Preventive Medicine 52 (2011) 218–222

Contents lists available at ScienceDirect

Preventive Medicine

j ourna l homepage: www.e lsev ie r.com/ locate /ypmed

Health behaviors and obesity among US children with attention deficit hyperactivitydisorder by gender and medication use

Juhee Kim a,⁎, Bala Mutyala a, Stamatis Agiovlasitis b, Bo Fernhall a

a Dept. of Kinesiology and Community Health, University of Illinois at Urbana Champaign, 1206 S. 4th St, 213 Huff Hall, Champaign, IL 61820, USAb Department of Kinesiology, Mississippi State University, 233 McCarthy Gym., Mississippi State, MS 39762, USA

⁎ Corresponding author at: 1206South Fourth St., 213HCommunityHealth,University of Illinois atUrbana-ChampFax: +1 217 333 2766.

E-mail address: juheekim@illinois.edu (J. Kim).

0091-7435/$ – see front matter © 2011 Elsevier Inc. Aldoi:10.1016/j.ypmed.2011.01.003

a b s t r a c t

a r t i c l e i n f o

Available online 15 January 2011

Keywords:Physical activitySedentary behaviorsDepressionObesityNational Survey of Children's Health (NSCH)ADHDMedication

Objective. We examined the levels of physical activity, sedentary behaviors, and obesity among childrenwith attention deficit hyperactivity disorder (ADHD) by gender and medication use and estimated theassociations between health behaviors and obesity.

Methods. Cross-sectional analysis of children 6–17 years-old enrolled in the National Survey of Children'sHealth 2003 (n=66,707). Odds ratios were adjusted for multistage-sampling and survey-design effects.

Results. ADHD prevalence was 8.6%. In general, children with ADHD engaged in less physical activity,organized sports, and reading than their counterparts. Children with ADHD had increased risk of obesity forboys [24.9% vs. 21.6%, OR(95% CI): 1.42(1.13–1.77)] and girls [21.9% vs. 16%, 1.85(1.26–2.73)], if notmedicated. Only girls with ADHD and not on medication were more likely to have higher media time (52.7%

vs. 42%) and this was associated with higher odds for obesity [27.7% vs. 19.5%, 2.51 (1.24–5.08)]. ChildrenwithADHD on medication had higher prevalence of depression than those not taking medication [boys: 29.5% vs.26.3%; girls: 30.9% vs. 23.6%] and the odds of being depressed remained significant after controlling for obesity[boys: 1.45 (1.09–1.94); girls: 2.27 (1.48–3.49)].

Conclusions. Health promotion and obesity prevention programs targeting children with ADHD shouldtake gender and medication use into consideration.

© 2011 Elsevier Inc. All rights reserved.

Introduction

Attention-deficit disorder/attention-deficit hyperactivity disorder(ADHD) is a neurobehavioral pathology affecting 7.8% of US childrenaged 4–17 years (Visser et al., 2007). ADHD often coexists withlearning disabilities, conduct disorder, anxiety, depression, bipolardisorder (National Institute of Mental Health, 2003; Waxmonsky,2003) and developmental coordination disorder (Harvey and Reid,2003;Watemberg et al., 2007). Youth with ADHD have problemswithinterpersonal relationships and employment, while their families mayhave higher risk for conflict, familial stress, and reduced socialparticipation (Stefanatos and Baron, 2007). These factors engenderfurther risk for unfavorable health behaviors and health outcomesamong ADHD-afflicted youth.

Although hyperactivity would seem to increase daily energyexpenditure, youth with ADHD show higher obesity prevalence thanyouth without ADHD (Holtkamp et al., 2004; Lam and Yang, 2007;Waring and Lapane, 2008). Obesity may result from ADHD-relatedphysiologic alterations (Biederman and Faraone, 2005; Faraone et al.,

uffHall, Dept. ofKinesiology andaign, Champaign, IL 61820, USA.

l rights reserved.

2003; Gadde et al., 2006; Waring and Lapane, 2008) such as changes indopamine receptors that could potentially cause binge eating (Corteseand Angriman, 2008) or from increased food intake associated withimpulsive behavior in boys with ADHD (Hubel et al., 2006) andexcessive daytime sleepiness (Cortese and Angriman, 2008). However,there are no epidemiological studies available to identify the high risksubgroups and the determinants of obesity among ADHD youth.

A recent study reported lower prevalence of obesity among ADHDyouth on medication than those not medicated (Waring and Lapane,2008). This is not surprising since stimulantmedications used for ADHDsymptoms cause appetite suppression and weight loss (Biederman andFaraone, 2005; National Institute ofMental Health, 2003; Pliszka, 2007).Nevertheless, only56%ofUSchildrenwithADHDreceivepharmacologictreatment (Biederman and Faraone, 2005). Thus, ADHDmedication usemay not completely account for the observed differences in obesityprevalence between individuals with and without ADHD. Factorscontributing to obesity may differ between boys and girls with ADHD.Girls with ADHD are prescribed medication at half the rate of boys(Faraone et al., 2003; Zuvekas et al., 2006), probably because of lowerhyperactivity levels and other externalizing behaviors (Biederman andFaraone, 2005). Furthermore, girls in the general population are lessactive than boys, especially during adolescence (Andersen et al., 1998;AnderssenandWold, 1992; CardonandDeBourdeaudhuij, 2008). Thesedifferences between boys and girls may impact health behaviors and

219J. Kim et al. / Preventive Medicine 52 (2011) 218–222

may result in higher odds of obesity among youth with ADHD. Inaddition to medication effects, other factors such as low physicalactivity, lack of familial supports, or neighborhood conditions that havebeen implied as contributors to obesityprevalence amongUSyouthmayalso play an important role in ADHD youth (Acevedo-Polakovich et al.,2006; Burdette and Whitaker, 2005).

This study examinedwhether parent reports of their child's physicalactivity, sedentary behaviors, depression, and home and neighborhoodconditions vary as a function of gender and medication use in ADHDchildren, and if children with ADHD were different from childrenwithout ADHD, using a nationally-representative sample of US youth.We also evaluated whether these health behaviors and contextualfactors are associated with obesity within the ADHD subgroups.

Methods

Study sample

The study population included children aged 6–17 years enrolled in theNational Survey of Children's Health (NSCH) 2003. These data were collectedby the US Department of Health and Human Services from all 50 states andthe District of Columbia by employing the State and Local Area IntegratedTelephone Survey sampling program (Blumberg et al., 2005). Trainedinterviewers collected data from parents or guardians (weighted responsewas 55.3%). After excluding children whose height and weight were notbiologically plausible (n=2023) and children with missing information onADHD (n=301), a total of 66,707 children were included. The protocol wasapproved by the Institutional Review Board of the University of Illinois atUrbana-Champaign.

Measures

The definition of ADHD was derived from two questions. “Has a doctor orhealth professional ever told you that your child has ADHD” and “Is your childcurrently taking medication for ADHD?” The exposure variable has threelevels: 1) “ADHD ever; child currently takesmedication” 2) “ADHD ever; childnot taking medication now” and 3) “Never told child has ADHD”. Obesity wasdichotomized as obese or not obese based on the CDC guidelines of weight-for-height ≥95th percentile (Centers for Disease Control and Prevention). Anecological model of obesity was used as a framework for the selection ofobesity-related variables for this study (Davison and Birch, 2001). Childhealth behaviors and contextual factors that place children at increased oddsof obesity such as family and community environment were selected forbivariate analysis to assess whether they were related to ADHD or obesity.

The level of physical activity was dichotomized as meeting or not meetingthe Healthy People 2010 goal— for instance, participating in vigorous activity20 min for 3+days per week or engaging in sedentary behavior such aswatching TV, videos, or playing video games more than two hours per schoolday. Additional variables were included such as: depression or anxietyproblems (yes/no), number of family mealtimes together (b5 days a week),family rules about the TV program (yes/no), pleasure reading (yes/no),computer use (yes/no), rode a bike, scooter, skateboard, roller skates, orrollerblades during the past 12 months (yes/no), participation in organizedsports during the past 12 months (yes/no), participation in clubs ororganizations during the past 12 months (yes/no), not enough sleep(b7 days a week or not), do not live in a supportive neighborhood (yes/no),do not feel safe about neighborhood (some/never and usually/always), anddo not feel safe at school (some/never and usually/always). The ‘supportiveneighborhood’ was a composite variable indicating parent's perceivedneighborhood social capital based on the four questions; “help each otherout”, “watch out each other's children”, “countable people in neighborhood”,and “neighbors help my child”. Socio-demographic variables such as age,family income, race/ethnicity, family structure, maternal education, andhousehold smoking were included in the analysis.

Analyses

All analyses were stratified by gender and weighted to generaterepresentative estimates of the US children by applying sampling weights.Bivariate analyses were performed to compare socio-demographic character-istics and obesity-related variables by ADHD. Among the obesity-related

variables of interest, only variables associated with ADHD remained in thefinal model and are presented in the tables. Regression analyses were appliedfor each health behavior and obesity-related conditions to determine thedifferences by ADHD condition (Table 2) and between medicated and not-medicated ADHD children (Table 3), simultaneously controlling for the socio-demographic variables such as age, race/ethnicity, maternal education, familystructure, household income, and household smoking. We also evaluated thepotential mediation effect of obesity on obesity-related behaviors by ADHDcondition (data not shown) and by the medication status of ADHD children(Table 3) by fitting the model with and without obesity. Furthermore,bivariate and multivariate logistic regression models were built to evaluatewhether each variable is associated with obesity status within each group ofADHD condition and the results were presented after controlling for socio-demographic variables and depression or anxiety problems (Table 4).Regression analyses were adjusted for the multistage sampling and surveydesign effects. Analyses were conducted with SAS 9.2 (SAS Institute, Cary,NC).

Results

The prevalence of ADHD (Table 1)was 8.6% in 2003 and differed bygender (boys, 12.3%; girls, 5%). The use of medication was similar forboys (59.7%) and girls (57.2%) with ADHD. The prevalence of obesitywas highest for both boys and girls with ADHD who were not onmedication when compared to those without the ADHD condition(Table 2). Boys whowere not onmedication for ADHD had 42% higherodds of being obese compared to boys without ADHD, aftercontrolling for socio-demographic variables. Girls, who were not onmedication for ADHD, had 85% higher odds of being obese comparedto those without ADHD (Table 2).

ADHD children on medication had 14 times higher odds of beingdiagnosedwith depression/anxiety compared to thosewithout ADHD.Both boys and girls with ADHD, regardless of medication status,engaged in less physical activity, organized sports, and pleasurereadingwhen compared to thosewithout ADHD. Girls with ADHD andon medication were less likely to participate in club activities. Girlswith ADHD and not taking medication were more likely to spend timeon media [OR(95% CI)=1.60(1.20–2.13)]. There was no difference incomputer use among girls, whereas boys with ADHDwho did not takemedication showed less computer use compared to those withoutADHD. Both boys and girls with ADHD were less likely to live insupportive neighborhoods compared to those without ADHD. Onlyboys with ADHD who were not taking medication did not feel safe atschool. There was no mediating effect of obesity on the obesity-related health behaviors and conditions except the ‘do not feel safeneighborhood’ among girls. After adding obesity in the model, the OR(95% CI) was changed to 1.51(0.99–2.31) for girls with ADHD not onmedication and to 1.38(0.90–2.17) for girls with ADHD onmedication.

The differences between medicated and not-medicated ADHDchildren were estimated (Table 3). Both boys and girls on ADHDmedication were more likely to have depression. The odds ratiosbecame stronger after adding obesity in the model, reflecting apotential additive effect of obesity on depression. Only boys withADHDmedication did not feel safe in the neighborhood, but this effectdisappeared when adding obesity in the model. Girls with medicatedADHD had lower media time compared to their non-medicated peers;obesity status did not change the odds of this behavior.

We evaluated whether health behaviors and other indicators wereassociated with obesity by ADHD condition (Table 4). Among boyswith ADHD who were not taking medication, ‘Not riding a bike’ wasassociated with being obese [OR(95% CI)=2.11(1.22–3.67)]. Partic-ipating in organized sports [OR(95% CI)=1.57(1.06–2.34)] and nothaving enough sleep [OR(95% CI)=0.62(0.41–0.94)] were associatedwith obesity for boys with ADHD and on medication. Watching mediamore than 2 h a day was also associated with increased odds of beingobese among girls with ADHD and not medicated, compared withthose without ADHD.

Table 1Socio-demographic characteristics by attention deficit hyperactivity disorder (ADHD) condition among children aged 6 to 17 years from the National Survey of Children's Health2003.

Boys Girls

No ADHD ADHD, not medicated ADHD, medicated No ADHD ADHD, not medicated ADHD, medicated

n (%) 29,801 (87.7) 1757 (5.3) 2599 (7.0) 30,772 (95.0) 733 (2.2) 981 (2.8)Race/ethnicity

Hispanic 3591 (16.8) 149 (8.9) 164 (5.8) 3505 (16.2) 61 (10.8) 59 (7.5)African American 2775 (13.8) 188 (17.8) 211 (13.3) 3018 (15.1) 68 (15.5) 72 (10.0)Other 2085 (7.2) 135 (7.0) 160 (4.8) 2229 (7.0) 64 (5.8) 63 (6.5)White 20,963 (62.0) 1270 (66.2) 2033 (76.2) 21,565 (61.6) 529 (67.9) 775 (76.1)

Family structureTwo parent biological/adopted 18,431 (60.6) 782 (42.6) 1242 (45.7) 18,701 (59.5) 320 (40.6) 494 (50.4)Two parent stepfamily 3053 (10.9) 281 (18.2) 417 (17.0) 3171 (11.2) 119 (21.5) 137 (17.1)Single mother 5956 (22.7) 483 (32.3) 674 (31.8) 6526 (24.9) 217 (31.4) 251 (28.1)other 1689 (5.8) 138 (6.9) 162 (5.5) 1520 (4.5) 42 (6.5) 57 (4.4)

More than high school graduates 22,468 (32.7) 1216 (40.4) 1907 (33.4) 23,017 (34.3) 545 (30.0) 748 (32.8)Household income

0–99% FPL 2759 (15.8) 259 (24.9) 307 (18.4) 2906 (15.9) 95 (17.3) 104 (14.5)100–199% FPL 5125 (21.6) 345 (21.5) 494 (22.7) 5475 (22.9) 140 (22.7) 192 (25.4)200–399% FPL 10,225 (34.6) 571 (30.5) 861 (32.9) 10,279 (33.7) 263 (37.7) 307 (29.7)400+% FPL 8992 (28.0) 427 (23.2) 771 (26.0) 9303 (27.6) 168 (22.3) 306 (30.4)

Household smoking, yes 8803 (28.7) 786 (42.8) 1,025 (43.5) 9282 (30.0) 321 (41.4) 377 (38.1)

All percentages were weighted to represent children aged 6 to 17 years nationally.Weighted percentages represent the column percentage except the prevalence of ADHD condition.

220 J. Kim et al. / Preventive Medicine 52 (2011) 218–222

Discussion

We evaluated whether the prevalence of health behaviors andobesity differed by gender and ADHD medication status from arepresentative sample of US youth and identified the potentialdeterminants for obesity in this study population. This study providesepidemiological evidence that children with ADHD, regardless ofmedication status and gender, are less likely to participate in vigorousphysical activity and organized sports compared to those withoutADHD. Previous studies on movement skills and motor performance

Table 2The prevalence and adjusted odds ratios for obesity, depression, and health behaviors by attenfrom the National Survey of Children's Health 2003.

% OR(95% CI) Boys

No ADHD ADHD, not medicated

Obesity 21.6 24.9Ref. 1.42 (1.13–1.77)

Depression/anxiety problems 2.8 26.3Ref. 10.3 (7.9–13.5)

Low physical activity (b3 days/week) 79.3 26.6Ref. 1.45 (1.23–1.70)

Media time (≥2 h a day) 46.4 52.3Ref. 1.09 (0.90–1.31)

Family rules about TV programs 15.1 18.1Ref. 0.92 (0.73–1.17)

No pleasure reading 19.4 34.0Ref. 1.65 (1.35–2.01)

No computer use 28.7 32.2Ref. 1.32 (1.06–1.65)

Not rode a bike, scooter, skateboard, etc. 15.2 16.7Ref. 0.88 (0.67–1.04)

Not participated in organized sports 36.6 49.0Ref. 1.50 (1.23–1.82)

Not participated in any clubs or organizations 51.0 54.3Ref. 1.09 (0.90–1.33)

Not enough sleep (b7 days a week) 30.1 38.1Ref. 1.22 (1.00–1.47)

Do not live in a supportive neighborhood 16.2 23.8Ref. 1.51 (1.18–1.91)

Do not feel safe neighborhood 12.2 14.7Ref. 1.54 (1.18–2.01)

Do not feel safe at school 9.9 15.1Ref. 1.44 (1.10–1.88)

All percentages were weighted to represent children aged 6 to 17 years nationally. The cethnicity, maternal education, family structure, household income, and household smoking

in ADHD children provide some insight into the underlying factors forthe observed low physical activity in this population. For example,children with ADHD exhibit low gross motor performance, physicalfitness, and delayed motor development (Harvey and Reid, 1997,2003). A recent qualitative study found that children with ADHD hadonly superficial knowledge about movement skills, paid littleattention to specific details and entertained negative feelings aboutphysical activity (Harvey et al., 2009).

We confirmed that obesity prevalence was higher among childrenwith ADHD (Holtkamp et al., 2004; Lam and Yang, 2007) and

tion deficit hyperactivity disorder (ADHD) condition among children aged 6 to 17 years

Girls

ADHD, medicated No ADHD ADHD, not medicated ADHD, medicated

20.5 16.0 21.9 18.60.89 (0.72–1.09) Ref. 1.85 (1.26–2.73) 1.21 (0.84–1.73)29.5 3.4 23.6 30.914.1 (11.4–17.6) Ref. 6.4 (4.6–9.1) 14.1 (10.9–18.2)25.8 68.9 42.7 35.61.36 (1.10–1.68) Ref. 1.71 (1.28–2.27) 1.29 (1.01–1.64)52.1 42.0 52.7 39.41.14 (0.98–1.32) Ref. 1.60 (1.20–2.13) 0.90 (0.70–1.15)10.0 85.4 82.9 86.10.65 (0.51–0.81) Ref. 1.01 (0.70–1.45) 1.03 (0.72–1.46)28.1 11.5 22.3 17.31.54 (1.30–1.81) Ref. 1.93 (1.37–2.70) 1.75 (1.32–2.32)30.2 27.6 27.6 31.41.15 (0.97–1.37) Ref. 1.20 (0.86–1.68) 1.23 (0.94–1.63)11.7 22.6 24.4 17.50.83 (0.99–1.28) Ref. 0.83 (0.61–1.14) 0.87 (0.65–1.17)48.8 45.4 55.7 54.61.67 (1.44–1.94) Ref. 1.39 (1.02–1.90) 1.63 (1.27–2.09)51.3 42.3 47.7 48.91.02 (0.88–1.19) Ref. 1.25 (0.93–1.69) 1.43 (1.10–1.83)33.6 31.2 36.7 35.41.15 (0.99–1.34) Ref. 1.17 (0.87–1.59) 1.19 (0.90–1.58)20.1 16.1 23.9 21.61.29 (1.04–1.60) Ref. 1.66 (1.13–2.43) 1.53 (1.13–2.08)11.2 13.5 16.2 14.11.14 (0.89–1.46) Ref. 1.62 (1.08–2.44) 1.45 (0.94–2.22)10.1 10.1 13.5 9.31.22 (0.87–1.72) Ref. 1.44 (0.98–2.11) 1.40 (0.86–2.27)

ovariates in the multivariate analyses were socio-demographic variables of age, race/. Bold type represents statistical significance at the 5 percent level.

Table 3The adjusted odds ratios for depression and health behaviors between the medicated and not medicated attention deficit hyperactivity disorder (ADHD) among children aged 6 to17 years from the National Survey of Children's Health 2003.

Boys Girls

Unit: OR(95% CI) ADHD, medicated (n=2599) vs. notmedicated (n=1757), withoutobesity

ADHD, medicated (n=2505) vs.not medicated (n=1696), withobesity

ADHD, medicated (n=981) vs. notmedicated (n=733), withoutobesity

ADHD, medicated (n=944) vs.not medicated (n=712), withobesity

Depression/anxietyproblems

1.36 (1.02–1.80) 1.45 (1.09–1.94) 2.04 (1.33–3.15) 2.27 (1.48–3.49)

Low physical activity(b3 days/week)

1.05 (0.81–1.35) 1.04 (0.81–1.35) 0.89 (0.62–1.29) 0.95 (0.65–1.39)

Media time (≥2 h a day) 1.03 (0.82–1.30) 1.01 (0.80–1.28) 0.64 (0.45–0.92) 0.66 (0.46–0.95)Family rules about TVprograms

0.65 (0.48–0.90) 0.66 (0.48–0.91) 0.97 (0.60–1.59) 0.94 (0.58–1.54)

No pleasure reading 0.90 (0.71–1.15) 0.90 (0.70–1.16) 0.87 (0.57–1.32) 0.85 (0.55–1.30)No computer use 0.87 (0.67–1.13) 0.89 (0.68–1.16) 1.09 (0.72–1.65) 1.03 (0.68–1.57)Not rode a bike, scooter,skateboard, etc.

0.85 (0.62–1.18) 0.90 (0.65–1.26) 1.02 (0.67–1.57) 1.08 (0.71–1.66)

Not participated inorganized sports

1.15 (0.91–1.47) 1.20 (0.94–1.53) 1.24 (0.84–1.84) 1.23 (0.82–1.83)

Not participated in any clubsor organizations

0.96 (0.76–1.21) 0.96 (0.76–1.21) 1.20 (9.82–1.76) 1.20 (0.81–1.77)

Not enough sleep(b7 days a week)

0.91 (0.72–1.15) 0.91 (0.72–1.16) 0.87 (0.60–1.28) 0.84 (0.57–1.22)

Do not live in a supportiveneighborhood

0.82 (0.60–1.12) 0.83 (0.60–1.14) 0.99 (0.62–1.57) 1.03 (0.64–1.66)

Do not feel safeneighborhood

0.69 (0.48–0.98) 0.71 (0.50–1.02) 0.90 (0.53–1.53) 0.95 (90.55–1.63)

Do not feel safe at school 0.80 (0.55–1.18) 0.80 (0.52–1.23) 0.99 (0.58–1.71) 0.97 (0.54–1.74)

The covariates in themultivariate analyseswere socio-demographic variables of age, race/ethnicity,maternal education, family structure, household income, andhousehold smoking. Boldtype represents statistical significance at the 5 percent level.

221J. Kim et al. / Preventive Medicine 52 (2011) 218–222

medication use was a protective factor for obesity among children withADHD (Waring and Lapane, 2008). We also found the odds of beingobese were higher among girls than boys with non-medicated ADHDcompared to those without ADHD. In addition, only health behaviorssuch as not participating in organized sports and lack of sleep wereassociated with obesity in boys with ADHD on medication, eventhough all children with ADHD were less likely to be physicallyactive. Interestingly, our study found that lack of sleep protectsagainst obesity in boys with ADHD on medication, which is

Table 4The adjusted odds ratios of depression and health behaviors for obesity by attention deficitNational Survey of Children's Health 2003.

Boys

Unit: OR(95% CI) No ADHD ADHD, not medicated AD

Depression/anxiety(yes vs. no)

1.49 (1.06–2.10) 1.44 (0.88–2.38) 1.1

Low physical activity(b3 days/week) vs. not

1.30 (1.13–1.50) 1.33 (0.83–2.12) 0.9

Media time (≥2 h a day) vs. not 1.13 (1.01–1.28) 1.24 (0.82–1.87) 1.0Family rules about TVprograms (yes vs. no)

1.14 (0.94–1.37) 1.37 (0.80–2.34) 0.9

Pleasure reading (no vs. yes) 1.21 (1.04–1.41) 1.10 (0.68–1.78) 0.8Computer use (no vs. yes) 0.99 (0.87–1.13) 1.53 (0.95–2.48) 0.7Rode a bike, scooter, skateboard,etc.(no vs. yes)

1.27 (1.06–1.52) 2.11 (1.22–3.67) 1.5

Participated in organized sports(no vs. yes)

1.29 (1.14–1.47) 1.23 (0.81–1.85) 1.5

Participated in any clubs ororganizations (no vs. yes)

1.03 (0.92–1.15) 1.35 (0.88–2.07) 1.1

Not enough sleep(b7 vs. 7 days a week)

1.02 (0.90–1.15) 0.77 (0.51–1.17) 0.6

Live in a supportive neighborhood(no vs. yes)

0.90 (0.77–1.06) 1.34 (0.78–2.29) 0.8

Feel safe neighborhood(no vs. yes)

1.04 (0.86–1.26) 1.21 (0.69–2.10) 0.6

Feel safe at school(no vs. yes)

0.91 (0.74–1.12) 0.87 (0.50–1.51) 0.7

All percentages were weighted to represent children aged 6 to 17 years nationally. The covmaternal education, family structure, household poverty, household smoking and depression

contradictory to the previous finding that lack of sleep is a risk factorfor obesity in the general population (Cappuccio et al., 2008; Chaputand Tremblay, 2009). Whether insomnia, a side effect of stimulantmedication for ADHD (Lerner and Wigal, 2008), changes the odds ofobesity among medicated ADHD youth warrants further research.While medication does not appear to alter movement skills in childrenwith ADHD (Harvey et al., 2007), it is unknown if medicationmitigates other unhealthy behaviors leading to obesity in our studypopulation.

hyperactivity disorder (ADHD) condition among children aged 6 to 17 years from the

Girls

HD, medicated No ADHD ADHD, not medicated ADHD, medicated

4 (0.73–1.78) 1.59 (1.13–2.26) 2.02 (0.89–4.62) 1.69 (0.88–3.23)

9 (0.67–1.46) 1.46 (1.27–1.69) 1.18 (0.60–2.32) 1.24 (0.66–2.31)

8 (0.73–1.58) 1.31 (1.14–1.50) 2.51 (1.24–5.08) 1.75 (0.92–3.30)4 (0.52–1.68) 1.16 (0.93–1.45) 0.77 (0.24–2.53) 0.76 (0.29–1.99)

3 (0.53–1.31) 0.84 (0.66–1.07) 0.91 (0.42–1.97) 1.35 (0.67–2.74)5 (0.48–1.17) 0.95 (0.82–1.11) 1.06 (0.44–2.59) 1.22 (0.63–2.36)4 (0.86–2.73) 1.43 (1.18–1.73) 1.10 (0.47–2.56) 1.56 (0.71–3.42)

7 (1.06–2.34) 1.35 (1.18–1.54) 2.01 (0.99–4.06) 1.90 (0.88–4.10)

7 (0.80–1.70) 1.05 (0.91–1.21) 1.57 (0.75–3.29) 0.83 (0.43–1.59)

2 (0.41–0.94) 0.90 (0.78–1.04) 1.42 (0.75–2.68) 1.69 (0.81–3.52)

9 (0.52–1.54) 1.19 (0.98–1.45) 1.35 (0.62–2.92) 1.63 (0.78–3.38)

4 (0.35–1.15) 1.23 (1.02–1.49) 1.33 (0.57–3.14) 1.97 (0.71–5.49)

4 (0.38–1.44) 1.16 (0.92–1.46) 2.00 (0.72–5.57) 2.51 (0.90–7.05)

ariates in the multivariate analyses were the socio-demographic variables of age, race,or anxiety problems. Bold type represents statistical significance at the 5 percent level.

222 J. Kim et al. / Preventive Medicine 52 (2011) 218–222

A novel finding in our study was that media time was associatedwith being obese among non-medicated girls with ADHD whencompared to those without ADHD; such association was not observedfor boys with ADHD. More than half of non-medicated girls withADHD spent more than 2 h/day on media and they were estimated tohave 60% more media time compared to girls without ADHD. Inaddition, girls with medicated ADHD spent less media time than girlswith non-medicated ADHD. Although media time may not impactattention and cognitive engagement abilities of elementary schoolchildren with ADHD (Acevedo-Polakovich et al., 2006), our studyfound that medicated ADHD girls had lower media time than non-medicated girls. Media use, especially television watching, indeedinfluences physical activity, body fat and weight among healthychildren (Andersen et al., 1998). Additionally, parental perception ofneighborhood safety influences media time in preschool children(Burdette and Whitaker, 2005). We found that non-medicated girlswith ADHD were more likely to live in less safe neighborhoods thangirls without ADHD, whereas this was not observed among girlsmedicated for ADHD. It is possible that parents of non-medicatedADHD girls may prefer their children to use media rather than engagein physical activity, a preference potentially contributing to obesity.

Depression was higher among youth with ADHD and their oddswere even higher with medication use, which is consistent withearlier studies (National Institute of Mental Health, 2003; Waring andLapane, 2008; Waxmonsky, 2003). In our study, the associationsbetween health behaviors and obesity among children with ADHDcondition remained strong even after controlling for depression.

Given the cross-sectional nature of the data, we cannot establishthe direction of the relationship. We cannot differentiate whetherchildren with ADHD, who are not obese, are more likely to be treatedwith medication or whether those on medication are more likely toreceive weight management program such as physical activity.Importantly, the NSCH study did not collect information on specificmedications used for ADHD; thus the effects of specific medicationson health behaviors and obesity could not be estimated. Furthermore,themain study outcomes, ADHD and obesity, were not based on DSM-IV criteria. In addition, our binary classification for continuousvariables provides an easy interpretation for program implementationand evaluation, but may not fully explain the nature of behaviors orperceptions on the study questions. For this reason, there is a potentialfor misclassification which could lead to either overestimation orunderestimation of the observed relationships.

Conclusions

ADHDchildrenare less likely to engage inphysical activity regardlessof gender and medication use compared to children without ADHD,whereas their odds of being obese are dependent on gender andmedication status. Behavioral modification and obesity interventionprograms should take these factors into consideration.

Conflict of interest statementThere are no financial or material support interests to disclose.

References

Acevedo-Polakovich, I.D., Lorch, E.P., Milich, R., Ashby, R.D., 2006. Disentangling therelation between television viewing and cognitive processes in children withattention-deficit/hyperactivity disorder and comparison children. Arch. Pediatr.Adolesc. Med. 160, 354–360.

Andersen, R.E., Crespo, C.J., Bartlett, S.J., Cheskin, L.J., Pratt, M., 1998. Relationship ofphysical activity and television watching with body weight and level of fatnessamong children: results from the Third National Health and Nutrition ExaminationSurvey. JAMA 279, 938–942.

Anderssen, N., Wold, B., 1992. Parental and peer influences on leisure-time physicalactivity in young adolescents. Res. Q. Exerc. Sport 63, 341–348.

Biederman, J., Faraone, S.V., 2005. Attention-deficit hyperactivity disorder. Lancet 366,237–248.

Blumberg, S.J., Olson, L., Frankel, M.R., Osborn, L., Srinath, K.P., Giambo, P., 2005. Designand Operation of the National Survey of Children's Health, 2003: Vital Health Stat.,43.

Burdette, H.L., Whitaker, R.C., 2005. A national study of neighborhood safety, outdoorplay, television viewing, and obesity in preschool children. Pediatrics 116,657–662.

Cappuccio, F.P., Taggart, F.M., Kandala, N.B., et al., 2008. Meta-analysis of short sleepduration and obesity in children and adults. Sleep 31, 619–626.

Cardon, G.M., De Bourdeaudhuij, I.M., 2008. Are preschool children active enough?Objectively measured physical activity levels. Res. Q. Exerc. Sport 79, 326–332.

Centers for Disease Control and Prevention, n. A SAS Program for the CDCGrowthChartsRetrieved Jan. 10, 2007, from http://www.cdc.gov/nccdphp/dnpa/growthcharts/sas.htm.

Chaput, J.P., Tremblay, A., 2009. Obesity and physical inactivity: the relevance ofreconsidering the notion of sedentariness. Obes. Facts 2, 249–254.

Cortese, S., Angriman, M., 2008. Attention-deficit/hyperactivity disorder and obesity:moving to the next research generation. Pediatrics 122, 1155 (author reply1155-6).

Davison, K.K., Birch, L.L., 2001. Childhood overweight: a contextual model andrecommendations for future research. Obes. Rev. 2, 159–171.

Faraone, S.V., Sergeant, J., Gillberg, C., Biederman, J., 2003. The worldwide prevalence ofADHD: is it an American condition? World Psychiatry 2, 104–113.

Gadde, K.M., Yonish, G.M., Wagner II, H.R., Foust, M.S., Allison, D.B., 2006. Atomoxetinefor weight reduction in obese women: a preliminary randomised controlled trial.Int. J. Obes. Lond 30, 1138–1142.

Harvey, W.J., Reid, G., 1997. Motor performance of children with attention-deficithyperactivity disorder: a preliminary investigation. Adapt. Phys. Act. Q. 14,189–202.

Harvey, W.J., Reid, G., 2003. Attention-deficit/hyperactivity disorder: a review ofresearch on movement skill performance and physical fitness. Adapt. Phys. Act. Q.20, 1–25.

Harvey, W.J., Reid, G., Grizenko, N., Mbekou, V., Ter-Stepanian, M., Joober, R., 2007.Fundamental movement skills and children with attention-deficit hyperactivitydisorder: peer comparisons and stimulant effects. J. Abnorm. Child Psychol. 35,871–882.

Harvey, W.J., Reid, G., Bloom, G.A., et al., 2009. Physical activity experiences of boys withand without ADHD. Adapt. Phys. Activ. Q. 26, 131–150.

Holtkamp, K., Konrad, K., Muller, B., et al., 2004. Overweight and obesity in childrenwith attention-deficit/hyperactivity disorder. Int. J. Obes. Relat. Metab. Disord. 28,685–689.

Hubel, R., Jass, J., Marcus, A., Laessle, R.G., 2006. Overweight and basal metabolic rate inboys with attention-deficit/hyperactivity disorder. Eat. Weight Disord. 11,139–146.

Lam, L.T., Yang, L., 2007. Overweight/obesity and attention deficit and hyperac-tivity disorder tendency among adolescents in China. Int. J. Obes. Lond 31,584–590.

Lerner, M., Wigal, T., 2008. Long-term safety of stimulant medications used to treatchildren with ADHD. Pediatr. Ann. 37, 37–45.

National Institute of Mental Health, 2003. Attention Deficit Hyperactivity Disorder. USDepartment of Health and Human Services, Bethesda, MD. Retrieved March 25,2008.

Pliszka, S.R., 2007. Pharmacologic treatment of attention-deficit/hyperactivity disorder:efficacy, safety and mechanisms of action. Neuropsychol. Rev. 17, 61–72.

Stefanatos, G.A., Baron, I.S., 2007. Attention-deficit/hyperactivity disorder: a neuropsy-chological perspective towards DSM-V. Neuropsychol. Rev. 17, 5–38.

Visser, S.N., Lesesne, C.A., Perou, R., 2007. National estimates and factors associated withmedication treatment for childhood attention-deficit/hyperactivity disorder.Pediatrics 119 (Suppl. 1), S99–S106.

Waring, M.E., Lapane, K.L., 2008. Overweight in children and adolescents in relation toattention-deficit/hyperactivity disorder: results from a national sample. Pediatrics122, e1–e6.

Watemberg, N., Waiserberg, N., Zuk, L., Lerman-Sagie, T., 2007. Developmentalcoordination disorder in children with attention-deficit-hyperactivity disorderand physical therapy intervention. Dev. Med. Child Neurol. 49, 920–925.

Waxmonsky, J., 2003. Assessment and treatment of attention deficit hyperactivitydisorder in children with comorbid psychiatric illness. Curr. Opin. Pediatr. 15,476–482.

Zuvekas, S.H., Vitiello, B., Norquist, G.S., 2006. Recent trends in stimulant medicationuse among US children. Am. J. Psychiatry 163, 579–585.