Physical activity and sedentary time in relation to academicachievement in children

(1)

UEF//eRepository

DSpace https://erepo.uef.fi

Rinnakkaistallenteet Terveystieteiden tiedekunta

2017

Physical activity and sedentary time in

relation to academicachievement in children

Haapala EA

info:eu-repo/semantics/article

info:eu-repo/semantics/publishedVersion

© Sports Medicine Australia

CC BY http://creativecommons.org/licenses/by/4.0/

http://dx.doi.org/10.1016/j.jsams.2016.11.003

https://erepo.uef.fi/handle/123456789/4274

Downloaded from University of Eastern Finland's eRepository

(2)

JournalofScienceandMedicineinSport20(2017)583–589

ContentslistsavailableatScienceDirect

Journal of Science and Medicine in Sport

j o u r n a l ho me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / j s a m s

Original research

Physical activity and sedentary time in relation to academic achievement in children

Eero A. Haapala

^a,b,∗

, Juuso Väistö

^a,c

, Niina Lintu

^a

, Kate Westgate

^d

, Ulf Ekelund

^d,e

, Anna-Maija Poikkeus

^f

, Soren Brage

^d

, Timo A. Lakka

^a,g,h

aInstituteofBiomedicine/Physiology,UniversityofEasternFinland,KuopioCampus,Finland

bChildhoodHealthandActiveLivingResearchGroup,DepartmentofBiologyofPhysicalActivity,UniversityofJyväskylä,Finland

cInstituteofDentistry,UniversityofEasternFinland,KuopioCampus,Finland

dMRCEpidemiologyUnit,UniversityofCambridge,UniversityofCambridgeSchoolofClinicalMedicine,InstituteofMetabolicScience,Cambridge BiomedicalCampus,UnitedKingdom

eDepartmentofSportMedicine,NorwegianSchoolofSportsScience,Norway

fDepartmentofTeacherEducation,UniversityofJyväskylä,Finland

gDepartmentofClinicalPhysiologyandNuclearMedicine,UniversityofEasternFinland,KuopioCampus,Finland

hKuopioResearchInstituteofExerciseMedicine,Finland

a r t i c l e i n f o

Articlehistory:

Received28March2016

Receivedinrevisedform21October2016 Accepted15November2016

Availableonline23November2016

Keywords:

Motoractivity Sedentarybehavior Academicperformance Cognition

Pediatrics

a b s t r a c t

Objectives:Toinvestigatetheindependentandcombinedassociationsofobjectivelymeasuredmoderate- to-vigorousphysicalactivity(MVPA)andsedentarytime(ST)withreadingandarithmeticskills.

Design:Cross-sectional/prospective.

Methods:Participantswere89boysand69girlsaged6–8years.MVPAandSTweremeasuredusingacom- binedheartrateandmovementsensorandbodyfatpercentagebydual-energyX-rayabsorptiometryin Grade1.Readingﬂuency,readingcomprehension,andarithmeticskillswereassessedusingstandardized testsinGrades1–3.Thedatawereanalyzedusinglinearregressionanalysesandanalysesofcovariance withrepeatedmeasures.

Results:Inboys,MVPAwasdirectlyandSTinverselyassociatedwithreadingfluencyinGrades1–3 andarithmeticskillsinGrade1(P<0.05).HigherlevelsofMVPAwerealsorelatedtobetterreading comprehensioninGrade1(P<0.05).MostoftheassociationsofMVPAandSTwithreadingandarithmetic skillsattenuatedaftermutualadjustmentforMVPAorST.Furthermore,boyswithacombinationoflower levelsofMVPAandhigherlevelsofSThadconsistentlypoorerreadingfluency(P=0.002)andreading comprehension(P=0.027)acrossGrades1–3thanotherboys.Ingirls,STwasdirectlyassociatedwith arithmeticskillsinGrade2(P<0.05).However,thisrelationshipofSTwitharithmeticskillswasnolonger significantafteradjustmentforbodyfatpercentage.

Conclusions:LowerlevelsofMVPAandhigherlevelsofSTandparticularlytheircombinationwererelated topoorerreadingskillsinboys.Ingirls,higherlevelsofSTwererelatedtobetterarithmeticskills.

1. Introduction

Physicalinactivityisamajorpublichealthproblemindeveloped countries.¹ Thereissomeevidencethathigherlevelsofphysical activity(PA)andlowerlevelsofsedentarytime(ST)areassociated withbetterbrainstructuresandfunctionsinchildren.^2,3Moreover, lowlevelsofPAandhighlevelsofST,particularlyTVwatching, havebeenlinkedtopooreracademicachievementinchildren.^4,5

∗Correspondingauthor.

E-mailaddress:eero.haapala@uef.ﬁ(E.A.Haapala).

However,theserelationshipsaremainlybasedondatafromcross- sectionalstudiesusingself-reportedPAandST.^4,6 Theabilityof self-reporttorankthelevelofPAcorrectlyhasbeenquestioned,and self-reportsarealsopronetorecallbias.⁷Therefore,evidenceon theassociationsofobjectivelymeasuredPAandSTwithacademic achievementinchildrenishighlywarranted.

Higher levels of objectively measured moderate-to-vigorous PA (MVPA) at the age of 11 were associated withbetter academic achievement at the ages of 11, 13, and 16 years⁸ but withpoorer academicachievementin children andadolescents aged 6–18 years.⁹ Some studies have found no relationship betweenobjectivelymeasuredMVPAandacademicachievementin http://dx.doi.org/10.1016/j.jsams.2016.11.003

(3)

584 E.A.Haapalaetal./JournalofScienceandMedicineinSport20(2017)583–589

children 10 years of age.^5,10 In recent studies, higher levels of PAhavebeenassociatedwithbetteracademicachievementonly amongboys.^11,12Incontrast,somestudieshaveobservedanassoci- ationbetweenhigherlevelsofPAandbettercognitiveperformance ingirlsbutnotinboys.^13,14Therearestudiesontheassociations ofMVPAwithacademicachievementduringchildhood,although MVPAaccounts only fora smallproportion of daily PA among children.¹⁵However,therearenostudiesontherelationshipsof objectivelymeasuredlightPAtoacademicachievementalthough childrenspentmostoftheirphysicallyactivetimeinlightPA.¹⁵

Fewavailablestudiesontheassociationsofobjectivelymea- sured ST with academic achievement suggest a weak positive or no relationship between ST and academic achievement in children⁵ andadolescents.^16,17Theseresultsareconsistentwith theobservationsindicatingthatsomesedentarybehaviors,such asTVwatching,areinverselyassociatedwhereassomesedentary behaviors,suchasreading,aredirectlyassociatedwithacademic achievementinchildrenthatmakestheinterpretationoftheasso- ciationsbetweentotalSTandacademicachievementdifﬁcult.⁴

Theevidenceontheindependentandcombinedrelationships ofobjectivelymeasuredPAandSTtoacademicachievementislim- ited.Furthermore,thereareonlyafewstudiesonthelongitudinal associationsofPAandSTwithacademicachievementduringthe ﬁrstschoolyears.Therefore,weinvestigatedtheindependentand combinedassociationsofobjectivelymeasuredMVPA,lightPA,and STinGrade1withreadingandarithmeticskillsinGrades1–3in Finnishboysandgirlsaged6–8years.

2. Methods

2.1. Studydesignandstudypopulation

DataforthepresentanalyseswereobtainedfromthePhysical ActivityandNutritioninChildren(PANIC)StudyandtheFirstSteps Study,twoindependentstudiesthatarebeingconductedsimul- taneouslyamong primaryschoolchildren intheCityofKuopio, Finland.¹⁸ Altogether207children fromtheCityofKuopiopar- ticipatedinboththePANICStudyandtheFirstStepsStudy.Data onPA,ST,andconfoundingfactorswerederivedfromthePANIC StudyinGrade1anddataonreadingandarithmeticskillsatthe endofGrades1–3werereceivedfromtheFirstStepsStudy.Com- pletedataonvariablesusedinthepresentanalyseswereavailable for153children(89boys,64girls)inGrade1,149children(87 boys,62girls) in Grade 2,and 145children (86 boys,59 girls) inGrade3.Childrenwhowereexcludedfromthepresentanaly- sesbecauseofincompletedatahadhigherlevelsofMVPA,lower levelsofST, and highermaximal workloadper leanbody mass inmaximalcycleergometerexercisetestthanthechildrenwho wereincluded(P<0.05).Therewerenodifferencesinotherchar- acteristicsbetweenchildreninthestudysampleandtheexcluded children.ThePANICStudyprotocolwasapprovedbytheResearch EthicsCommitteeoftheHospitalDistrictofNorthernSavo,Kuopio, andtheFirstStepsStudyprotocolwasapprovedbytheResearch EthicsCommitteeoftheUniversityofJyväskylä.Allparticipating childrenandtheirparentsprovidedwritteninformedconsent.The fundingsourceshadnoroleinthecollection,analysis,orinterpreta- tionofthedataorintheapprovalordisapprovalofthepublication.

2.2. Assessmentofacademicachievement

Readingﬂuencywasassessedusingagroup-administeredsub- testofthenationallynormedreadingachievementtestbatteryfor primaryschoolscalledAla-asteenlukutesti(ALLU)inFinnish.¹⁹The testscorewasthenumberofcorrectanswers,rangingfrom0to80, duringa2-mintimelimitforitemsthatinvolvedidentifyingthe

correctwordfromfourphonologicallysimilaralternativeslinked toanadjoiningpicture.

Reading comprehension was assessed with a group- administeredsubtestfromtheALLUtestbattery.¹⁹Afterreading ashorttext,childrenwereaskedtoanswerto12multiple-choice questionsrelatingtofacts,causalrelationships,interpretations,or conclusionsdrawnfromthetext.Thetestscorewasthenumber ofcorrectanswers,rangingfrom0to12,duringthe30-mintest periodwhenchildrenwereallowedtorefertotheoriginaltext.

Arithmeticskillswereassessedusingabasicarithmetictestwith asetofvisuallypresentedadditionandsubtractiontasks.²⁰Chil- drenwereaskedtoperformasmanycalculationsastheycould duringthe3-min timelimit. Thetest scorewasthenumber of correctanswers,rangingfrom0to28.

2.3. Assessmentofphysicalactivityandsedentarytime

PAwasobjectivelyassessedusingacombinedheartrateand movement sensor (Actiheart^®, CamNtech Ltd.,Papworth, UK)²¹ whichwasattachedtothechildren’schestwithtwostandardECG electrodes.Thechildrenwereaskedtowearthesensorcontinu- ouslyforaminimumoffourdays(includingsleepandwater-based activities)withoutchangingtheirusualbehavior.Theheartrate datawereindividuallycalibratedwithdatafromamaximalcycle ergometerexercisetest.WedeﬁnedMVPAasactivitiesexceeding theintensityof4metabolicequivalents(METs),lightPAasactivities thatwereperformedattheintensityof1.5–4metabolicequivalents (METs),andSTasactivitiesthatwereperformedbelowtheinten- sityof1.5METs.Weincludeddataonchildrenwhohadatleast48h (32hduringweekdays,16hduringweekenddays,representedby

≥12hofmorning,noon,afternoon,andeveningweartime)ofvalid activityrecordingintheanalyses.²²

2.4. Otherassessments

Body height and body weight were measured by standard procedures.²³Pubertalstatuswasassessedusingthefivestagecri- teriadescribedbyTanner.²⁴Bodyfatmass,bodyfatpercentage,and leanbodymassweremeasuredusingtheLunarProdigyAdvance^® DXAdevice.²⁵Cardiovascularfitness(maximalworkloadperlean bodymass)wasassessedbyamaximalcycleergometertestand motor performance by 50-m shuttle run test time.²⁵ Parental education,householdincome, PAat differentsettingsincluding unsupervisedPA,organizedsports,othersupervisedexercise,PA atrecess,andphysicallyactiveschooltransportation,habitualtime spentwatchingTV,usingcomputer,andreadingwereassessedby aquestionnaire.²⁶Theriskofreadingdisabilitywasassessedinthe FirstStepsStudyasdescribedearlier.²⁷Children’swellbeingwas assessedbyaquestionnairethatwasfilledoutbytheparentsand included37questionsonthefrequencyofthecomponentsofphys- ical,psychological,andsocialwellbeingscoredbetween1(never) and5(everydayoralmosteveryday).Overallwellbeingscorewas computedasasumofallthesemeasuresthatrangedbetween37 and185,lowerscoreindicatingbetteroverallwellbeing.

2.5. Statisticalmethods

WeperformedalldataanalysesusingSPSSStatistics,Version 21.0(IBMCorp.,Armonk,NY,USA).Basiccharacteristicsbetween boys and girls were compared using the Student’s t-test, the Mann–WhitneyU-test,orthechisquare-test.Theassociationsof MVPA,lightPA,andSTinGrade1withreadingﬂuency,reading comprehension,and arithmeticskills in Grades1–3werestud- iedusinglinearregressionanalyses.MVPA,LightPA,andSTwere enteredintothemodelonebyoneanddatawereadjustedforage, sex,andsensorweartime.Thedataweremutuallyadjustedfor

(4)

E.A.Haapalaetal./JournalofScienceandMedicineinSport20(2017)583–589 585

Table1

Basiccharacteristics.

All Boys Girls P

Backgroundcharacteristics

Age(years) 7.7(0.4) 7.7(0.4) 7.6(0.3) 0.021

Prepubertal(%) 96.8 96.7 97.0 0.935

Bodyheight(cm) 130(5.5) 130.0(5.8) 128.5(5.0) 0.052

Bodyweight(kg)^† 25.8(6.1) 26.4(5.8) 25.5(6.0) 0.250

Bodyfatpercentage^† 18.1(11.2) 14.8(11.6) 19.9(10.4) <0.001

Maximalworkload/leanbodymass(W/kg) 3.6(0.5) 3.7(0.5) 3.4(0.4) <0.001

50-mshuttleruntesttime(s) 24.1(2.2) 23.8(2.1) 24.7(2.1) 0.012

Parentaleducation(%)

Vocationalschoolorless 20.0 23.9 14.7 0.105

Polytechic 38.1 31.5 47.1

Universitydegree 41.9 44.6 38.2

Householdincome(%)

≤30,000 20.3 21.1 19.1 0.616

>30,000–60,000 45.6 42.2 50.0

>60,000 34.2 36.7 30.9

Riskofreadingdisabilities 13.8 18.5 7.4 0.043

Physicalactivityandsedentarytime

Sensorweartimeduringweekdays(hours)^† 98.6(49.4) 92.6(45.1) 113.6(52.5) 0.223

Sensorweartimeduringweekend(hours)^† 40.8(24.1) 38.7(25.4) 45.0(19.8) 0.170

Sedentarytime(min/d)^† 226(192) 226(199) 223(202) 0.694

Lightphysicalactivity(min/d) 501(117) 489(112) 517(122) 0.140

Moderate-to-vigorousphysicalactivity(min/d)^† 95.8(82.7) 108(88.9) 77.7(62.6) <0.001

Academicachievement Grade1

Readingﬂuency(range0–28) 18.7(9.5) 17.8(10.4) 19.9(8.0) 0.147

Readingcomprehension(range0–12)^† 5.0(5.0) 4.0(6.0) 5.0(6.0) 0.134

Arithmeticskills(range0–28) 10.3(4.2) 10.3(4.4) 10.2(4.0) 0.856

Grade2

Readingﬂuency 24.8(8.4) 24.3(9.5) 25.5(6.8) 0.337

Readingcomprehension 8.0(5.0) 7.0(5.0) 9.0(4.0) 0.003

Arithmeticskills 15.4(5.1) 15.0(5.5) 16.0(4.4) 0.176

Grade3

Readingﬂuency 36.4(9.0) 34.7(9.3) 38.6(8.3) 0.007

Readingcomprehension 9.0(2.0) 9.0(3.0) 10.0(2.0) 0.126

Arithmeticskills 19.7(4.8) 19.5(4.6) 20.0(5.1) 0.508

DataarefromtheStudentt-testorMann–WhitneyUtestforcontinuousvariablesandchi-squaretestforcategoricalvariablesandaredisplayedasmeans(SD)fornormally distributedvariables,medians(IQR)^†forslightlyskewedvariables,orpercentages(%).Pvaluesrefertostatisticalsigniﬁcancefordifferencesbetweenboysandgirls.

MVPA,light PA,andST inorder toinvestigatetheindependent associationsofthesemeasureswithacademicachievement.We additionallyadjustedthedataontheassociationsofPAandSTwith academicskillsinGrades2–3foracademicskillsinGrade1.

The combined associations of MVPA and ST in Grade 1 withacademicachievementinGrades1–3wereinvestigatedby dichotomizingMVPAandSTattheirsex-speciﬁcmediansandcom- paringacademicachievementofchildren withacombinationof lowerlevelsofMVPA(≤109min/dforboys,≤79min/dforgirls) alongwithhigherlevelsofST(>215min/dforboys,>227min/d forgirls)toallotherchildren usinganalysesofcovariance with repeatedmeasuresadjustedforage,sex,andsensorweartime.We investigatedtheinteractionsofsexwithPAandSTonacademic skillsusinggenerallinearmodels.

Inadditiontotheadjustmentforage,sex,andsensorweartime, alldatawerefurtheradjustedforparentaleducation,household income,maximalworkloadachievedinexercisetest,50-mshuttle runtesttime,body fatpercentage,PAatdifferentsettings,time spentwatchingTV,usingacomputer,readingorwriting,riskfor readingdisability,orthewellbeingscore.

3. Results

Boyswereslightlyolder,had alowerbody fatpercentage, a highermaximalworkloadachievedinexercisetest,andafaster50-

mshuttleruntesttime,andweremorelikelytohaveincreasedrisk forreadingdisabilitythangirls(Table1).Boysalsohadhigherlev- elsofMVPA,poorerreadingcomprehensioninGrade2,andpoorer readingﬂuencyinGrade3thangirls.

Inallchildren,MVPAinGrade1wasdirectlyassociatedwith readingfluencyinGrades2and3(Table2).MVPAorSThadno otherstatisticallysignificantly associationswithacademicskills inallchildren(Table2).Furthermore,sexmodifiedmanyassoci- ationsofMVPA,lightPA,andSTwithreadingandarithmeticskills (Table2).

In boys,higher levels of MVPA in Grade 1 were related to betterreadingfluencyinGrades1–3andbetterreadingcompre- hensionandarithmeticskillsinGrade1afteradjustmentforage and sensorwear time(Table2).Higherlevels ofST in Grade1 wereassociatedwithpoorerreadingfluencyin Grades1–3and poorerarithmeticskillsinGrade1afteradjustmentfor ageand sensorweartime.LightPAinGrade1wasrelatedtobetterread- ingfluencyinGrade3.However,thisassociationwasnolonger statisticallysignificantafterfurtheradjustmentforMVPA,ST,or bodyfatpercentage(P>0.100).OnlytheassociationofMVPAwith readingfluencyinGrade2remainedstatisticallysignificantafter mutualadjustmentforMVPAandST.Furtheradjustmentforread- ingfluencyinGrade1attenuatedtheassociationsofMVPAorST inGrade1withreadingfluencyinGrade2(P>0.150)butithad noeffectontherelationshipsofMVPAorSTtoreadingfluencyin

(5)

586E.A.Haapalaetal./JournalofScienceandMedicineinSport20(2017)583–589

Table2

Associationsofphysicalactivityandsedentarytimewithacademicachievement.

Readingﬂuency Readingcomprehension Arithmeticskills

All Boys Girls Pforinteraction All Boys Girls Pforinteraction All Boys Girls Pforinteraction

Grade1 Sedentarytime

(min/d)

−0.111 −0.247^* 0.147 0.016 −0.097 −0.209 0.034 0.132 −0.089 −0.229^* 0.149 0.063 Lightphysical

activity(min/d)

0.028 0.138 −0.146 0.133 0.067 0.109 0.035 0.653 0.045 0.174 −0.166 0.109

Moderate-to- vigorousphysical activity(min/d)

0.163 0.285^** −0.171 0.011 0.105 0.255^* −0.211 0.007 0.133 0.227^* −0.108 0.094

(min/d)

−0.129 −0.225^* 0.028 0.119 −0.103 −0.118 −0.133 0.892 0.028 −0.152 0.379^** 0.009 Lightphysical

activity(min/d)

0.042 0.071 −0.001 0.715 0.076 0.033 0.191 0.438 −0.025 0.164 −0.385^** 0.006

0.192^* 0.309^** −0.183 0.008 0.113 0.190 −0.056 0.148 0.014 0.067 −0.155 0.309

(min/d)

−0.146 −0.354^** 0.158 0.004 −0.080 −0.134 0.066 0.324 −0.032 −0.155 0.163 0.111 Lightphysical

activity(min/d)

0.035 0.227^* −0.249 0.015 0.078 0.120 −0.037 0.455 −0.002 0.129 −0.203 0.105

0.275^** 0.350^** 0.100 0.238 0.130 0.156 0.001 0.500 0.086 0.134 −0.036 0.431

Dataarestandardizedregressioncoefﬁcientfrommultivariatelinearregressionanalysesadjustedforage,sex,andsensorweartime.Moderate-to-vigorousphysicalactivitywasdeﬁnedasactivitiesexceedingtheintensityof4 metabolicequivalents(METs),lightphysicalactivityasactivitiesthatwereperformedattheintensityof1.5–4metabolicequivalents(METs),andsedentarytimeasactivitiesthatwereperformedbelowtheintensityof1.5METs.

Therewere153children(89boys,64girls)inGrade1,149children(87boys,62girls)inGrade2,and145children(86boys,59girls)inGrade3.

Statisticallysigniﬁcantassociationsarebolded.

*P≤0.05.

**P≤0.01.

(6)

E.A.Haapalaetal./JournalofScienceandMedicineinSport20(2017)583–589 587

Fig.1. Differencesinreadingﬂuency,readingcomprehension,andarithmeticskillsinGrades1–3between30boyswithlowlevelsofMVPAandhighlevelsofSTandother 56boysandbetween22girlswithlowlevelsofMVPAandhighlevelsofSTandother37girlsadjustedforageandsensorweartime.Thedataarepresentedasestimated marginalmeansandtheir95%conﬁdenceintervals.

Grade3(P<0.030).Furthermore,therelationshipsofMVPAandST inGrade1toreadingandarithmeticskillsinGrade1werenolonger statisticallysigniﬁcantafteradditionaladjustmentforbodyfatper- centageor50-mshuttleruntesttime(P>0.065).Therelationship betweenMVPAandarithmeticskillsinGrade1wasalsoattenu- atedaftercontrollingforthewellbeingscore(P=0.092).Additional adjustmentshadnoeffectsontheseassociations(datanotshown).

Ingirls,lowerlevelsoflightPAandhigherlevelsofSTinGrade1 wererelatedtobetterarithmeticskillsinGrade2afteradjustment forageandsensorweartime(Table2).Theseassociationswereno longerstatisticallysigniﬁcantaftermutualadjustmentforlightPA andST(datanotshown).TherelationshipofSTinGrade1toarith- meticskillsinGrade2wasnolongerstatisticallysigniﬁcantafter

furthercontrollingforbodyfatpercentage.Additionaladjustments hadnoeffectonthisassociation.

BoyswithacombinationoflowerlevelsofMVPAandhigher levelsofSTinGrade1hadpoorerreadingfluency(meandiffer- enceacrossGrades1–3=−6.6,95% CI forthedifference=−10.6 to−2.6,P=0.002)andreadingcomprehension(meandifference acrossGrades1–3=−1.3,95%CIforthedifference=−2.5to−0.2, P=0.027)thanallotherboysafteradjustmentforageand sensorweartime(Fig.1).Boyswitha combinationoflower levels ofMVPAandhigherlevelsofSTalsohadpoorerreadingfluency inGrades2–3thanallotherboysafteradjustmentforage,sensorweartime,andreading fluencyinGrade1(meandifference acrossGrades2–3=−3.0,95%CIforthedifference=−5.4to−0.6,

(7)

588 E.A.Haapalaetal./JournalofScienceandMedicineinSport20(2017)583–589

P=0.014).Ingirls,nostatisticallysigniﬁcantcombinedassociations ofMVPAandSTinGrade1withreadingandarithmeticskillsin Grades1–3werefound.Furtheradjustmentshadnoeffectsonthese differences(datanotshown).

4. Discussion

Inthepresent studyamong Finnishprimaryschoolchildren aged6–8years,wefoundthatlowerlevelsofMVPA,higherlevels ofST,andparticularlytheircombination,wererelatedtopoorer readingskillsinboys.Furthermore,weobservedthathigherlevels ofMVPAandlowerlevelsofSTwererelatedtobetterreadingskills inGrades2and3,independentofreadingskillsinGrade1among boys.WefoundfewassociationsofPAandSTwithacademicskill ingirls.LowerlevelsoflightPAandhigherlevelsofSTwereassoci- atedwithbetterarithmeticskillsbuttherelationshipbetweenST andarithmeticskillsweakenedmarkedlyaftercontrollingforbody fatpercentage.

Previous investigations suggest either direct,⁸ non- significant,^5,10,17 or inverse⁹ relationships between objectively measured MVPA and academic achievement in children and adolescents.We founda direct association betweenMVPA and academicachievementin boys,while therelationship wassta- tisticallynon-significant orinverse in girls.Our observationon the direct associations of MVPA with reading skills in boys is in linewiththe resultsof studies showingdirect relationships of PA and cardiorespiratory fitness to academic achievement andworking memoryinboys butnot in girlsat 6–10years of age.^11,12 Nevertheless, some studies in adolescents suggest a direct association between PA and academic achievement in girls but not in boys.^13,14 It has been hypothesized that these sexdifferencesmaybeduetomoreadvancedsexualmaturation andhigherplasma levelsofcirculatingsex-hormones,orbetter psychosocialwellbeingamonggirlsthanboys.^12,28Anotherreason forthesesex-dependentassociationsmaybethattheeffects of PA onself-esteem or friendshipnetworks, which may increase schoolconnectednessandschooladjustmentandtherebyimprove academicachievement,arelargerinboysthaningirls6–8yearsof age.^29,30Furthermore,girlsmayreceivemoreeducationalsupport fromtheirparentsthanboys³¹whichmayhaveaneffectonthe associationsofPAandSTwithacademicachievementinchildren duringearlyschoolyears.Weakandinconsistentassociationsof PAand ST withacademicachievementamonggirl mayalsobe partlydue toa smallernumberof girlsthan boysinourstudy sample and therefore limited statistical power in theanalyses.

Furthermore,body fat percentage weakenedthe association of STwitharithmeticskillsingirls.Moreresearchiswarrantedon theeffects of these factorsonthe associationbetween PA and academicachievement.

BiologicalmechanismbywhichhigherlevelsofPAmayimprove academicachievementinboysincludeimprovedneuroelectricpro- cessing,increasedhippocampalvolumesandplasmaneurotrophic factor concentrations, enhanced blood ﬂow in the brain, and improvedattentionandworkingmemory.^2,3 Furthermore,some evidencesuggeststhathighlevelsofSTthatresultinlowenergy expendituremayimpairlearningbydecreasingplasmalevelsof brain-derivedneurotrophicfactor.²PAandSTlikelyhavesimilar effectsonthebrainingirlsandinboys,butotherfactors,suchas parentaleducationalsupport,peeracceptance,teachers’positive attitudeforthestudent,andchildren’smotivationtowardsschool aremoreimportantcorrelatesofacademicachievementthanPA andSTamonggirls.^32,33

HigherlevelsofTVwatchingandlowerlevelsofreadinghave beenassociatedwithpooreracademicachievementinchildren.^4,11 Therearefewstudiesontheassociationsofobjectivelymeasured

STwithacademicachievementinchildren. Someofthesestud- ieshaveshownaweakpositiverelationshipandotherstudiesno associationbetweenSTandacademicachievement.^5,16,17Wefound thathigherlevelsofobjectivelymeasuredSTwererelatedtopoorer reading ﬂuencyin boysindependentof screen-based sedentary behaviorandreading.However,theseassociationswereweakened aftercontrollingforMVPA,suggestingthatboyswithhigherlev- elsofSTalsohadlowerlevelsofMVPA.Accordingly,boyswitha combinationoflowerlevelsofMVPAandhigherlevelsofSThada 6.5pointlowerscoreinreadingﬂuencyanda1.3pointlowerscore inreadingcomprehensionthanotherboys.Thus,ourdatasuggest thatacombination oflowMVPAandhighST mightbeparticu- larlyharmfulforthedevelopmentofacademicskillsinboysand thatincreasingMVPA,reducingST,orespeciallybothofthemmay improveacademicachievement.

WefoundthattheassociationsofMVPAandSTwithreading andarithmeticskillsinGrade1,butnotinGrades2–3,weakened aftercontrollingforbodyfatcontentandmotorperformance.It iswellcharacterizedthatPA,adiposity,and motorperformance areinterrelated.³⁴Motorperformancehasbeendirectlyrelatedto academicskillsinchildren,particularlyinboys^18,25andadipos- ityhasbeeninverselyassociated withacademicachievementin children.³⁵Althoughourresultssuggestthatmoreactiveboyshad alowerbodyfatpercentage,bettermotorperformance,andbetter readingskillsinGrade1thanotherboys,physicallyactivelifestyle inGrade1mayimprovereadingskillslaterregardlessofthelevel ofadiposityandmotorperformance.Furthermore,higherlevelsof MVPAinGrade1wereassociatedwithbetterreadingﬂuencyin Grade3andthecombinationofhighlevelsofMVPAandlowlevels ofSTwasrelatedtobetterreadingﬂuencyinGrades2–3indepen- dentofreadingskillsinGrade1and otherconfoundingfactors.

Thus,inlinewiththeﬁndingsbyBoothetal.,⁸ ourresultssug- gestthatboyswithlowerlevelsofMVPAandhigherlevelsofSTin Grade1hadsmallerimprovementsinreadingskillsduringtheﬁrst schoolyearsthanotherboys.Anexplanationforourobservations mightbethathigherlevelsofMVPAandlowerlevelsofSTinGrade 1arerelatedtoimprovedbrainstructuresandfunctions,better cognitivefunctions,self-esteem,schoolconnectedness,andschool adjustmentthatmayimprovelearningparticularlyinboys.^3,28

Thestrengthsofthepresentsstudyaretheobjectivemeasures ofPAandST,theassessmentofreadingandarithmeticskillsusing standardizedtests,andtheprospectivestudydesign.Wealsoused acomprehensivebatteryofimportantcovariatesintheanalyses.

However,ourdatadonotallowustodrawconclusionsoncausal relationships, becausewe assessed PA and ST only in Grade 1.

We alsohad a relativelysmallsamplesizeparticularlyforgirls thatdecreasedpowertodetectstatisticallysigniﬁcantassociations.

Nevertheless,thecharacteristicsof thestudysamplewererela- tively comparable totheexcluded children. Becauseof a small samplesize,wewerenotabletoinvestigatethecombinedasso- ciationsofMVPAandSTwithacademicskillsusingfourpossible groupsofMVPAandSTbutcomparedchildrenwithlowerlevelsof MVPAandhigherlevelsofSTwithallotherchildren.

5. Conclusion

Inconclusion, lowerlevelsofMVPA,higher levelsofST, and especiallytheircombinationwererelatedtopoorerreadingskills inGrades1–3amongboys.Wefoundweakandinconsistentasso- ciationsofPAandSTwithacademicskillsingirls.Thus,ourresults providesomeevidencethatpromotingaphysicallymoreactive lifestylemaybeneﬁtthedevelopmentofreadingskillsinboysdur- ingtheﬁrstschoolyears.Moreresearchisneededtoinvestigate whethertheassociations ofPA andST withacademicskillsare differentinboysandgirls.

(8)

E.A.Haapalaetal./JournalofScienceandMedicineinSport20(2017)583–589 589 Practicalimplications

–Increasingdailyphysicalactivityanddecreasingsedentarytime mayimproveacademicperformanceparticularlyinboys.

–Acombinationoflowlevelsofphysicalactivityandhighlevels ofsedentarywasstronglyrelatedtopooracademicperformance inboys.

–Physicalactivityorsedentarytimehadsmallifanyassociation withacademicperformanceingirls.

Acknowledgements

ThisstudywasﬁnanciallysupportedbygrantsfromtheJenny andAnttiWihuriFoundation,theSinikkaandSakariSohlbergFoun- dation,theMinistryofSocial Affairs and Healthof Finland,the MinistryofEducationandCultureofFinland,theFinnishInnovation FundSitra,theSocialInsuranceInstitutionofFinland,theFinnish CulturalFoundation,theJuhoVainioFoundation,theYrjöJahns- sonFoundation,theFoundationforPaediatricResearch,thePaavo NurmiFoundation,thePauloFoundation,theDiabetesResearch Foundation,cityofKuopio,theKuopioUniversityHospital(EVO- fundingnumber 5031343)and theResearch Committeeof the KuopioUniversity HospitalCatchmentArea (the StateResearch Funding).Wethankthevoluntarysubjectsandtheirfamilieswho participatedinthisstudy.Wearealsogratefullyindebtedtothe PANICStudyresearchteammembersfortheirskillfulcontribution inperforming thestudy.We thankStefanieHollidge, MRCEpi- demiologyUnit,Cambridge,forherassistancewiththeobjective physicalactivitydataprocessing.TheworkofKW,UE,andSBis supportedbytheUKMedicalResearchCouncil(MCUU12015/3).

References

[1].LeeIM,ShiromaEJ,LobeloFetal.Effectofphysicalinactivityonmajornon- communicablediseasesworldwide:ananalysisofburdenofdiseaseandlife expectancy.Lancet2012;380(9838):219–229.

[2].VossMW,CarrLJ,ClarkRetal.RevengeofthesitII:doeslifestyleimpactneuronal andcognitivehealththroughdistinctmechanismsassociatedwithsedentary behaviorandphysicalactivity?MentalHealthPhysActivity2014;7(1):9–24.

[3].DonnellyJE,HillmanCH,CastelliDetal.Physicalactivity,ﬁtness,cognitive function,andacademicachievementinchildren.MedSciSportExerc2016;

48(6):1197–1222.

[4].Tremblay MS,LeBlanc AG,Kho MEetal. Systematic review ofsedentary behaviourandhealthindicatorsinschool-agedchildrenandyouth.IntJBehav NutrPhysActivity2011;8(1):98.

[5].SyväojaHJ,KantomaaMT,AhonenTetal.Physicalactivity,sedentarybehavior,andacademicperformanceinFinnishchildren.MedSciSportsExerc2013;

45(11):2098–2104.

[6].SinghA,UijtdewilligenL,TwiskJWRetal.Physicalactivityandperformanceat school.Asystematicreviewoftheliteratureincludingamethodologicalquality assessment.ArchPediatrAdolescMed2012;166(1):49–55.

[7].CorderK,vanSluijsEMF,WrightAetal.Isitpossibletoassessfree-livingphysical activityandenergyexpenditureinyoungpeoplebyself-report?AmJClinNutr 2009;89:862–870.

[8].BoothJN,LearySD,JoinsonCetal.Associationsbetweenobjectivelymeasured physicalactivityandacademicattainmentinadolescentsfromaUKcohort.BrJ SportsMed2014;48(3):265–270.

[9].Esteban-CornejoI,Tejero-GonzálezCM,Martinez-GomezDetal.Objectively measuredphysicalactivityhasanegativebutweakassociationwithacademic performanceinchildrenandadolescents.ActaPaediatr2014;103(11):501–506.

[10].LeBlancMM,MartinCK,HanHetal.Adiposityandphysicalactivityarenot relatedtoacademicachievementinschool-agedchildren.JDevBehavPediatr 2012;33(6):486–494.

[11].HaapalaEA,PoikkeusA-M,Kukkonen-HarjulaKetal.Associationsofphysical activityandsedentarybehaviorwithacademicskills—afollow-upstudyamong primaryschoolchildren.PLoSOne2014;10:e107031.

[12].DrolletteES,ScudderMR,RaineLBetal.Thesexualdimorphicassociation ofcardiorespiratoryﬁtnessto workingmemoryinchildren.DevSci2016;

19(1):90–108.

[13].KwakL,KremersSPJ,BergmanPetal.Associationsbetweenphysicalactivity, ﬁtness,andacademicachievement.JPediatr2009;155(6):914–918.

[14].Martínez-GómezD,RuizJR,Gómez-MartínezSetal.Activecommutingtoschool andcognitiveperformanceinadolescents:theAVENAstudy.ArchPediatrAdolesc Med2011;165(4):300–305.

[15].PoitrasVJ,GrayCE,BorgheseMMetal.Systematicreviewoftherelationships betweenobjectivelymeasuredphysicalactivityandhealthindicatorsinschool- agedchildrenandyouth.ApplPhysiolNutrMetab2016;41:197–239.

[16].Esteban-CornejoI,Martinez-GomezD,SallisJFetal.Objectivelymeasuredand self-reportedleisure-timesedentarybehaviorandacademicperformancein youth:theUP&DOWNstudy.PrevMed2015;77:106–111.

[17].CorderK,AtkinAJ,BamberDJetal.Revisingontherunorstudyingonthesofa:

prospectiveassociationsbetweenphysicalactivity,sedentarybehaviour,and examresultsinBritishadolescents.IntJBehavNutrPhysActivity2015;12(1):106.

[18].HaapalaEA,PoikkeusA-M,TompuriTetal.Associationsofmotorandcardiovas- cularperformancewithacademicskillsinchildren.MedSciSportsExerc2014;

46(5):1016–1024.

[19].LindemanJ.ALLU-ala-asteenlukutesti(standardizedreadingtestforcomprehen- siveschool),Turku,CenterforLearningResearch.UniversityofTurku,1998.

[20].RäsänenP,AunolaK.Testofarithmetics.Unpublishedtestmaterialdevelopedin theﬁrststepsfollow-up,Jyväskylä,UniversityofJyväskylä,2007.

[21].BrageS,BrageN,FranksPWetal.Reliabilityandvalidityofthecombinedheart rateandmovementsensorActiheart.EurJClinNutr2005;59:561–570.

[22].CollingsPJ,WijndaeleK,CorderKetal.Objectivelymeasuredphysicalactivity andlongitudinalchangesinadolescentbodyfatness:anobservationalcohort study.PediatrObes2015;11(2):107–114.

[23].ElorantaAM,LindiV,SchwabUetal.Dietaryfactorsassociatedwithoverweight andbodyadiposityinFinnishchildrenaged6–8years:thePANICstudy.IntJObes 2012;36(7):950–955.

[24].TannerJM.Growthatadolescence,Oxford,Blackwell,1962.

[25].HaapalaEA,LintuN,VäistöJetal.Associationsofphysicalperformanceandadi- positywithcognitioninchildren.MedSciSportsExerc2015;47(10):2166–2174.

[26].VäistöJ,ElorantaA-M,ViitasaloAetal.Physicalactivityandsedentarybehaviour inrelationtocardiometabolicriskinchildren:cross-sectionalﬁndingsfromthe physicalactivityandnutritioninchildren(PANIC)study.IntJBehavNutrPhys Activity2014;11(1):55.

[27].KiuruN,AunolaK,TorppaMetal.Theroleofparentingstylesandteacherinter- actionalstylesinchildren’sreadingandspellingdevelopment.JSchoolPsychol 2012;50(6):799–823.

[28].TomporowskiPD,LambourneK,OkumuraMS.Physicalactivityinterventions andchildren’smentalfunction:anintroductionandoverview.PrevMed2011;

52(Suppl.1):3–9.

[29].TrudeauF,ShephardRJ.Physicaleducation,schoolphysicalactivity,school sportsandacademicperformance.IntJBehavNutrPhysActivity2006;10:1–10.

[30].SawkaKJ,McCormackGR,Nettel-AguirreAetal.Friendshipnetworksandphys- icalactivityandsedentarybehavioramongyouth:asystematizedreview.IntJ BehavNutrPhysActivity2013;10:130.

[31].KristjánsonÁ, SigfúsdóttirI. Therole ofparentalsupport, parentalmoni- toring,andtimespentwithparentsinadolescentacademicachievementin Iceland:astructuralmodelofgenderdifferences.SandinavianJEducRes2009;

53(5):481–496.

[32].KiuruN,LerkkanenMK,NiemiPetal.Theroleofreadingdisabilityriskand environmentalprotectivefactorsinstudents’readingﬂuencyingrade4.Read ResQ2013;48(4):349–368.

[33].Aunola K,NurmiJ-E,Niemi Petal. Developmentaldynamicsof achieve- mentstrategies,readingperformance,andparentalbeliefs.ReadResQ2002;

37(3):310–327.

[34].Robinson LE, Stodden DF, Barnett LM et al. Motor competence and its effect on positive developmental trajectories of health. Sport Med 2015;

45(9):1273–1284.

[35].ReinertKRS,Po’eEK,BarkinSL.Therelationshipbetweenexecutivefunction andobesityinchildrenandadolescents:asystematicliteraturereview.JObes 2013;2013(2):820956.