing was carried outunder the square root transformation.

RESULTSFrequency of Consumption andAssociations WithDemographic Characteristics

Overall, sports and energy drinks wereconsumed regularly (at least 1/wk) by37.9% and 14.7% of the adolescentsample, respectively. Regular con-sumption of sports drinks amongboys was 44.9% and 31.6% amonggirls (P < .001). For energy drinks, reg-ular consumption was 17.1% amongboys and 12.5% among girls(P < .001). There was a moderate tolow correspondence between intakeof sports and energy drinks, with10.6% of adolescents regularlyconsuming both sports and energydrinks, 26.8% regularly consumingsports drinks but not energy drinks,and 3.6% consuming energy drinksbut not sports drinks. Consumptionof sports drinks was highest amongboys and girls who reported that theirrace/ethnicity was black or mixed/other (Table 1). However, similar dif-ferences by race/ethnicity in energydrink consumption were statisticallysignificant only among girls.

Table 1. Demographic Variables and Frequency of Sports and Energy DrinkConsumption Among Adolescent Boys (n ¼ 1,307) and Girls (n ¼ 1,486)

Characteristic

Sports Drink Intake Energy Drink Intake

Boys Girls Boys Girls

Race/ethnicityWhite 37.4 20.5 15.8 8.6Black 52.8 36.3 19.1 15.5Hispanic 44.9 32.8 13.8 12.8Asian 35.0 27.3 15.1 8.8Mixed/other 55.0 38.4 22.3 15.6P < .001 < .001 .14 .01

Socioeconomic statusLow 45.6 30.1 18.3 12.6Middle-low 48.8 37.3 17.3 12.8Middle 42.5 30.5 16.7 13.7Middle-high 46.7 27.1 12.6 8.4High 38.1 26.1 15.0 12.5P .33 .10 .52 .58

Body mass index categoryNot overweight 43.5 31.1 17.8 11.7Overweight 46.0 31.9 16.9 10.7Obese 46.5 32.0 14.8 16.6P .60 .95 .50 .06

Grade levelMiddle school 43.9 33.8 16.4 13.8High school 45.7 29.7 17.9 11.4P .52 .09 .50 .17

Note: Data represent the percentage of adolescents who had $ 1 drink/wk.P represents overall testing by sex for independence of demographic categoriza-tion and weekly sport or energy drink consumption using the chi-square test.

184 Larson et al Journal of Nutrition Education and Behavior � Volume 46, Number 3, 2014

Sports Drink Consumption andHealth Behaviors

Regular sports drink consumption wasassociated with measures of physicalactivity, media use, other beverageintake, and cigarette smoking(Table 2). For both genders, regularsports drink consumption was associ-ated with higher MVPA, greater likeli-hood of past-year participation in anorganized sport, and higher videogame use. Among boys, regular sportsdrink consumption was also signifi-cantly associated with higher TVviewing; boys who regularlyconsumed sports drinks spent about1 additional hour per week watchingTV, compared with boys whoconsumed sports drinks less thanonce per week. With regard to dietarybehaviors, among both genders therewas a significant positive associationbetween regular sports drink con-sumption and daily consumption ofsugar-sweetened soft drinks and fruit

drinks. In addition, there was a signif-icant association between regularsports drink consumption and everhaving smoked cigarettes for bothgenders.

Energy Drink Consumption andHealth Behaviors

Regular energy drink consumptionwas similarly associated with mea-sures of media use, other beverageintake, and cigarette use, but wasunrelated to measures of physicalactivity (Table 3). For both genders,regular energy drink consumptionwas positively associated with weeklyvideo game use. As an example, ina usual week, boys who regularlyconsumed energy drinks spentapproximately 4 additional hoursplaying video games, compared withthose who consumed energy drinksless than once per week (Table 3).Among the dietary factors examined,

for both genders, there was a signifi-cant association between regularenergy drink consumption and higherdaily intake of sugar-sweetened softdrinks and fruit drinks. For girls only,there was also a significant associationbetween regular energy drink con-sumption and lower frequency ofbreakfast. In addition, there was a sig-nificant positive association betweenever having smoked cigarettes andregular energy drink consumptionfor both genders.

DISCUSSION

The results of this study indicate thatat least weekly consumption of bothsports and energy drinks amongadolescents is significantly associatedwith higher consumption of othersugar-sweetened beverages, cigarettesmoking, and screen media use. Theobserved associations betweenconsumption of sports and energydrinks and these unhealthy behaviorsare troubling because they may indi-cate a clustering of problem behaviorsamong some adolescents. Given thesefindings, evidence of increasingadvertising of sports and energydrinks to youth is particularly con-cerning. One study found that expo-sure to TV advertisements for energydrinks increased 23% among childrenand 20% among adolescents from2008 to 2010.25 Research further indi-cated that sports and energy drinks aredisproportionately targeted to blackyouth; in a media exposure study,black children and adolescents wereexposed to over twice as many adver-tisements for sports and energy drinksin 2010 as white youth, after adjust-ing for TV viewing time.25 Based onthe results of this media exposureresearch and the current study, thereis a need for advocacy efforts to limitadvertising to all racial/ethnic groupsof adolescents and complementpublic health interventions thatdirectly target sports and energy drinkconsumption. Recently, energy drinkcompanies have hired lobbyists tofight the US Food and Drug Adminis-tration investigation into the safetyof their products, which signals aneven greater need for public healthadvocacy in this area.26

In addition to unhealthy behav-iors, sports drink consumption was

Table 2. Sports Drink Consumption Frequency and Health Behaviors Among Adolescent Boys and Girls

Boys’ Sports Drink Intake Girls’ Sports Drink Intake

Behavior< 1/Wk, Mean

(95% CI)$ 1/Wk, Mean

(95% CI) P< 1/Wk, Mean

(95% CI)$ 1/Wk, Mean

(95% CI) P

Physical activityModerate to vigorous activity, h/wk 4.9 (4.0–5.7) 6.9 (6.3–7.5) .003 3.9 (3.1–4.6) 4.9 (4.0–5.7) .004Sport participation (%) 56 (52–60) 76 (72–80) .003 48 (44–2) 66 (61–71) .004

Media useVideo games, h/wk 15.2 (12.6–17.7) 18.4 (16.0–20.7) .008 6.1 (4.5–7.7) 7.8 (5.9–9.7) .02Television viewing, h/wk 18.2 (16.5–19.8) 19.2 (17.6–20.7) .03 17.3 (15.4–19.2) 17.6 (15.5–19.6) .60

Dietary factorsSugared soda, servings/d 0.37 (0.26–0.49) 0.55 (0.46–0.63) < .001 0.46 (0.39–0.52) 0.59 (0.52–0.66) .02Fruit drinks, servings/d 0.50 (0.35–0.64) 0.66 (0.52–0.79) .002 0.59 (0.47–0.70) 0.78 (0.62–0.93) .003Fruit juice, servings/d 0.56 (0.32–0.83) 0.81 (0.57–1.0) .005 0.92 (0.64–1.2) 1.3 (1.0–1.5) .005Milk, servings/d 1.4 (1.1–1.7) 1.6 (1.1–1.9) .46 1.3 (1.1–1.4) 1.4 (1.1–1.6) .42Breakfast, frequency/wk 4.7 (4.2–5.2) 4.5 (3.9–5.2) .30 3.8 (3.2–4.3) 3.6 (3.0–4.2) .20

Other behaviorsSleep, mean h/d 9.3 (8.9–9.7) 9.2 (8.8–9.6) .21 9.6 (9.3–9.9) 9.4 (9.0–9.7) .01Ever smoked cigarettes (%) 7.1 (5.1–9.1) 13.5 (10.5–16.5) .02 7.4 (5.4–9.4) 13.7 (11.1–16.2) .02

CI indicates confidence interval for the estimated means of the behavior. P values represent significance of differences of means.Note: All tests controlled for race/ethnicity, grade, and socioeconomic status.

Journal of Nutrition Education and Behavior � Volume 46, Number 3, 2014 Larson et al 185

also significantly associated withhigher levels of MVPA. These resultsmay reflect evidence demonstratingthat sports drinks are heavily mar-keted to adolescent athletes.25,27 Themajority of advertisements for sportsand energy drinks directed towardadolescents depict these drinks as

Table 3. Energy Drink Consumption Frequ

Behavior<

Physical activityModerate to vigorous activity, h/wk 5.Sport participation (%) 6

Media useVideo games, h/wk 16.Television viewing, h/wk 18.

Dietary factorsSugared soda, servings/d 0.4Fruit drinks, servings/d 0.5Fruit juice, servings/d 0.6Milk, servings/d 1.Breakfast, frequency/wk 4.

Other behaviorsSleep, mean h/d 9.Ever smoked cigarettes (%) 8.

CI indicates confidence interval for the estimNote: All tests controlled for race/ethnicity

‘‘hip or cool,’’ promoting optimalfitness, or having beneficialnutrients.25 For example, the Gator-ade Company and the NationalAthletic Trainers' Association recentlypartnered to establish NationalRecovery Day and promote the con-sumption of fluids with sodium (ie,

ency and Health Behaviors Among Adolesce

Boys’ Energy Drink Intake

1/Wk, Mean(95% CI)

$ 1/Wk, Mean(95% CI) P

< 1

7 (4.9–6.5) 6.4 (5.7–7.0) .31 4.3 (57–66) 70 (64–76) .08 5

2 (13.7–18.6) 20.5 (18.2–22.8) .005 6.5 (17.1–19.9) 19.7 (17.9–21.6) .19 17.

2 (0.33–0.51) 0.70 (0.52–0.87) .01 0.67 (0.44–0.70) 0.72 (0.55–0.88) .03 0.45 (0.38–0.92) 0.90 (0.55–1.2) .03 0.95 (1.1–1.8) 1.5 (1.1–2.0) .70 1.6 (4.0–5.2) 4.4 (3.8–5.0) .28 3.

3 (8.9–9.7) 9.0 (8.6–9.4) .10 9.7 (6.7–10.7) 19.7 (14.6–24.7) .005 7.

ated means of the behavior. P values repres, grade, and socioeconomic status.

sports drinks) before, during, and afterexercise.27 The campaign messagespromoting consumption of sportsdrinks were widely disseminated toathletes of all ages despite contradic-tion with the AAP recommendationsfor adolescent use of sports drinks,which emphasize they are not

nt Boys and Girls

Girls’ Energy Drink Intake

/Wk, Mean(95% CI)

$ 1/Wk, Mean(95% CI) P

3 (3.5–5.0) 4.5 (3.4–5.7) .283 (50–56) 61 (52–70) .10

5 (4.7–8.2) 8.9 (6.6–11.2) .0030 (15.0–19.0) 18.3 (15.5–21.1) .46

1 (0.41–0.53) 0.86 (0.62–0.85) .0017 (0.49–0.72) 0.74 (0.70–1.0) .018 (0.74–1.2) 1.2 (0.94–1.5) .044 (1.2–1.5) 1.3 (1.0–1.5) .458 (3.2–4.3) 3.2 (2.5–3.7) .01

5 (9.2–9.8) 9.3 (8.9–9.6) .068 (6.1–9.5) 21.8 (15.8–27.8) .01

ent significance of differences of means.

186 Larson et al Journal of Nutrition Education and Behavior � Volume 46, Number 3, 2014

necessary for the average adolescentathlete.9,27,28 Thus, exposure toadvertisements for sports drinks maylead adolescents to believe that thesebeverages are beneficial for activitiesthat do not involve vigorous andprolonged exertion.

IMPLICATIONS FORRESEARCH ANDPRACTICE

Although this study used a compre-hensive survey with reliable measuresto examine characteristics associatedwith sports and energy drinkconsumption among adolescents, notall survey measures were validatedand additional research is needed toaddress certain limitations. Singleitems were used to separately assessoverall intake of sports and energydrinks for the current study, but thereare numerous types of these drinkson the market that vary considerablyin nutritional composition (eg, low-or no-sugar beverages). Because of thecross-sectional design, it is not possibleto establish temporality. In addition,although the sample was large anddiverse, the results may not apply toother adolescent populations withhigher levels of sports and energydrink consumption for whom theremay be larger associations with adversehealth-related outcomes. Futurestudies should incorporate longitudi-nal designs when feasible and examineadditional factors of relevance toconsumption (eg, stress, substanceuse) as well as reasons for observed as-sociations using validated measures inother population-based samples.Because of the finding that electronicgaming was relevant for both sportsand energy drink consumption, it willbe particularly important for studiesto develop measures for investigatingspecific linkages to various gamingplatforms (eg, mobile devices, videogame consoles). Research should alsoevaluate intervention designs thatencourage appropriate hydrationamong adolescents, in particular waterconsumption, and target linkagesbetween sports and energy drink useand other unhealthy behaviors.

In conclusion, the findings of thisstudy and other research indicatethat sports and energy drinkconsumption are prevalent among ad-olescents.5,10 National data indicate

that 12% of adolescents consume asports or energy drink on a givenday.5 The national prevalence ofconsuming other sugar-sweetenedbeverages such as soft drinks (51%)and fruit drinks (23%) remains muchhigher among adolescents; however,consumption of sports and energydrinks is increasing and the potentialfor serious consequences when energydrinks are consumed in combinationwith alcohol justifies the need forpublic health concern.5,10 Althoughsports drink consumption was foundto be associated with higher levels ofMVPA, it is unlikely that consum-ption only followed vigorous,prolonged activity as recommendedby the AAP, and both sports andenergy drink consumption were alsoassociated with several unhealthybehaviors.9 These findings suggestthat future sugar-sweetened beverageinterventions should consider target-ing the clustering of unhealthy behav-iors among some adolescents, such assmoking, high screen media use, andbreakfast skipping. Nutrition profes-sionals, pediatricians, coaches, andteachers can also make an impactthrough continued public healthadvocacy focusing on the marketingand availability of sports and energydrink products to youth. Furthermore,these professionals can work toeducate youth and parents about thepotential consequences associatedwith consumption of sports and en-ergy drinks, and encourage healthyhydration habits.

ACKNOWLEDGMENT

This work was supported by GrantR01HL084064 from the NationalHeart, Lung, and Blood Institute (PI:Dianne Neumark-Sztainer). Thecontent is solely the responsibility ofthe authors and does not necessarilyrepresent the official views of theNational Heart, Lung, and BloodInstitute or the National Institutes ofHealth.

REFERENCES

1. Malik V, Popkin B, Bray G, Despr�es J,Hu F. Sugar-sweetened beverages,obesity, type 2 diabetes mellitus, andcardiovascular disease risk. Circulation.2010;121:1356-1364.

2. Ebbeling C, Feldman H, Chomitz V,et al. A randomized trial of sugar-sweetened beverages and adolescentbody weight. N Engl J Med. 2012;367:1407-1416.

3. Berkey CS, Rockett HR, Field AE,Gillman MW, Colditz GA. Sugar-added beverages and adolescent weightchange. Obes Res. 2004;12:778-788.

4. Shenkin J, Heller K, Warren J,Marshall T. Soft drink consumptionand caries risk in children and adoles-cents. Gen Dent. 2003;51:30-36.

5. Han E, Powell L. Consumption pat-terns of sugar-sweetened beverages inthe United States. J Acad Nutr Diet.2013;113:43-53.

6. Galemore C. Sports drinks and energydrinks for children and adolescents—are they appropriate? A summary ofthe clinical report. NASN Sch Nurse.2011;26:320-321.

7. Rath M. Energy drinks: what is all thehype? The dangers of energy drink con-sumption. J Am Acad Nurse Pract. 2012;24:70-76.

8. Reissig C, Strain E, GriffithsR. Caffein-ated energy drinks—a growing prob-lem. Drug Alcohol Depend. 2009;99:1-10.

9. Committee on Nutrition and theCouncil on Sports Medicine andFitness. Sports drinks and energy drinksfor children and adolescents: are theyappropriate? Pediatrics. 2011;127:1182-1189.

10. Seifert S, Schaechter J, Hershorin E,Lipshultz S. Health effects of energydrinks on children, adolescents, andyoung adults. Pediatrics. 2011;127:511-528.

11. Ranjit N, Evans M, Byrd-Williams C,Evans A, Hoelscher D. Dietary andactivity correlates of sugar-sweetenedbeverage consumption among adoles-cents. Pediatrics. 2010;126:e754-e761.

12. Neumark-Sztainer D, Wall M,Larson N, et al. Secular trends in weightstatus and weight-related attitudes andbehaviors in adolescents from 1999 to2010. Prev Med. 2012;54:77-81.

13. Larson N, Wall M, Story M, Neu-mark-Sztainer D. Home/family, peer,school, and neighborhood correlatesof obesity in adolescents.Obesity (SilverSpring). 2013;21:1858-1869.

14. Berge J, Wall M, Larson N,Eisenberg M, Loth K, Neumark-Sztainer D. The unique and additiveassociations of family functioning andparenting practices with disorderedeating behaviors in diverse adolescents.J Behav Med. 2013;21:1858-1869.

Journal of Nutrition Education and Behavior � Volume 46, Number 3, 2014 Larson et al 187

15. Godin G, Shephard R. A simplemethod to assess exercise behavior inthe community. Can J Appl Sport Sci.1985;10:141-146.

16. Brener N, Collins J, Kann L,Warren C,Williams B. Reliability of the YouthRisk Behavior Survey Questionnaire.Am J Epidemiol. 1995;141:575-580.

17. Gortmaker SL, Peterson K, Wiecha J,et al. Reducing obesity via a school-based interdisciplinary interventionamong youth: Planet Health. ArchPediatr Adolesc Med. 1999;153:409-418.

18. Nelson M, Neumark-Sztainer D,Hannan P, Sirard J, Story M. Longitu-dinal and secular trends in physicalactivity and sedentary behavior duringadolescence. Pediatrics. 2006;118:e1627-e1634.

19. Pasch K, Laska M, Lytle L, Moe S.Adolescent sleep, risk behaviors, anddepressive symptoms: are they

linked? Am J Health Behav. 2010;34:237-248.

20. Meyer K, Wall M, Larson N, Laska M,Neumark-Sztainer D. Sleep durationand BMI in a sample of young adults.Obesity (Silver Spring). 2012;20:1279-1287.

21. Rockett H, Wolf A, Colditz G.Development and reproducibility ofa food frequency questionnaire toassess diets of older children and ado-lescents. J Am Diet Assoc. 1995;95:336-340.

22. Rockett H, Breitenbach M, Frazier A,et al. Validation of a youth/adolescentfood frequency questionnaire. PrevMed. 1997;26:808-816.

23. Kuczmarski R, Ogden C, Grummer-Strawn L, et al. CDC growth charts:United States.AdvData. 2000;314:1-27.

24. Neumark-Sztainer D, Story M,Hannan PJ, Croll J. Overweight status

and eating patterns among adolescents:where do youths stand in comparisonwith the Healthy People 2010 objec-tives? Am J Public Health. 2002;92:844-851.

25. Sugary Drink f.a.c.t.s.: Food Adver-tising to Children and Teens Score.Yale Rudd Center for Food Policy andObesity. http://www.sugarydrinkfacts.org/resources/SugaryDrinkFACTS_Report.pdf. Accessed January 24, 2014.

26. Wilkie C. Energy drinks hire lobbyiststo fend off regulation. http://www.huffingtonpost.com/2013/01/24/energy-drinks-lobbyists-regulation_n_2546622.html. Accessed January 24, 2014.

27. Cohen D. The truth about sportsdrinks. BMJ. 2012;345:e4737.

28. Heneghan C, Howick J, O’Neill B,et al. The evidence underpinning sportsperformance products: a systematicassessment. BMJ Open. 2012;2.