Sleep problems and language development in toddlers with Williams syndrome
Emma L. Axelsson
a, Catherine M. Hill
b, Avi Sadeh
c, Dagmara Dimitriou
d,*
aUniversityofNewSouthWales,SchoolofPsychiatry,FacultyofMedicine,Australia
bDivisionofClinicalNeuroscience,SchoolofMedicine,UniversityofSouthampton,UK
cSchoolofPsychologicalSciences,TelAvivUniversity,Israel
dDepartmentofPsychologyandHumanDevelopment,InstituteofEducation,London,UK
1. Introduction
Sufficientsleepisdefinedasanamountthatisconducivetoeffectivesocialandneuropsychologicalfunctioningandis importanttohealthydevelopmentinchildren(Dahl,1999;Hill,Hogan,&Karmiloff-Smith,2007).Thereisanextensivebody ofevidencedemonstratingthatinsufficientsleepinterfereswithhighercognitivefunctioning(e.g.,Randazzo,Muehlbach, Schweitzer,&Walsh,1998;Walker&Stickgold,2006),andcontributestochildbehaviouralproblems(e.g.,Sadeh,Gruber,&
Raviv,2002).Therehavebeenfewerstudiesexaminingtheeffectsofsleepqualityintypicallydeveloping(TD)infantsand ARTICLE INFO
Articlehistory:
Received9July2013
Receivedinrevisedform8August2013 Accepted14August2013
Availableonline10September2013
Keywords:
Williamssyndrome Sleepproblems Maternaldepression Developmentaldisorders
ABSTRACT
Sleepandrelatedmaternalbeliefswereassessedinanarrowagerangeof18childrenwith Williamssyndrome(WS)and18typicallydeveloping(TD)children.WSisararegenetic disordercharacterisedbyacomplexphysical,cognitiveandbehaviouralphenotype.High prevalenceofsleepdifficultiesinolderchildrenandadultswithWShavebeenreported.
Parentscompleted6questionnaires:theBriefInfantSleepQuestionnaire,InfantSleep VignettesInterpretationScale,PittsburghSleepQualityIndexofParents,ChildBehaviour Checklist,MacArthurCommunicativeDevelopmentInventoryforInfants–Wordsand Gestures,andtheMajor(ICD-10)DepressionInventory.ComparedtoTDchildren,those withWS hadshorter night sleep,morenight wakingsand wakefulnessaccording to parentalreport.Regressionanalysesrevealedthataproportionofthevarianceinlanguage developmentscoresinWSchildrencouldbeexplainedbynightsleepduration.Compared tocontrolparents,themothersoftheWSgroupweremorelikelytodescribetheirchild’s sleepasproblematicandhadhigherratesofinvolvementwithchildsleep,yettheyhada lessertendencytointerpretsleepproblemsassignsofdistressandagreatertendencyto emphasiselimitsetting.Approximatelyhalfofbothgroupsofmothersexperiencedpoor sleepquality.This wasalso related tomaternalmood, andnight wakefulnessinthe childrenwithWS.Thisisthefirststudytoquantifysleepdifficultiesinyoungchildrenwith WSinanarrowagerangeusingmaternalreport.Thepossiblenegativeeffectsonmaternal sleepandmood,andthelinkbetweennightsleepandlanguagedevelopmentinyoung childrenwithWS,requiresfurtherdetailedinvestigation.
ß2013ElsevierLtd.Allrightsreserved.
*Correspondingauthorat:InstituteofEducation,UniversityofLondon,DepartmentofPsychologyandHumanDevelopment,25WoburnSquare,London WC1H0AA,UK.Tel.:+442076126229.
E-mailaddress:[email protected](D.Dimitriou).
ContentslistsavailableatScienceDirect
Research in Developmental Disabilities
0891-4222/$–seefrontmatterß2013ElsevierLtd.Allrightsreserved.
http://dx.doi.org/10.1016/j.ridd.2013.08.018
toddlersbutmountingevidencesuggeststhatinsufficientsleep,andfrequentnightwakingsintheearlyyearsareassociated withavarietyofshort-andlong-termnegativeoutcomessuchaslowerperformanceinlanguageandspatialtasks,higher levelsofimpulsivityandhyperactivity(Dearing,McCartney,Marshall,&Warner,2001;Touchetteetal.,2007),largerweight for length(Tikotzkyet al., 2010), lowerscores on measuresofmental and motor development(Dearinget al., 2001;
Freudigman&Thoman,1993;Scher,2005).
EstimatesofsleepdifficultiesamongstTDinfantsandtoddlersinAustralia,UK,US,Canada,andanumberofAsian countriessuggestthat20–50%ofparentsreportsomedegreeofdifficultywiththeirchild’ssleep,posingalargecostto healthservices(e.g.,Armstrong,Quinn,&Dadds,1994;Hiscock&Wake,2001;Johnson,1991;Martin,Hiscock,Hardy, Davey, &Wake,2007;Mindell, Sadeh,Wiegand,How, &Goh,2010),andreportsof associated negativementaland physicalhealthforparents(e.g.,Martinetal.,2007).Earlysleepproblemsareoftenpresumedtobetemporaryhowever, thisistypicallynotthecase,andmaypersistintochildhood(Touchetteetal.,2007;Zuckerman,Stevenson,&Bailey, 1987).Touchetteetal.(2005)foundthatathirdofinfantswhoweresleepinglessthan6hpernightat5and17months, continuedtohavethisproblemat29months.Evenwheninsufficientsleepat2.5yearsincreasedtosufficientlevelsby 3.5years,lowerscoresonaspatialtask,andhigherscoresonahyperactivity–impulsivitymeasurewerefoundatage6 comparedtothosesleepingmorethan10hanight.ThisisofparticularconcernnotonlyforTDchildren,butalsothose withdevelopmentaldisordersassleepproblemsmaycompoundtheirexistingcognitiveandbehaviouraldifficulties, andtheyareinherentlymorevulnerabletosleepdifficulties(Bartlett,Rooney,&Spedding,1985;Lancioni,O’Reilly,&
Basili, 1999).
ThemainaimofthecurrentstudywastoinvestigatesleepintoddlerswithWS.Todate,thereappeartobenopublished studiesprovidingquantifiablereportsinvestigatingsleepspecificallyintoddlerswiththisraredevelopmentaldisorder.WS iscausedbyahemizygousmicrodeletionofsome28genesonchromosome7q11.23(Tassabehji,2003).TheincidenceofWS is approximately 1 in 20,000 live births (Morris, Demsey, Leonard, Dilts, & Blackburn, 1988). The main cognitive characteristicsofWSincludeoverallIQscoresbetween55and69(Mervisetal.,2000;Searcyetal.,2004),a‘hyper-social’
personalityprofile,relativelygoodfacerecognitionandlanguageskills,andpoorvisuo-spatialskills(Annaz,Karmiloff- Smith,Johnson,&Thomas,2009;Donnai&Karmiloff-Smith,2000).
Studies examiningsleepin individualswith WS indicatethat sleep is problematic for this population. Annazand colleagues(2011)reportedthat97%ofschool-agedchildrenwithWSexperiencedbedtimeresistance,sleepanxiety,night wakings,anddaytimesleepiness.Morerecently,actigraphymeasuresrevealedthat,comparedtoTDchildrenandchildren withDownsyndrome,childrenwithWS(agerange:6–12years)hadsignificantlylongersleeplatenciesaswellasparental reports of bed-wetting and body pains (Ashworth, Hill, Karmiloff-Smith, & Dimitriou, 2013). A large proportion of adolescentsandadults(agerange:17–35years)withWSarealsoreportedtoexperiencedaytimesleepiness,nocturnalleg discomfort, and fragmented sleep as measured by actigraphy (Goldman, Malow, Newman, Roof, & Dykens, 2009).
PolysomnographystudieshavedemonstrateddifferencesinsleeparchitecturebetweenTDandWSchildren.Arensetal.
(1998)foundthatchildrenwithWSspentdoubletheamountoftimeawakeaftersleeponset,moretimeinstages3and4 (slowwavesleep),andlessinstages1and2.Reducedsleepefficiency(sleeponsettooffsettimeasaproportionoftimein bed),andREMsleep,aswellasincreasedslowwavesleep,respiratory-relatedarousal,andrestlessnesshavebeenseenin studieswithchildren(agerange:2–18years,Masonetal.,2011),andadolescentsandadults(agerange:14–29years,Bo´dizs, Gombos,&Kova´cs,2012;Gombos,Bo´dizs,&Kova´cs,2011).
DuetothemountingevidenceseeninchildrenandadultswithWS,Annaz,Hill,Ashworth,Holley,andKarmiloff-Smith (2011)arguedthatsleepdifficultiesshouldbeconsideredoneofthedefiningsymptomsofWS,asentimentalsosharedby someclinicalprofessionals.Theagesoftheparticipantsinthestudiesreviewedabovewerewideinrange,andthereappear tobenostudiesinvestigatingsleepspecificallyinWSinthefirstfewyearsoflife.Investigatingsleepintheearlyyearsis essentialasitisunclearifsleepproblemsareaphenotypiccharacteristicofWS(Bo´dizsetal.,2012)orwhethersleep difficultiesdevelopasaresultofotherdomainspecificfactorsassociatedwithWSsuchasissueswithattention(e.g.,Scerif, Cornish,Wilding,Driver,&Karmiloff-Smith,2004),hypersociabilityorbehaviouralproblems(Bo´dizsetal.,2012;Mason etal.,2011).
Sleep involves finely tuned multidimensional processes such as psychological processes, biochemistry, genetics, and responses to external environmental cues (Hill, 2011). Therefore multiple types of regulation may affect the complexsleep-wakesystem,whichconsequentlymayimpactonwakinglifeandfamilyfunctioning.Sleepproblemsin earlychildhoodmayhaveanegativeimpactonmultiplefactorssuchasbehaviour,cognition,language,andhealthofa child (e.g., Dearing et al., 2001; Scher, 2005; Tikotzky et al.,2010; Touchette et al., 2007). Furthermore, parental involvementmaycontributetoinfantsleepdisturbancesandinfants’abilitiestoself-sooth(Sadeh,Flint-Ofir,Tirosh,&
Tikotzky,2007).Thismayparticularlybethecaseinprimarycaregiversofchildrenwithdevelopmentaldisorders(e.g., Meltzer,2008)astheymaybeatincreasedriskofassociatedstress,andexperiencemoreparentalanxiety.Theremay bebidirectionaleffectswheresleepdifficultiesinchildrenorparentsmayimpactoneachother(e.g.,Meltzer&Mindell, 2007;Wayte,McCaughey,Holley,Annaz,&Hill,2012).Therefore,inthecurrentstudy,arangefactorswereexplored in six questionnaires assessing child sleep, parental cognition in relationto early child sleep, maternal sleep, and mood measures.The maingoalsof thestudywere:(i) tocomparesleep in TDtoddlers andthosewith WSfroma narrowagerange,usingparentalreports;(ii)toexploretherelationshipsbetweensleepandbehaviourandlanguage outcomes;and(iii)toexaminetherelationshipsbetweenchildsleepandrelatedmaternalbeliefs,maternalsleep,and maternalmood.
2. Method 2.1. Participants
Twenty-eightparentsoftoddlerswithWSwerecontactedthroughtheWilliamsSyndromeFoundation,UKdatabase(88%
ofchildrenupto48monthsofageregistered).FourteenparentsofWSchildren(10female,4male)returnedthecompleted questionnaires.Afurther4parentswereinterviewedbytelephonefortheBriefInfantSleepQuestionnaire(BISQ,Sadeh, 2004).Theparentsof18TDchildren(12female,6male)alsocompletedthequestionnaires(4ofwhomonlycompletedthe BISQ).TheTDchildrenwereindividuallyagematchedtothechildrenwithWS.Thepairedparticipantsdifferedinageon averageby8.33days(SD=5.36,range4–22days,seeTable1).Therewerenosignificantdifferencesinchronologicalage between thepaired children (M difference 0.12m, SD=0.31m, t(17)=1.56,p=.14, d=0.01).Sixty-onepercent ofthe participants(11pairs)werealsomatchedongender.Alloftheparticipantswerebornfull-termapartfromoneineachgroup.
A priori chi-square analyses revealed that the two groups did not differ on gender (
x
2(1)=0.47, p=.49), ethnicity (x
2(3)=6.00,p=.11),maternaleducation(x
2(3)=5.01,p=.15),andparentaloccupation(x
2(2)=0.80,p=.67).Therewasa differenceinrelationtobreastfeedingasmoreoftheTDtoddlershadbeenbreastfed(WS:57%,TD:100%;x
2(1)=7.64, p=.006),andforalongertime(WS,Mmonths:4.25months,SD:6.61;TD,Mmonths:9.14,SD:6.52;t(13)=2.19,p=.047, d=0.74).ChildrenwithWShadbeendiagnosedclinically,aswellasbymeansofthefluorescenceinsituhybridisation(FISH)genetic testfordeletionofonecopyoftheElastingene.Allindividualshadnormalorcorrected-to-normalvision.Parentswereasked abouttheirchild’scurrentandpasthealthissues,medicationuse,anddiet.Notethatnoneofthechildrenhadanyformal diagnosisofasleepdisorder,althoughonechildwithWShadbeenprescribedmelatoninforsleepdifficulties.Thestudywas approvedbytheWilliamsSyndromeFoundation,UKandMiddlesexUniversityEthicsCommittee,andwrittenconsentwas providedfromalloftheparents.
2.2. Sleep-relatedquestionnaires
1.BriefInfantSleepQuestionnaire(BISQ,Sadeh,2004).Basedonparentalreports,thisquestionnaireprovidesinformation onsleepandthecircumstancessurroundingsleep,suchasduration(nightandday),numberanddurationofnightwakings, settlingtime, sleeplatency,sleeplocation(e.g.,own bed,parent’sbed), position(e.g.,prone, supine,side),conditions surroundingsleeponset,andparents’perceptionoftheirchild’ssleepquality.Ithasbeenfoundtocorrelatepositivelywith sleepdiariesandactigraphy(Sadeh,2004).
2.InfantSleepVignettesInterpretationScale(ISVIS,Sadehetal.,2007)examinesparentalunderlyingbeliefsaboutinfant sleep,andhasbeenfoundtobepredictiveofinfants’nightwakings(Tikotzky&Sadeh,2009).Fourteenbriefvignettesof infantandtoddlersleepscenariosarepresentedandcaregiversareaskedtoratethreequestionsona6-pointscale.Thethree questionsareaimedatgaugingparents’beliefsaboutthedegreeofparentalinvolvementrequired,limitsetting,andthe extenttowhichagivensleepissueisrelatedtochildtemperament.Note,theISVISlargelyreferstoinfantsandtoddlersupto 24monthsofagebutduetothelowgeneralfunctioningoftheWSgroup,itwasdeemedusefultousematerialsthatwould covertheirleveloffunctioning.
3.PittsburghSleepQualityIndexofParents(PSQIP,Buysse,ReynoldsIII,Monk,Berman,&Kupfer,1989)exploresadults’
sleepwithquestionsratedona4-pointscaleandyieldssevencomponentscoresaswellasaglobalscore.Itisreportedto haveasensitivityof0.90andspecificityof0.87inthedetectionofsleepdisturbances(Buysseetal.,1989).Itgeneratesa maximumdifficultyscoreof21andascoreabove5indicatespoorsleepquality.
4.ChildBehaviourChecklist1.5–5(CBCL1.5–5,Achenbach&Rescorla,2000)isawidelyusedquestionnairecompletedby parentswith99 questionsratedon a 3-pointscale.Thequestions yieldscoreson 7 sub-scales(EmotionallyReactive, Anxious/Depressed,SomaticComplaints,Withdrawn,SleepProblems,AttentionProblems,AggressiveBehaviour,&Other) for1.5 to5-year-oldchildren. Thesescoresare comparedtonormsthatcorrespond tonormal, borderline(93rd–97th percentiles)andclinicalranges(scoresabovethe97thpercentile).
5.MacArthurCommunicativeDevelopmentInventoryforInfants(MCDI)–WordsandGestures(Fensonetal.,1993)isa standardisedmeasureoflanguagedevelopment,suitablefor8-to30-month-olds.Itisbasedonparentalreportsofinfants’
andtoddlers’receptivelanguageandtheirproductionofgestures,wordsandsentences.Sleepanddaytimenappingin
Table1
AgeinMonthsfortheChildMeasures:BISQ,ISVIS,CBCL,MCDI.
AgeMonths n M SD Range
BISQonly
Williams 18 30.43 9.53 15.40–48.20
Typicallydeveloping 18 30.31 9.44 15.57–47.47
Allchildmeasures
Williams 14 31.67 9.58 18.50–48.20
Typicallydeveloping 14 31.55 9.50 18.03–47.47
infancyhavebeenfoundtocorrelatewithlaterscoresonlanguagedevelopment(Dearingetal.,2001;Touchetteetal.,2007) andlanguagelearning(Go´mez,Bootzin,&Nadel,2006).
6.Major(ICD-10)DepressionInventory(MDI,Bech,Rasmussen,RaabækOlsen,Noerholm,&Abildgaard,2001)isashort questionnairewithitemsratedon6-pointLikertscales.Thequestionsareaimedatreflectingmoderatetoseveresymptoms of depression listedin the Diagnostic and Statistical Manual of MentalDisorders-IV (DSM-IV) and the International ClassificationofDiseases(ICD-10).Ithasasensitivityof0.86–0.92andspecificityof0.82–0.86inthediagnosisofclinical depression.Totalscoresequatetoclinicalsymptomsasfollows:‘milddepression’(20–24),‘moderate’(25–29)or‘severe’
(30+).
3. Results
DatawereanalysedusingIBMSPSSforWindows,Version20(SPSSInc.,Chicago,IL).Datawerescreenedforoutliersusing z-scores, and Cook’s distances. Age-matched analyses (paired t-tests, logistic regressions) were performed with the continuousdatabetweentheWSandTDgroupsfortheBISQ,ISVIS,PSQIP,andMDI.ThecategoricalvariablesoftheBISQ were analysed using chi-square, and thelogistic regression analyses were based on whether the children with WS experiencedagivencharacteristiccomparedtotheirage-matchedTDpair.SomeoftheBISQvariables,asbasedonthe groupeddata(asopposedtotheage-matchedanalyses),werenotnormallydistributed.Therefore,Kendall’s
t
correlations wereperformed(duetothesmallsamplesizes)andthevariablesinthehierarchicalmultipleregressionanalyseswerelog transformed.1.BISQAccordingtoparentalreports,thechildrenwithWShadsignificantlyshorternightsleepduration(t(17)=3.21, p=.005,d=0.99),hadsignificantlymorenightwakings(t(17)=2.22,p=.04,d=0.83),longernightwakefulness(t(17)=3.14, p=.006, d=1.02), tooklonger tosettle(t(16)=2.79,p=.01, d=0.90)and fell asleeplater at night(t(17)=2.20, p=04, d=0.61),comparedtoage-matchedTDchildren(Table2).Therewerenosignificantdifferencesindaytimesleepduration (t(14)=0.93,p=.37,d=0.44).
Forthecategoricalvariables,theoptionsforanumberoftheitemsweredichotimisedduetothesamplesizes(e.g.,sleep withaparentornot;sleepasasmallorseriousproblemversusnotatall;fallingasleepaloneorwithaparent,seeSadeh, 2004).Groupcomparisonsusingchi-squareandlogisticregressionanalysesrevealedthatsignificantlymorechildrenwith WSsleptwithaparent(
x
2(1)=4.43,p=.035)andmoremothersfromtheWSgroupreportedthattheirchildhadasmallor serioussleepproblem(x
2(1)=10.6,p=.001),whichwasalsoreflectedinahighoddsratio(seeTable3).Thatis,parentsof childrenwithWSwere21timesmorelikelytodescribetheirchild’ssleepasasmallorseriousproblemthanparentsofTD children.Therewerenosignificantdifferencesinthefrequencyofparentalinvolvementwhenfallingasleeporinsleep positions(seeTable3).2.ISVISMothersofchildrenwithWShadsignificantlylowerscoresthanmothersofTDchildren(maternalscoresmatched bychronologicalageofthechildren)onexpectationsofdistressinrelationtochildsleep(t(12)=2.51,p=.027,d=0.86)and significantlyhigherscoresonlimit-settingbeliefs(t(12)=3.80,p=.003,d=1.23)(seeTable2).Relationshipsbetweenthe ISVISscoresandnightsleepduration,numberofnightwakings,anddurationofnightwakefulnessfromtheBISQwere
Table2
BISQ,ISVIS,PSQIP,andMDIScores.
WSgroup TDgroup
n M SD n M SD
BISQ
Durationsleepnight(7pm–7am) 18 9:20 2:23 18 11:06 0:50
Durationsleepday(7am–7pm) 15 1:06 0:52 18 0:45 0:42
Numbernightwakings 18 1.92 1.88 18 0.72 0.83
Durationnightwakefulness 18 0:49 0:59 18 0:06 0:07
Settlingduration 17 0:41 0:47 18 0:11 0:06
Timefallasleep 17 20:13 1:08 18 19:41 0:30
ISVIS
Infantdistressed 13 2.97 0.73 14 3.71 0.97
Needtolimitinterference 13 4.28 0.53 14 3.36 0.92
Infanttemperamentascause 13 3.07 0.80 14 3.01 1.01
PSQIPcomponentscores
1Subjectivesleepquality 14 1.07 0.83 14 1.00 0.39
2Sleeplatency 14 0.57 0.51 14 0.93 0.73
3Sleepduration 14 0.79 0.80 14 0.43 0.51
4Habitualsleepefficiency 14 0.36 0.74 14 0.21 0.43
5Sleepdisturbances 14 1.00 0.39 14 1.07 0.47
6Sleepmedication 14 0.00 0.00 14 0.14 0.53
7Daytimedysfunction 14 0.93 0.92 14 0.50 0.65
PSQIPglobalscore 14 4.71 3.07 14 4.29 2.33
MDI 11 7.55 5.81 11 7.36 7.61
assessedseparatelyfortheWSandTDgroups.1IntheTDgroup,therewasasignificantpositiverelationshipbetweenthe meannumberofchildnightwakingsandperceptionsofdistress(
t
(12)=.55,p=.015),andasignificantnegativerelationship betweenthenumberofnightwakingsandattitudestowardslimitsetting(t
(12)= .56,p=.013).Therefore,theattitudesof themothersofTDchildrenthatreflectgreaterinterpretationsofdistressintheirchildrenandlesslimit-settingmaybe linkedtomorenightwakingsintheTDchildren.Incontrast,therewerenosignificantrelationshipsseenintheWSgroup.Thus, the attitudes towards sleepin the mothers of the children with WS may be unrelated to the reported sleep disturbances.
3.PSQIPPairedsamplest-testsshowednosignificantdifferencesbetweenthemothersofTDandWSchildrenonall7 componentsofthisscale,northeglobalscalesuggestingthatthesleepqualityofthemothersinbothgroupsweresimilar (seeTable2).PSQIPGlobalscoresabove5areindicativeoftheexistenceofsleepdifficulties.HalfofthemothersofWS children(n=7)and36%ofthemothersofTDchildren(n=5)hadscoresabove5,anon-significantdifferenceinfrequency.
Kendall’s
t
correlations1wereperformedseparatelyforeachgroupbetweentheglobalPSQIPscoreandBISQvariables:night sleepduration,meannumberofnightwakings,anddurationofnightwakefulness.FortheWSgroup,thereweresignificant positivecorrelationsbetweenthePSQIPglobalscoreandthemeannumberofnightwakings(t
(12)=.58,p=.01)andnight wakefulnessinthechildren(t
(12)=.53,p=.016).ThissuggeststhatasthemeannightwakingsandwakefulnessintheWS childrenincreased,thepoorerthesleepqualityofthemothers.Table3
BISQFrequencesandOddsRatiosandCBCLFrequencies.
WS TD
BISQ n n B SE OddsRatio
(WSonlyvs.TD)
LCL UCL
Howfallasleep
Alone 8 12
Parentinvolved 10 6 0.92 0.69 0.40 0.10 1.54
Constant 1.32 1.05
Sleeparrangement
Alone/shareroomwithsibling 12 17
Withparent 6 1 2.14 1.14 8.5 0.90 80.03
Constant 1.79 1.08
Sleepproblem
No 8 17
Small/serious 10 1 3.06 1.13 21.25 2.31 195.80
Sleepposition
Onbelly 7 7
Onside 8 4 0.85 0.87 0.43 0.77 2.37
Onback 3 7 1.54 0.92 0.21 0.04 1.31
Constant 0.85 0.69 2.33
CBCLsub-scales Emotionallyreactive
Normal 8 14
Borderline/clinical 6 0
Anxious/depressed
Normal 12 14
Borderline/clinical 2 0
Somaticcomplaints
Normal 11 14
Borderline/clinical 3 0
Withdrawn
Normal 12 14
Borderline/clinical 2 0
Sleepproblems
Normal 10 14
Borderline/clinical 4 4
Attentionproblems
Normal 7 14
Borderline/clinical 7 0
Aggressiveproblems
Normal 11 14
Borderline/clinical 3 0
1 WithBonferroniadjustments.
4.CBCL 1.5–5.Scores abovethe93rdand97thpercentilesforeach sub-scaleareclassifiedasborderlineorclinical.
Comparedtocontrols,significantlyhigherfrequenciesofchildrenwithWSwereclassifiedasborderline/clinicalonthe Emotionally Reactive (
x
2(1)=7.63,p=.006),SleepProblems (x
2(1)=4.67,p=.031)and Attention Problems sub-scales (x
2(1)=9.33,p=.002)(seeTable3).Point-biserialcorrelations1wereperformedbetweenthesesub-scalesandthemean night sleepdurationand meannumber ofnight wakings withtheWS group.OnlytheSleepProblemssub-scalewas significantlynegativelyrelatedtothelogtransformednightsleepduration(rpb(12)= .77,p=.001).Thushigherscoreson thisscale(indicatinggreaterseverity)wereassociatedwithlesssleepasrecordedintheBISQ.5.MCDIAsmostoftheTDchildrenwereatceilingforthis measureonlythedatafortheWSgroupwasanalysed.
Hierarchicalmultipleregressionanalysiswasperformedassessingthevarianceintherawscoreforthenumberofwords producedandunderstood(M=120.71,SD=137.76,range:7–345)inrelationtotheageoftheparticipantsandnightsleep duration(logtransformed).Chronologicalagewasasignificantpredictor,butsleepdurationalsoexplainedasignificant portionofthevarianceintheMCDIscore(seeTable4).
6.MDIOnemotherfromtheTDgroupmetthecriteriafordepressionwithascorethatfellwithintherangeformild depression.TherewerenosignificantdifferencesfoundbetweentheWSandTDgroupsontheMDITotalscore.However, significantpositiverelationshipswerefoundinbothgroupsofmothersbetweentheMDIandPSQIP(WS:r(11)=.74, p=.004,TD:r(11)=.56,p=.045).Multipleregression analyseswereperformedseparatelyfortheWSandTDgroups assessing therelationship betweentheMDIandthelogtransformedBISQnightsleepdurationandnumberofnight wakingsvariables.Thenumberofnightwakingswasasignificantpredictorofmothers’MDIscoresfortheWSgroup(see Table5).
4. Discussion
ThisisthefirststudytoexploresleepintoddlerswithWS,araredevelopmentaldisorder.Wesampledover50%ofthe toddlersupto48monthsofageregisteredontheWSFoundationUKdatabase.Thefindingshereprovideevidencethatsleep problemsareprevalentearlyinlifeinchildrenwithWS.Accordingtoparentalreport,theWSchildrenherehadshorternight sleepduration,morenightwakings,morenightwakefulness,tooklongertosettle,andhadlaterbedtimesthanage-matched TDchildren.ThisisconsistentwithpreviousstudieswitholderchildrenwithWS(e.g.,Annazetal.,2011;Arensetal.,1998;
Bo´dizsetal.,2012;Masonetal.,2011).Additionally,ahighproportionofmothersofchildrenwithWSconsideredtheir child’ssleeptobeproblematic.
MoreofthechildrenwithWSsharedabedwithaparentbutthereweresimilarratesofparentalinvolvementinbedtime routinesacrossbothgroups.However,theparentsofchildrenwithWShadpossiblyadaptedtochild’ssleepdifficultiesas theirscoresoninterpretationsofdistressinchildsleepdifficultieswerelowerthanparentsofTDchildren,andtheirscores onbeliefsaboutlimitsettingaroundchildsleepwerehigher.SimilartoTikotzkyandSadeh(2009),relationshipsbetween childnightwakingsandhigherperceptionsofdistressandlowerscoresonattitudestowardslimitsettingamongstthe parentsofTDchildrenwerefound.Interestingly,thiswasnotthecaseamongsttheparentsofWSchildren,suggestingthat parentalattitudesareunlikelytoexplainthefrequencyofnightwakingsintheWSchildren.
Table4
MultipleregressionanalyseswithMCDIUnderstandsandSaysrawscore,age,andnighttimesleep.
B SEB b
Step1
Constant 270.81 69.97
Age 12.36 2.12 .86**
Step2
Constant 2687.70 900.95
Age 10.91 1.80 .76**
Lognightsleep 541.21 201.35 .34*
R2=.74forStep1,DR2=.1forStep2(p=.021).
* p<.05.
** p<.001.
Table5
MultipleregressionanalyseswithmaternalMDIandWSchildren’sBISQnightsleepandnumberofwakings.
B SEB b
Constant 42.73 77.53
Logsleepnight 9.96 16.83 .15
Lognightwakings 16.76 5.29 .80*
R2=.53.
* p<.05.
Thepresentfindingsrevealedaninterestingassociationbetweennightsleepdurationandlanguagedevelopmentinthe childrenwithWS. Nighttimesleepdurationaccountedfor10%ofthevarianceinameasureoflanguagedevelopment, beyondthataccountedforbyage.RelationshipsbetweensleepandlanguagedevelopmenthavebeenfoundwithTDtoddlers (e.g.,Dearingetal.,2001;Touchetteetal.,2007),butthisisthefirstfindingthatweareawareofwithtoddlerswithWS.
Languagedevelopment in children withWS is relatively goodyet atypical (Laing etal., 2002). Furtherinvestigations preferablyintheformofalongitudinalstudyandalternativelanguagemeasureswouldbeofinterest.
ConsistentwithMasonandcolleagues(2011),wedidnotfindanyrelationshipbetweensleepanddaytimebehaviour (CBCL).ThisisperhapssurprisingaschildrenwithWStypicallyhaveproblemswithsustainedattention,andupto65%are diagnosedwithbehaviouralproblemsandattention-deficithyperactivitydisorder(ADHD)(Dykens,2003;Einfeld,Tonge,&
Florio,1997;Leyfer,Woodruff-Borden,Klein-Tasman,Fricke,&Mervis,2006;Pober&Morris,2007).Sleepproblemshave alsobeenfoundtobestronglyrelatedtobehaviouralproblemssuchasADHDinchildren(Cohen-Zion&Ancoli-Israel,2004).
However,Masonetal.didnotfinddifferencesinsleepvariablesbetweenWSparticipantswithADHDfeaturesandthose without.Itispossiblethatlargersamplesizeswithnarroweragerangesarerequiredtodetecttheeffectsofsleepon behaviour.Perceptionsofproblematicbehaviourarelikelytovaryacrossdifferentagerangesthusrenderingthedatahighly variable.Eveninthecurrentstudy,theagerangewasrelativelynarrowcomparedtopreviousstudiesbutbehaviourvaries dramaticallyintheearlyyears.Moresensitivemeasuresofbehaviour,cognition,andlanguagearealsorequired.InolderTD children,experimentalsleeprestrictionshavebeenreportedtonegativelyaffectclassroombehaviourandattention(Fallone, Acebo,Seifer,&Carskadon,2005)aswellasdecreasedverbalcreativityand abstractthinking(Randazzoetal., 1998).
Associationshave also beenfoundbetween poorer sleepefficiency and visuospatialworkingmemory in TD children (Steenarietal.,2003).Incontrast,sleep-dependentmemory,aneffectwhereproceduralanddeclarativeperformanceon recentlylearnedtasksisenhancedfollowingaperiodofsleepascomparedtoanequalnumberofwakinghours,isfrequently foundinadults(Walker&Stickgold,2006),andinsomestudieswithchildren (e.g.,Ashworth,Hill,Karmiloff-Smith,&
Dimitriou,2013b;Wilhelm,Prehn-Kristensen,&Born,2012).However,thiseffectwasrecentlynotobservedinchildrenwith WS(Dimitriou,Karmiloff-Smith,Ashworth,&Hill,2013).Therefore,moreexperimentalmeasureswithyoungchildrenwith WSarenecessarytocharacterisethespecificeffectsofsleepdifficultiesoncognition,language,andbehaviour.
Therewerenosignificantdifferencesseeninthemeasuresofmaternalsleepqualityandmood(asmeasuredbytheMDI) betweenthemothersofTDandWSchildren.However,upto50%ofthemothersinbothgroupshadscoresindicativeofpoor sleepquality.Furthermore,thesleepqualityscoresofthemothersofchildrenwithWSweresignificantlyrelatedtotheir child’snightwakingsandnightwakefulness.Gressetal.(2010)alsofoundthatinfantnightwakingswasapredictorof maternalsubjectiveratingsoftheirownsleepquality.Amoreobjectivemeasureofmaternalsleepqualitywouldbeusefulto determineifmaternal sleepisin factdisturbed orwhethertheyperceive their sleeptobepoor becauseof thesleep difficultiesoftheir children.Equally,itispossiblethatmotherswithpoorsleepqualityaremorevigilantatnightand thereforemorelikelytobeawareoftheirchild’snightwakings.Interestingly,thenumberofnightwakingsinthechildren withWSaccountedforasignificantproportionofvarianceinthematernalmoodscores(MDI).Wehypothesisethatnight wakingsinchildrenwithWSmaydisruptmaternalsleepandinturnimpactonmaternalmood.Sleeplossinadultsis detrimentaltotheirmoods,whichmayaffecttheirowndaytimefunctioningandabilitytocopewithparenting(Banks&
Dinges,2007;Moore,David,Murray,Child,&Arkwright,2005).Inturn,thismayaffectparentalsleepandfuturestudieswith objectivemeasurescouldbeusedtoquantifythisassociation.
ApartfromDimitriouandcolleagues’(2013)study,therelationshipsbetweensleepinchildrenwithWSandcognitive domains,andbehaviourhavebeenimplicatedbutnotsupported.Apossiblerelationshipbetweennightsleepdurationand languagedevelopmentwasfoundhere,butnotbetweensleepandbehaviour.Giventhenumerousstudieslinkingsleepand behaviour(e.g.,Jansenetal.,2011;Touchetteetal.,2007),andtheattenuationofbehaviouralproblemsfollowingsleep interventions(Chervinetal.,2006),thisissomewhatsurprising.Further,bothADHDandPLMSarereportedtobeprevalent intheWSparticipantsinsleepstudiesbutclearrelationshipswithsleephavenotbeenfound(e.g.,Arensetal.,1998;Mason etal.,2011).Addingtothecomplexityofthisissue,someoftheparentsofthechildrenwithWSinthisstudyreportedthat theirchildsleptwell,yethadobservedhighlevelsofmovementduringnighttimesleep.Nightwakingscouldbeareflection ofincreasedsleepspindleactivityassociatedwithincreasedthalamocorticaloscillatorydynamicsinWSassuggestedby Bo´dizsetal.(2012).Equally,behaviouralinsomnia(ICSD)suchaslimitsettingdisorderorsleeponsetassociationdisorder maybeinvolved(Hill,2011).
4.1. Limitationsofthecurrentstudy
Relyingonparentalreportsisofcoursenotideal.TheBISQhasbeenfoundtocorrelatewellwithactigraphy(Sadeh,2004), however,comparedtosleepdiariesrecordedbyparents,morenightawakeningsandshortersleephavebeenfoundamongst infantswithactigraphy(Aceboetal.,2005;Sadeh,1994,1996;Sadeh,Acebo,Seifer,Aytur,&Carskadon,1995;So,Adamson,
&Horne,2007;So,Buckley,Adamson,&Horne,2005).Therefore,thereportsheremayunderestimatethesleepdifficulties experiencedbythechildrenwithWSandpossiblytheTDchildren.Parentsarereportedtobeoftenunawareoftheirchild’s sleepproblems(Owens&Witmans,2004butseeGringrasetal.,2012fordiscussionongoodsleepestimatesusingsleep diaries).Masonetal.(2011)foundthatsomeWSchildrenwithlowsleepefficiency,accordingtopolysomnography,were describedassleepingsufficientlybytheirparents.Futurestudieswouldbenefitfromtheuseofobjectivemeasuresbutone needstobearinmindthedifficultiesoftestingchildrenwithdevelopmentaldisorders.Thisstudyisthefirsttoprovide
evidenceofsleepproblemsearlyindevelopmentforWSwithanarrowagerangeusingsubjectivemeasures.Thenextstepto extendourknowledgewouldbetouseobjectivemeasuressuchasactigraphyandexaminesleepintoddlerswithdifferent syndromes.
4.2. Clinicalandeducationalimplications
Significantchangesoccurduringearlycentralnervoussystemdevelopment,andassleepmakesupamajorpartoflifein theearlyyearsitisthushugelyimportantforbraindevelopment(Kohyama,1998;Owens&Witmans,2004).Interventions andeducationforparentsontheimportanceofsleephygieneinearlychildhoodhasbeenshowntobebeneficial(Kerr, Jowett, &Smith,1996; Tikotzky&Sadeh, 2009; Wolfson,Lacks, &Futterman,1992). Despitethe prevalenceof sleep difficulties,paediatriciansandGPsreportlackingconfidenceindiagnosingandtreatingsleepproblemsintheearlyyears (Chervin, Archbold, Panahi, & Pituch, 2001; Owens, 2001; Owens & Witmans, 2004). Ideally, greater awareness, interventionsandrecommendationsshouldbeprovidedforparentsofchildrenwithWSintheearlymonths.Thismay contributesignificantlytolong-termoutcomesandmayalsohelptopreventotherissuessuchassleepresistancebehaviours laterinlife(Annazetal.,2011).
Acknowledgements
WewouldliketothanktheparentsandchildrenfromtheUKWilliamsSyndromeFoundationUKandparentsoftypically developingchildreninSEEnglandforparticipating.ThisstudywasfundedbyAssociation‘AutourdesWilliams’,France.We wouldalsoliketothanktheUniversityofSussexWORDLabmembers:DrJessicaHorst,KatherineTwomey,MatthewHilton, JemimaScholfield,LucyMetcalfe,andNicoleJones.
References
Acebo,C.,Sadeh,A.,Seifer,R.,Tzischinsky,O.,Hafer,A.,&Carskadon,M.A.(2005).Sleep/wakepatternsderivedfromactivitymonitoringandmaternalreportfor healthy1-to5-year-oldchildren.Sleep,28(12),1568–1577.
Achenbach,T.M.,&Rescorla,L.A.(2000).ManualfortheASEBAPreschoolForms&Profiles.Burlington,VT:UniversityofVermont,ResearchCenterforChildren, Youth,andFamilies.
Annaz,D.,Hill,C.M.,Ashworth,A.,Holley,S.,&Karmiloff-Smith,A.(2011).CharacterisationofsleepproblemsinchildrenwithWilliamssyndrome.Researchin DevelopmentalDisabilities,32,164–169.
Annaz,D.,Karmiloff-Smith,A.,Johnson,M.H.,&Thomas,M.S.C.(2009).Across-syndromestudyofthedevelopmentofholisticfacerecognitioninchildrenwith autism,DownsyndromeandWilliamssyndrome.JournalofExperimentalChildPsychology,102(4),456–486.
Arens,R.,Wright,B.,Elliot,J.,Zhao,H.,Wang,P.P.,Brown,L.W.,&Kaplan,P.(1998).PeriodiclimbmovementinsleepinchildrenwithWilliamssyndrome.The JournalofPediatrics,133(5),670–674.
Armstrong,K.L.,Quinn,R.A.,&Dadds,M.R.(1994).Thesleeppatternsofnormalchildren.MedicalJournalofAustralia,161,202–206.
Ashworth,A.,Hill,C.M.,Karmiloff-Smith,A.,&Dimitriou,D.(2013a).CrosssyndromecomparisonofsleepproblemsinchildrenwithDownsyndromeand Williamssyndrome.ResearchinDevelopmentalDisabilities,34,1572–1580.
Ashworth,A.,Hill,C.M.,Karmiloff-Smith,A.,&Dimitriou,D.(2013b).Sleepenhanceslearninginchildren.JournalofChildExperimentalPsychology(inpress).
Banks,S.,&Dinges,D.F.(2007).Behavioralandphysiologicalconsequencesofsleeprestriction.JournalofClinicalSleepMedicine,3(5),519–528.
Bartlett,L.B.,Rooney,V.,&Spedding,S.(1985).Nocturnaldifficultiesinapopulationofmentallyhandicappedchildren.BritishJournalofMentalSubnormality,31, 54–59.
Bech,P.,Rasmussen,N.-A.,RaabækOlsen,R.,Noerholm,V.,&Abildgaard,W.(2001).ThesensitivityandspecificityoftheMajorDepressionInventoryusingthe PresentStateExaminationastheindexofdiagnosticvalidity.JournalofAffectiveDisorders,66,159–164.
Bo´dizs,R.,Gombos,F.,&Kova´cs,I.(2012).SleepEEGfingerprintsrevealacceleratedthalamocorticaloscillatorydynamicsinWilliamssyndrome.Researchin DevelopmentalDisabilities,33,153–164.
Buysse,D.J.,Reynolds,III,Monk,C.F.,Berman,T.H.,&Kupfer,S.R.D.J.(1989).ThePittsburghSleepQualityIndex:Anewinstrumentforpsychiatricpracticeand research.PsychiatryResearch,28,193–213.
Chervin,R.D.,Archbold,K.H.,Panahi,P.,&Pituch,K.J.(2001).Sleepproblemsseldomaddressedattwogeneralpediatricclinics.Pediatrics107.
Chervin,R.D.,Ruzicka,D.L.,Giordani,B.J.,Weatherly,R.A.,Dillon,J.E.,Hodges,E.K.,etal.(2006).Sleep-disorderedbreathing,behavior,andcognitioninchildren beforeandafteradenotonsillectomy.Pediatrics117.
Cohen-Zion,M.,&Ancoli-Israel,S.(2004).Sleepinchildrenwithattention-deficithyperactivitydisorder(ADHD):Areviewofnaturalisticandstimulant interventionstudies.SleepMedicineReviews,8,379–402.
Dahl,R.E.(1999).Theconsequencesofinsufficientsleepforadolescents:Linksbetweensleepandemotionalregulation.PhiDeltaKappan,80,354–359.
Dearing,E.,McCartney,K.,Marshall,N.L.,&Warner,R.M.(2001).Parentalreportsofchildren’ssleepandwakefulness:Longitudinalassociationswithcognitive andlanguageoutcomes.InfantBehavior&Development,24,151–170.
Dimitriou,D.,Karmiloff-Smith,A.,Ashworth,A.,&Hill,C.M.(2013).SleepandlearninginWilliamssyndrome.PediatricsResearchInternationalJournal(inpress).
Donnai,D.,&Karmiloff-Smith,A.(2000).Williamssyndrome:fromgenotypethroughtothecognitivephenotype.AmericanJournalofMedicalGenetics,97(2), 164–171.
Dykens,E.M.(2003).Anxiety,fearsandphobiasinpersonswithWilliamssyndrome.DevelopmentalNeuropsychology,23,291–316.
Einfeld,S.L.,Tonge,B.J.,&Florio,T.(1997).BehavioralandemotionaldisturbanceinindividualswithWilliamssyndrome.AmericanJournalonMentalRetardation, 102(1),45–53.
Fallone,G.,Acebo,C.,Seifer,R.,&Carskadon,M.A.(2005).Experimentalrestrictionofsleepopportunityinchildren:Effectsonteacherratings.Sleep,28(12), 1561–1567.
Fenson,L.,Dale,S.P.,Reznick,J.S.,Thal,D.,Bates,E.,Hartung,J.P.,etal.(1993).TheMacArthurCommunicativeDevelopmentInventory:User’sGuideandTechnical Manual.SanDiego:SingularPublishingCompany.
Freudigman,K.A.,&Thoman,E.B.(1993).Infantsleepduringthefirstpostnatalday:Anopportunityforassessmentofvulnerability.Pediatrics,92(3),373–379.
Goldman,S.E.,Malow,B.A.,Newman,K.D.,Roof,E.,&Dykens,E.M.(2009).SleeppatternsanddaytimesleepinessinadolescentsandyoungadultswithWilliams syndrome.JournalofIntellectualDisabilitiesResearch,53(2),182–188.
Gombos,F.,Bo´dizs,R.,&Kova´cs,I.(2011).AtypicalsleeparchitectureandalteredEEGspectrainWilliamssyndrome.JournalofIntellectualDisabilitiesResearch, 55(3),255–262.
Go´mez,R.L.,Bootzin,R.R.,&Nadel,L.(2006).Napspromoteabstractioninlanguage-learninginfants.PsychologicalScience,17,670–674.
