assert the autism symptom self report for adolescents and adults bifactor analysis and validation in a large adolescent population

Lundervoldb,d,e a Department of Child and Adolescent Psychiatry, Haukeland University Hospital, 5021 Bergen, Norway b Regional Centre for Child and Youth Mental Health and Child Welfare,

ASSERT – The Autism Symptom SElf-ReporT for adolescents

and adults: Bifactor analysis and validation in a large

Maj-Britt Posseruda,b,e,* , Kyrre Breivikb, Christopher Gillbergc,

Astri J Lundervoldb,d,e

a Department of Child and Adolescent Psychiatry, Haukeland University Hospital, 5021 Bergen, Norway

b

Regional Centre for Child and Youth Mental Health and Child Welfare, Uni Health, Uni Research, P.O Box 7800, 5020 Bergen, Norway

c

Gillberg Neuropsychiatry Centre, Institute of Neuroscience and Physiology, University of Gothenburg, 411 19 Go¨teborg, Sweden

d

Department of Biological and Medical Psychology, University of Bergen, P.O Box 7800, Bergen, Norway

e

K.G Jebsen Centre for Research on Neuropsychiatric Disorders, University of Bergen, P.O Box 7800, Bergen, Norway

1 Introduction

Theconceptofautismhasevolvedfromthedescriptionofseverecasesofinfantileautismaffectingabout0.02%(Kanner,

1943),tothemoderndayautismspectrumdisorder(ASD)encompassinganestimated1%ofthepopulation(Bairdetal., 2006;Brughaetal.,2011;Posserud,Lundervold,Lie,&Gillberg,2010).Needlesstosay,the‘‘1%ASD’’isnotthesameas‘‘0.02% infantileautism’’.ThemajoritywithASDfunctionsatnormalornexttonormallevelscognitively,and manyalsolead independentlivesinadultage.Thebroadeningoftheconceptandgrowingpublicawarenesshasledtoasituationwhere

A R T I C L E I N F O

Article history:

Received 26 July 2013

Received in revised form 17 September 2013

Accepted 19 September 2013

Available online 28 October 2013

Keywords:

Autism

ASD

Autism symptoms

Screen

Adolescents

Adults

Factor analysis

ASSERT

Self-report

A B S T R A C T Withaviewtodevelopingabriefscreeninginstrumentforautismsymptomsinageneral populationofadolescents,sevenitemsfromtheAspergersyndrome(andhigh-functioning autism)diagnosticinterviewwereadaptedforuseasself-reportinanonlinequestionnaire for youths aged 16–19 years (N=10,220) The selected items target lack of social understanding(4items)andrigidandrepetitivebehaviorandinterests(RRBI;3items) Factoranalyseswereperformed,andthesevenitemswerealsovalidatedagainst self-reportedASDdiagnosis.Beststatisticalmodelfitwasfoundforabifactormodelwithone generalfactorandtwodomainspecificfactorstiedtosocialdifficultiesandRRBI.Boththe generalandthedomainspecificfactorswereassociatedwithself-reportedASDdiagnoses Thescale(referredtoastheAutismSymptomSElf-ReporTforAdolescentsandAdults– ASSERT)hadgoodscreeningpropertieswithareceiveroperatingcurve-areaunderthe curve(ROC-AUC)of0.87andadiagnosticoddsratio(DOR)of15.8.Applyingamodified scoringofthescalefurtherimprovedthescreeningpropertiesleadingtoaROC-AUCof 0.89andaDORof24.9.TheASSERTholdspromiseasabriefself-reportscreenforautism symptomsinadolescents,andfurtherstudiesshouldexploreitsusefulnessforadults

ß2013TheAuthors.PublishedbyElsevierLtd.Allrightsreserved

§

This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution and reproduction in any medium, provided the original author and source are credited.

* Corresponding author at: PBU, Haukeland University Hospital, 5021 Bergen, Norway Tel.: +47 97641843.

E-mail addresses: maj-britt.posserud@uni.no (M.-B Posserud), kyrre.breivik@uni.no (K Breivik), christopher.gillberg@gnc.gu.se (C Gillberg),

astri.lundervold@psybp.uib.no (A.J Lundervold).

ContentslistsavailableatScienceDirect

0891-4222/$ – see front matter ß 2013 The Authors Published by Elsevier Ltd All rights reserved.

adults,whohavenotbeendiagnosedinchildhood,seekhelpfortheirproblemswith(Brughaetal.,2011)isolationand feelingsofinadequacy.AdultserviceshavenotyetdevelopedtomeettheneedsforadultswithASD,and fewsupport programsarein placethattargetthespecificneeds ofthoseindividuals(Howlin,Alcock,&Burkin,2005).Adultsmay thereforeaccessservicesthatareatlossaswhattodo,sometimesevenoutrightuncooperative,duetolackofknowledge,and

adearthofadequatetoolsandinterventionsforthisgroup.Theresearchcommunityandpublicservicesneedtoadapttothe newrealityofarelativelylargegroupofpeoplewithASD,orautismsymptomsthatperhapsdonotquitesurpassthelevel requiredforadisorderdiagnosis,who,withjustabitofsupportandadequateunderstanding,mightfunctionwellwiththeir socialdisability,butwho,ifnotproperlyunderstood,mightsuffergreatly

Giventhatautismhastraditionallybeenconceptualizedasachildhooddisorder,thereisalackofinstrumentstoscreen for,assessanddiagnoseautisminadults.Mostdiagnosticinterviewsareintendedforcompletion/interviewby/withaparent

orsomeoneelsewithintimatefirst-handknowledgeaboutthepersonaffected,includinginformationabouthis/herfirst yearsoflife.GiventhatadultswithsuspectedASDmaynotevenhavealivingparent,itmaybeverydifficulttoassessthe socialskillsbeforetheageofthree(diagnosticrequirementintheDSM-IV,butlessstringentlydefinedundertheDSM-5) (Diagnosticandstatisticalmanualofmentaldisorders:DSM-IV,2011).Whenitcomestoself-ratinginstruments,theautism quotient(AQ)anditsshorterversion(AQ-Short)aretheimportantexceptionstothelackofsuchinstruments(Baron-Cohen, Wheelwright,Skinner,Martin,&Clubley,2001;Hoekstraetal.,2011;Woodbury-Smith,Robinson,Wheelwright,& Baron-Cohen,2005) AlthoughtheAQ existsfor adolescents,this version is tobefilled in/completedby parents ofaffected individuals(Baron-Cohen,Hoekstra,Knickmeyer,&Wheelwright,2006).Tothebestofourknowledge,therearenoASD self-reportinstrumentsforadolescents.Mostadolescents,atleastintheory,dohaveanadulttoanswerforthem,butthereare instanceswhereanadultmaynotbeavailable,asitistypicallydifficulttogetholdofandbeabletocooperatewithpatients andparentstogetherinthelaterteenageyearsandyoungadulthood(Sanci,Sawyer,Kang,Haller,&Patton,2005).Infact,the largemajorityofadolescentsreportthatthelackofconfidentialhealthservicesimpedesthemfromseekinghelpfortheir problems(Thralletal.,2000)

Thegoalofthepresentstudywasthereforetoformulateandevaluateasetofself-reportitemsthatwouldvalidlycapture thelackofsocialunderstandingandrigidandrepetitivebehaviorandinterests(RRBI)thatsignalASDinadolescentsand youngadults(andthroughoutthelife-span).ItemsfromtheAspergersyndrome(andhigh-functioningautism)diagnostic interview(ASDI)(Gillberg,Rastam,&Wentz,2001)wereadaptedforthispurpose.AlthoughtheASDIisaninvestigator-rated interview,itemshadalreadybeenadaptedforself-reportandcomparedtotheparentalASDIinapreviousstudyofyoung adultsmaleswithAspergersyndrome(AS),showinggoodagreementon theseitemsacrossparentand patientratings (Cederlund,Hagberg,&Gillberg,2010).Wefurtheradaptedsevenitemscoveringsocialimpairment(4items)andRRBI(3 items)tofitourNorwegianpopulation-basedadolescentsurveyusinganonlinequestionnaireandrenamedthescaleAutism SymptomSElf-ReporTforAdolescentsandAdults(ASSERT)toreflecttheintendeduseoftheseitems.Theaimsofthecurrent studywastoinvestigatethepsychometricpropertiesoftheASSERTanditsusabilityasascreeninginstrumentforthe presenceofautisticsymptoms

PreviousstudieshavetendedtofindsupportforthefactthatASDconsistsoftwoormoredimensions/factorsthatareonly modestlycorrelatedwitheachother(Happe&Ronald,2008;Mandy&Skuse,2008;Shuster,Perry,Bebko,&Toplak,2013) Thesefindingshavecontributedtothe‘‘fractionableautism’’hypothesiswhereproponentsarguethattheASDdimensions arelargelyindependentofeachotherwithlargelyseparatecauses.TherelationshipbetweenASD,socialdifficultiesandRRBI mustbesaidtobeunclear(Mandy&Skuse,2008),butinspiteofthis,thetwodomainshavenowbeeninseparablylinkedto ASDintheDSM-5,asadiagnosisofASDcannotbemadewithouthavingRRBIsymptoms(McPartland,Reichow,&Volkmar,

2012).ManyfactoranalyseshavebeenperformedonASDsymptoms,buttoourknowledge,abifactormodelhasnotbeen applied(Shusteretal.,2013).Apotentialadvantagewiththebifactormodeloverthecorrelatedfactormodel(wherethe dimensionsaretreatedascorrelatedbutseparate)isthatitprovidesinformationaboutwhatalloftheitemshaveincommon

aswellasuniquesymptomdimensions.Ittherebyprovidesarationalstructureexplainingbothoverlapandseparability betweendimensionsinamodel,andcouldthusbeusefultoexplorethecontradictoryfindingsregardingASD,RRBIand sociability.Tothisaimandtoexaminethepsychometricpropertiesofthescale,weappliedbothaconventionalexploratory factoranalysis(EFA)andaconfirmatoryfactoranalysis(CFA)usingabifactormodel.Wewerealsointerestedinwhetherthe generalfactorpredictedself-reportedASDdiagnosisupandabovewhatwaspredictedfromtheuniquevariancetiedtothe subdomains(controlledforthegeneralfactor)

2 Materialandmethods

2.1 Populationsample

ThebackdropofthestudywasthefourthwaveofthelongitudinalBergenChildStudy(BCS).Inthiswave,theoriginal targetpopulationwasextendedtoincludeallyoungpeoplebornin1993–1995(age16–19yearsatthetimeofthestudy) residinginthecountyofHordalandwhereBergencityissituated(N=19121).Thiscross-sectionalstudyofalargergroupof adolescentswasnamed‘‘ung@hordaland’’(young@hordaland).Thecollectionofdatawasperformedinthespringof2012, and10,220youngpeopleparticipated(withacorrespondingresponserateof54%).Allyouthswereinvitedtoparticipate,but thegreatmajorityofresponsescamefromadolescentsattendingschools(97.8%),whereaschoollessonwassetasideto allowforthecompletionoftheonlinequestionnaire(bothprivateandpublicschools)

2.2 Instruments

Theung@hordalandquestionnairewasdevelopedspecificallyforthisstudywithaviewtocoveringawiderangeof mentalhealthproblemsandassociatedissues.Toscreenforautismsymptoms,sevenitemsfromtheAspergersyndrome (andhigh-functioningautism)diagnosticinterview(ASDI)(Gillbergetal.,2001)wereadaptedtogetherwiththemain developerofthatinstrument,ChristopherGillberg.TheASDIisasemi-structuredinvestigator-baseddiagnosticinterview including20items,andhasbeenusedinpreviousstudiesasavalidandreliabletooltodiagnoseAsperger/high-functioning autisminadults(Gillbergetal.,2001).Someoftheitemsareratedbytheinvestigatoraccordingtoobservedbehaviorduring theinterview.TheitemsfromtheASDIthatarenotinvestigator-ratedwereadaptedforanearlierstudyofyoungadultmales withAspergersyndrome(AS)inwhichtheywereusedasself-reportitemsandcomparedtoparentalreportsonthesame items(Cederlundetal.,2010).Theauthorsfoundthatmanywerequiteawareoftheirowndifficultiesinsomeareas,and arguedforincreasedconsiderationofthepatient’sownreportinthediagnosticwork-upandinterventionplanningfor patientswithASD.ThesamesevenitemsweretranslatedintoNorwegianandadaptedforuseintheonlineself-report questionnaireforadolescents16–19yearsofage.Fouritemstargetingsocialsymptoms(items1–4intheASDI)andthree itemstargetingrigidandrepetitivebehaviorandinterests(RRBI;items5,8and9intheASDI)wereincluded(Table1

Responseoptionswere‘‘nottrue’’(score0)–‘‘somewhattrue’’(score1)–‘‘certainlytrue’’(score2),leadingtoascorerange

of0–14p.ontheASSERT

Theadolescentswerealsoaskedtoreportonthepresenceofpsychiatricdiagnoses:‘‘Haveyoubeendiagnosedwithany mentalhealthproblems?(e.g.ADHD,anxiety,depression,autism)’’

2.3 Statisticalanalyses

ReportsmissingmorethanoneofthesevenASDIitemswerenotincludedintheanalyses(N=228).Receiveroperating curve(ROC)analyseswereperformedusingallsevenitemscombinedintoonescale(ASSERT)withself-reportedASDasstate variable.Descriptiveanalyses,Cronbach’salpha(a)andROCanalyseswereperformedusingIBMSPSSStatistics19.Mplus version 6.0 wasusedforother correlationanalysesandfactor analyses(Muthe´n &Muthe´n,1998–2012) The robust-weightedleastsquareestimator(WLSMV)wasusedinthefactoranalysesbecauseoftheskewedcategoricaldata(ordinal datawiththreeoptions).Usingpolychoriccorrelationsforestimation,theWLSMVseemsrelativelyrobusttoviolationsof normality(Dumenci&Achenbach,2008;Flora&Curran,2004).Thechi-squarevalueisnotreportedasmeasureofmodelfit

asthisisnotexactwhenusingtheWLMSVestimator.Therefore,weusedBentler’scomparativefitindex(CFI;Bentler,1990), Tucker–Lewisindex(TLI;Tucker&Lewis,1973)andtheroot-mean-squareerrorofapproximation(RMSEA;Steiger&Lind,

1980)withcut-off values for CFI0.96, TLI0.95and RMSEA0.05 toindicate goodnessof fit (Yu,2002) EFA was performedwithgeominrotation(defaultobliquerotationinMplus).MissingdataononeASSERTitemwasreplacedwiththe meanoftheremainingsixitemsandincludedintheROCanalysesand correlationanalysesoftheentirescale Inthe remaininganalyses,missingvaluesweretreatedwithpairwisedeletionfortheanalysesperformedinMplus(default)and withlistwisedeletionforanalysesperformedinSPSS(default)

3 Results

3.1 Responses

Themeanscorefortheentirescalewas2.60(SD2.22,N=9992).DistributionofresponsesisshowninFig.1anditem responsefrequenciesinTable2

MostindividualshadverylowscoresontheASSERT,and55%scored2points.Forty-fiveindividualsreportedhaving been diagnosed with an ASD (11 autism, 29 Asperger syndrome, 4 atypical autism/PDD-NOS, 1 possible autism), correspondingtoaprevalenceof0.45%self-reporteddiagnosedASD

Table 1

The seven self-report items of the Autism Symptom SElf-ReporT (ASSERT) adapted from the Asperger syndrome and high functioning autism diagnostic interview (ASDI).

Item abbreviation and translated content of original Norwegian item

S1 Do you find it difficult to socialize with, or to get in touch with people, especially people your own age?

S2 Do you prefer to be alone rather than being together with other people?

S3 Do you have difficulties perceiving social cues?

S4 Do other people tell you that your behavior or your emotional responses are inappropriate or hurtful?

R1 Do you have a strong interest or hobby that absorbs so much of your time that it hampers other activities?

R2 Do you or do other people feel that you have very set routines or that you are very immersed in your own interests? R3 Do you or do other people feel that you impose your routines or interests on others?

3.2 Factoranalyses

Thethree-factorEFAsolutionshowedexcellentfit(CFI=1.00,TLI=1.00,RMSEA=0.00)forasolutionincludinga one-itemfactor(Table3 Thetwo-factormodelalmostmetpretestcriteria(CFI=0.98,TLI=0.94,RMSEA=0.06)andhadhigher itemloadings(Table2)whiletheone-factormodelwasdefinitelydiscarded(CFI=0.67,TLI=0.51andRMSEA=0.18).Table3

showstheitemloadingsforallthreeEFAfactormodels.Thecorrelationbetweenthefirst(social)andsecond(RRBI)factorin thetwo-factormodelwasr=0.23

Thebifactormodelwithonegeneralfactorandtwosubdomains(socialandRRBI)showedverygoodstatisticalmodelfit withCFI=0.996,TLI=0.987andRMSEA=0.030.ThemodelwithitemloadingsisshowninFig.2.Evenifallitemloadings weresignificant(p<0.001,exceptS4loadingontothesocialsubdomain,withp=0.019),severaloftheloadingswererather weak(<0.50).Theitemsassessingdifficultysocializingwithpeople,stronginterestorhobbiesandhavingverysetroutines allhadloadingslowerthan0.40onthegeneralfactor.Thefactorswerecorrelatedwiththeself-reporteddiagnosisofASDto

0 200 400 600 800 1000 1200 1400 1600 1800 2000

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

N

ASS ERT sc ore

Fig 1 Distribution of ASSERT scores in adolescents 16–19 years old (N = 9992).

Table 2

Response frequencies for each ASSERT item in percentages and numbers.

Table 3

Exploratory factor analysis – item loadings for the one-factor (F1), two-factor (F2) and the three factor (F3) solutions Loadings above 0.45 are high-lighted (signaling 20% overlapping variance).

First factor Second factor First factor Second factor Third factor

examinetheirrelevanceforadiagnosisofASD.Boththegeneralfactorandthesocialfactorhadmoderatecorrelationswith theASDdiagnosis(0.47and0.39respectively),indicatingthattheybothcontributeuniquelytopredictingthepresenceofan ASD,whiletheRRBIfactoronlycorrelated0.20withself-reporteddiagnosisofASD

3.3 ASSERTscaleproperties

Cronbach’salpha(a)fortheentirescalewas0.62,probablyreflectingthemultidimensionalityofthescale.Cronbach’s alphawas0.63forboth thesocial subscale(ASSERTsoc)and therepetitiveand stereotypebehavior/interestssubscale (ASSERTrrbi),partlyanindicationoftheratherfewitemsincludedinthesubscales(4and3respectively)

3.4 ValidityofASSERT

TheASSERTshowedgoodscreeningpropertiesversusself-reporteddiagnosisofASD,withanareaunderthecurve(AUC)

intheROC-analysis(receiveroperatingcurve)of0.87(95%CI0.83–0.92)(Fig.3andTable4 Scoring5pointsontheASSERT hadasensitivityof0.80andspecificityof0.81forself-reportedASD.Nineadolescentswithself-reportedASDscored<5,but onlyoneoftheseadolescentsscoredexclusivelyontheASSERTrrbiitems,whereas33%oftheadolescentswithout self-reporteddiagnosisofASDandascoreof2–4endorsedexclusivelyonthoseitems.Thecorrelationpatternbetweenthe factorsinthebifactorsolutionandself-reporteddiagnosisofASDindicatedthattheASSERTrrbiitemsareimportantmainly

inconjunctionwiththesocialitems.Basedontheseresultswedecidedtoexploretheeffectofdifferentialweightingofthe itemsintheASSERT,bydoublingthescoreofthesocialitemswhilemaintainingthescoringoftheASSERTrrbiitems.This improvedthescreeningpropertiesfurtherleadingtoanAUCof0.89(95%CI0.84–0.93),sensitivityof0.80andspecificityof 0.86forascoreof8(Table4andFig.3 TheROCanalyseswerererunwiththeyouthswhohadresponded‘‘yes’’totheitem

‘‘Haveyoubeendiagnosedwithanymentalhealthproblems?(e.g.ADHD,anxiety,depression,autism)’’(N=724)toexamine thediscriminativepoweroftheASSERTforASDvs.othermentalhealthdisorders.AUCforthisgroupwas0.80(95%CI0.74– 0.86),andtheROCindicatedascoreof8(modifiedscoring)tobeoptimalalsointhisgroup

3.5 ScreeningutilityofASSERT

TofurtherevaluatetheutilityofASSERTasdiagnostictestwecalculatedthediagnosticoddsratio(DOR).Thismeasureis moreinformativethanthepredictivevalueofatest,whichisinfluencedbythebaseprevalenceinthesampleinvestigated

difficultsocialize

prefers being alone

perceiving social cues

inappropriate behav

setroutines/immersed strong interest

imposing routines

Repetitive interests

General factor

Social difficulties

0.74

0.82

0.40

0.89

0.42

0.08 0.45

0.33

0.46

0.62

0.67

0.47 0.22

0.28

0.55

Self-reported ASD

0.39

0.20

Fig 2 Bifactor model with item loadings and correlations between factors and self-reported ASD.

TheDORisameasureofadiagnostictest’soverallaccuracy,andisgeneratedbydividingthenumberofcorrectlyclassifiedby thenumberofincorrectlyclassifiedindividuals(Glas,Lijmer,Prins,Bonsel,&Bossuyt,2003;Haynes,2006).Inotherwords,it tellsyouhowgoodthe‘‘sieve’’youareusingactuallyis.AnotheradvantageofreportingtheDORisthatitfacilitatesthe comparisonsoftestsformeta-analyses(Glasetal.,2003).ADORvalueof20ormoreindicatesthataninstrumenthasuseful screeningproperties(Fischer,Bachmann,&Jaeschke,2003).Usingthecut-offof5ontheASSERTproducesaDORof15.8, whileusingthecut-offof8onthemodifiedASSERTscoregivesaDORof24.9.Thisclearlyshowsthatalthoughthechange

inspecificity ofusingthemodified ASSERTmayseem modest,it wasimprovedwithintact sensitivity,thereby much increasingthecorrectclassificationrateofthescale

4 Discussion

ThepresentstudyshowedthatASSERTseemstoworkwellasaself-reportscreenforautisticsymptomsinthislarge population-basedsampleofadolescentsage16–19yearsold.Theitemsseemedtobereadilyunderstood,witharesponse rateof98.8%.Furthermore,theAUCfortheROCanalysisusingself-reporteddiagnosisofASDasoutcomewasashighas0.89, witha sensitivityof 0.80and aspecificity of0.86forscores8,supportingthevalidityof theASSERTas ascreening instrumentofASDsymptomsinadolescentsandyoungadults.Usingself-reportedASDdiagnosisforvalidationofASSERT wasnotideal,buttheoverallrateof0.45%ofself-reportedASDisalmostidenticaltooneofthepreviouslyreportedestimates fromtheBergenChildStudy(BCS)of0.44%basedontheDAWBA(developmentandwell-beingassessment)(Heiervangetal.,

2007),andself-reportshaveprovenvalidforarangeofmentalhealthdisorders(Halmoyetal.,2010).Asthepresentfigureis basedonself-reportsfrommainlyhigh-schoolstudents,weassumethatthefigureof0.45%onlyincludeshigh-functioning individualswithanASD,andthattherateofadolescentswithanyASDshouldbehigher(Brughaetal.,2012)

Fig 3 Receiver operating curves for ASSERT and modified ASSERT versus self-reported presence of ASD (N = 9992).

Table 4

Sensitivity and specificity for the ASSERT scores.

ASDsareconceptualizedintheDSM-5ascontainingtwocoregroupsofsymptoms,viz.socialimpairmentandRRBI.Many factoranalyseshavebeenperformedinthefieldofautism,bothinclinicalsamplesandlargerpopulation-basedsamples (Shusteretal.,2013),butnopreviousstudy(toourknowledge)hasappliedabifactormodel,inspiteofresearchindicating bothseparabilityofthedomainsinASD(geneticallyandinthepopulation)andunityofdomains(theASDs).Thebifactor modelhadgoodstatisticalfit,confirmingthatthesocialandRRBIdimensionsbothshareandhaveuniquevariancetiedto them.Despitethemodestcorrelation(r=0.23)betweenthetwosubdimensions(socialandRRBI)whenmodeledastwo correlatedfactors,thegeneralfactor(reflectingthesharedvariance)predictedself-reportedASDdiagnosis(r=0.47)over andabovewhatwaspredictedfromthetwodomainspecificfactors.ThisunderscorestheimportanceofnottreatingtheASD dimensionsastotallyseparatedimensionseveniftheyarenotstronglycorrelated.Atraditionalcorrelatedfactormodel wouldonlyhaveexploredhowwelleachofthetwosubdimensionswouldhavepredictedself-reportedASDdiagnosisby itselforcontrolledforeachother(theiruniqueprediction).Thisinformationisalsogivenbythebifactormodelshowingthat thedomainspecificsocialdifficultiesfactorhadastrongerassociationwithself-reportedASDdiagnosis(r=0.39)thanthe RRBIdomainspecificfactor(0.20).However,thebifactormodelfurtherindicatesthatthesharedvariancebetweenthetwo ASD subdimensions seemsto be equallyimportant as the unique contributionfrom social difficulties dimensionsin predictingself-reportedASDdiagnosis

WearehesitanttoregardthegeneralfactorasrepresentingtheoverarchingASDconcept.Thebifactormodelhasoften beenappliedwhenrepresentingmultidimensionalconstructssuchasADHDandintelligencewhereitishypothesizedtobea dominant/overarching general factor in addition to some smaller domain specific factors (Ullebo, Breivik, Gillberg, Lundervold,&Posserud,2012) Thepresentstudygivessomesupporttothefractionable autismhypothesisas sucha dominantfactordoesnotseemtoexistregardingASDinthisgeneralpopulationsample.Ourgeneralfactorwasratherweak withlowloadingsofseveralitems,includingcoreASDsymptoms,suchashavingdifficultiessocializingwithotherpeople Thethreeitemswithstrongestloadingsontothegeneralfactorsuggestitcouldrepresenttheoryofmind(ToM)related difficulties more specifically(difficulties perceivingsocial cues (S3), and other peoples responsesto therespondents behavior(S4,R3)).Moreresearchisneededtobringconceptualclaritytowhatthegeneralfactorrepresentsandwhetherthe ratherweakloadingscouldbeduetoapplyingthemodelofanarrowdisordertoageneralpopulation.Themodelshouldbe replicatedinotherpopulationsandusingotherASDinstrumentsbeforeanystrongconclusionscanbedrawn

AlthoughtherelationshipbetweenASD,socialdifficultiesandRRBIisunclearandinsufficientlyexplored(Mandy&Skuse,

2008),intheDSM-5,theASDdiagnosiscannotbemadeintheabsenceofRRBI(McPartlandetal.,2012).Simultaneously,a disorderofsocialcommunicationisintroduced.Thischangehasraisedcriticismandconcernamongresearchers,clinicians andparents,assomechildrenwithaclearsocialhandicapbutwithoutRRBIwillnolongerbelongwithintheASDcategory, withimplicationsforresearchandtreatment.SomestudieshaveshownthatanumberofchildrenwithASDaccordingtothe DSM-IVcriteria,donotmeetDSM-5criteriaforASD(Mayes,Black,&Tierney,2013).Someofthemwillmeetcriteriaforthe newsocialcommunicationdisorder.ItisuncertainwhatthedifferencebetweenASDandthenewsocialcommunication disorderis, apartfromnot includingRRBI(Skuse,2012; Tanguay, 2011) The correlationbetween theASSERTsocand ASSERTrrbiwaslow(r=0.23)butidenticaltootherpopulation-basedstudies(Ronald,Happe,&Plomin,2005;Ronaldetal.,

2006),suggestingalowdegreeofoverlapbetweenthetwofactors.ThehighercorrelationwithanASDdiagnosisofthe generalfactorandthesocialfactormayindicatethattheoverlapbetweenthesocialandtheRRBIdomainsmayberather specifictoASDproblems.However,thespecificsocialfactorstillshowedamoderatecorrelation(r=0.39)withthe self-reportedASDdiagnosis,afteraccountingforthegeneralfactor.ThissuggeststhatanASDdiagnosismaybejustifiedalsoin theabsenceofclearRRBIsymptoms.Ifonebelongstothe‘‘splitter’’-advocateswithinthediagnosticsystem(ratherthan

‘‘lumpers’’)(McKusick,1969),onecouldalsoarguethatitsupportstheuseofaseparatedisorderincludingonlysocial difficulties,butthenatleastinthisstudytheyarenotdistinguishableatthesymptomlevel,otherthannotincludingRRBI symptoms

TheDSM-5hasreplacedtheseparateASDswithonecategory.Inlinewiththisframeworkshift,theself-reportedautism diagnosesinthepresentstudywerecombinedanddichotomizedintopresenceorabsenceofASD.TheDSM-5furthermore indicatesthathavingASDanditsseverityissignaledbythepresenceanddegreeofimpairment.Thisintroducesacategorical elementwhichissituationalandnot(necessarily)directlyrelatedtotheunderlyingtrait.Similartootherlarge population-basedstudies(Constantinoetal.,2003;Posserud,Lundervold,&Gillberg,2006)thedistributionofautismsymptomsinthis sampleofadolescentsshowsagradualshiftfromnosymptomsuptohighsymptoms,supportingadimensionaldistribution However,thegoodfitofthebifactormodelincombinationwithaweakgeneralfactorcouldmeanthatthereareseparate populationswithinthepopulationfollowingdifferentdistributions.Thedimensionalmodelmaythusbeinsufficientin explaining the structure and nature of ASD symptoms, and models allowing for both categorical and dimensional distributionscouldbemoreadequate

Althoughthetargetsampleinthisstudywas16–19yearsofage,thesameitemscouldmostlikelysuccessfullybeused acrosstheadultagespan,asindicatedbyapreviousstudyshowinggoodpropertiesofthesameitemsinyoungadults (Cederlundetal.,2010).Aself-reportinstrumentforASDvalidatedfromtheageof16isimportantasyouthsatthatagecan accesshealthservicesindependentlyanddenyparentscontactwiththesameservices.Althoughmostadolescentsdohavea livingparentwhocouldfilloutaquestionnaireforthem,itisnotalwayseasytoachieveasmanyadolescentsdonotdesire parentalinvolvement(Thralletal.,2000)

Morethan50%respondedpositivelytotheitem‘‘Doyouhaveastronginterestorhobbythatabsorbssomuchofyour timethatithampersotheractivities?’’indicatingpoordiscriminatorypropertiesforthisitemversusASD.Whenformulating

thisitemweconsideredotherwordingincluding‘‘narrow’’or‘‘circumscribed’’interesttoconveytheautisticnatureofthe interest.Whileitmighthaveimproveditsdiscriminatoryproperties,itmadetheitemdifficulttounderstand.Asautistic interestsarenotnecessarilycircumscribed,andtheborderlinebetweentheintensityofanautisticinterestandanintense normalinterestmaybehardtodefine,weoptedforreadabilityratherthanhighspecificity.Anintenseinterestintheabsence

ofsocialdifficultiesisnottoberegardedasanautisticsymptom,andalowASSERTscorewhereallscoresstemfromR1toR3 canprobablybedisregarded.Thebifactormodelsolutionsupportsthis,asRRBIwhenincludedinthegeneralfactorwas relatedtoanASD,butafteraccountingforthegeneralfactoronlycorrelated0.20withreportedASD.Furthermore,modifying thescoringoftheASSERTbyweighingthesocialitemsmorethanRRBIitemsalsoimprovedspecificitywithintactsensitivity forASD

4.1 Limitations

Thestudyreliedexclusivelyontheself-reported diagnosisofASD.Whilenot likelytoproducefalse positives,this procedureverylikelyproducesfalsenegativeanswers,e.g.fromindividualswhoindeedhaveanASDbutdonotreportitor whohavenotyetbeendiagnosed.Undiagnosedadolescentsmayalsobelessawareoftheirownproblems(thusscoringlow

ontheASSERT).Thepresentprocedurecould,intheory,thereforebothunderestimateandinflatethepropertiesofthescale FuturestudiesshouldincludeclinicalassessmentofASDtoevaluatethescreeningpropertiesofASSERT

Thefewitemsincludedmayhaveincreasedthelikelihoodofspuriousordistortedfindingsinthefactoranalyses.Thisalso appliestothescreeningpropertiesofthescale;moreitemscouldhaveincreasedthescreeningpropertiesandincreasedits usefulnessasaninformativemeasureofASDsymptomsaswell.However,brevitycanalsobeofmerit,especiallyregarding adolescentsandindividualswithcognitiveormentalhealthproblems,wherestayingfocusedonataskatlengthisoftena problem.However,furtherdevelopmentofthescaleincludingitemstargetingforinstancesocialcommunicationcouldbe advantageous

Theprevalenceof0.45%forASDisprobablyanunderestimateonlyincludinghigherfunctioningindividualswithASD,but thisisalsothemostprevalentkindofASDandthegroupwhereself-reportprobablywouldbemostuseful

5 Summary

Thepresentstudypresentsandvalidatestheuseofaself-reportscreenforautismsymptomsinadolescents–theAutism Symptom SElf-ReporT (ASSERT) The scale is briefand easily comprehended,and seemsto bevalid and useful with discriminationforthepresenceofanASDinboththegeneralpopulationofadolescentsandamongmoreclinicallyaffected adolescents.StudiesincludingclinicalassessmentareneededtodetermineitspotentialasascreenforASD.Abifactormodel

ofthescalesupportedageneralfactorandtwoindependentfactorsofsocialdifficultiesandRRBI,butthegeneralfactorwas weak.ThemodelholdssomepromiseindescribingthestructureofASDtraitsinthegeneralpopulationbutitandother modelsshouldbeexploredfurther

Acknowledgements

WethanktheBergenChildStudyresearchgroupthathasworkedtogetherformanyyears,andwhosejointefforthas madethestudypossible.AspecialthankstoKjellMortenStormarkandMariHysingintheBergenChildStudyleaderteam,to TormodBøeandHildeSackariassenfordatabaseworkandadministrativeeffort.Wearegratefulforthesupportofthe RegionalCentreforChildandYouthMentalHealthandChildWelfare,UniHealth,forhostingthestudyforalltheseyears.We alsothanktheHordalandCountyCouncilforcollaboratinginthestudyandallowingustoperformthestudyincollaboration withtheschools.WearegratefultoAnnaSpyrouforproof-readingthefinalmanuscript.Inparticular,wewouldliketothank alltheadolescents,parentsandteachersforparticipatinginstudy

References

Baird, G., Simonoff, E., Pickles, A., Chandler, S., Loucas, T., Meldrum, D., et al (2006) Prevalence of disorders of the autism spectrum in a population cohort of children in South Thames: The special needs and autism project (SNAP) Lancet, 368, 210–215.

Baron-Cohen, S., Hoekstra, R A., Knickmeyer, R., & Wheelwright, S (2006) The autism-spectrum quotient (AQ) – adolescent version Journal of Autism and Developmental Disorders, 36, 343–350.

Baron-Cohen, S., Wheelwright, S., Skinner, R., Martin, J., & Clubley, E (2001) The autism-spectrum quotient (AQ): Evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians Journal of Autism and Developmental Disorders, 31, 5–17.

Bentler, P M (1990) Comparative fit indexes in structural models Psychological Bulletin, 107, 238–246.

Brugha, T., Cooper, S., McManus, S., Purdon, S., Smith, J., Scott, F., et al (2012) Estimating the prevalence of autism spectrum conditions in adults: Extending the 2007 adult psychiatric morbidity survey England: The NHS Health and Social Care Information Centre.

Brugha, T S., McManus, S., Bankart, J., Scott, F., Purdon, S., Smith, J., et al (2011) Epidemiology of autism spectrum disorders in adults in the community in England Archives of General Psychiatry, 68, 459.

Cederlund, M., Hagberg, B., & Gillberg, C (2010) Asperger syndrome in adolescent and young adult males Interview, self- and parent assessment of social, emotional, and cognitive problems Research in Developmental Disabilities, 31, 287–298.

Constantino, J N., Davis, S A., Todd, R D., Schindler, M K., Gross, M M., Brophy, S L., et al (2003) Validation of a brief quantitative measure of autistic traits: Comparison of the social responsiveness scale with the autism diagnostic interview-revised Journal of Autism and Developmental Disorders, 33, 427–433.

Dumenci, L., & Achenbach, T M (2008) Effects of estimation methods on making trait-level inferences from ordered categorical items for assessing psychopathology Psychological Assessment, 20, 55–62.

Fischer, J E., Bachmann, L M., & Jaeschke, R (2003) A readers’ guide to the interpretation of diagnostic test properties: Clinical example of sepsis Intensive Care Medicine, 29, 1043–1051.

Flora, D B., & Curran, P J (2004) An empirical evaluation of alternative methods of estimation for confirmatory factor analysis with ordinal data Psychological Methods, 9, 466–491.

Gillberg, C., Rastam, M., & Wentz, E (2001) The Asperger syndrome (and high-functioning autism) diagnostic interview (ASDI): A preliminary study of a new structured clinical interview Autism: The International Journal of Research and Practice, 5, 57–66.

Glas, A S., Lijmer, J G., Prins, M H., Bonsel, G J., & Bossuyt, P M M (2003) The diagnostic odds ratio: A single indicator of test performance Journal of Clinical Epidemiology, 56, 1129–1135.

Halmoy, A., Halleland, H., Dramsdahl, M., Bergsholm, P., Fasmer, O B., & Haavik, J (2010) Bipolar symptoms in adult attention-deficit/hyperactivity disorder: A cross-sectional study of 510 clinically diagnosed patients and 417 population-based controls Journal of Clinical Psychiatry, 71, 48–57.

Happe, F., & Ronald, A (2008) The fractionable autism triad: A review of evidence from behavioural, genetic, cognitive and neural research Neuropsychology Review, 18, 287–304.

Haynes, R B (2006) Clinical epidemiology: How to do clinical practice research Philadelphia: Lippincott Williams & Wilkins.

Heiervang, E., Stormark, K M., Lundervold, A J., Heimann, M., Goodman, R., Posserud, M B., et al (2007) Psychiatric disorders in Norwegian 8- to 10-year-olds: An epidemiological survey of prevalence, risk factors, and service use Journal of the American Academy of Child and Adolescent Psychiatry, 46, 438–447.

Hoekstra, R A., Vinkhuyzen, A A., Wheelwright, S., Bartels, M., Boomsma, D I., Baron-Cohen, S., et al (2011) The construction and validation of an abridged version of the autism-spectrum quotient (AQ-Short) Journal of Autism and Developmental Disorders, 41, 589–596.

Howlin, P., Alcock, J., & Burkin, C (2005) An 8 year follow-up of a specialist supported employment service for high-ability adults with autism or Asperger syndrome Autism: The International Journal of Research and Practice, 9, 533–549.

Kanner, L (1943) Autistic disturbances of affective contact Nervous Child, 2, 217–250.

Mandy, W P., & Skuse, D H (2008) Research review: What is the association between the social-communication element of autism and repetitive interests, behaviours and activities? Journal of Child Psychology and Psychiatry, and Allied Disciplines, 49, 795–808.

Mayes, S D., Black, A., & Tierney, C D (2013) DSM-5 under-identifies PDDNOS: Diagnostic agreement between the DSM-5, DSM-IV, and checklist for autism spectrum disorder Research in Autism Spectrum Disorders, 7, 298–306.

McKusick, V A (1969) On lumpers and splitters, or the nosology of genetic disease Perspectives in Biology and Medicine, 12, 298–312.

McPartland, J C., Reichow, B., & Volkmar, F R (2012) Sensitivity and specificity of proposed DSM-5 diagnostic criteria for autism spectrum disorder Journal of the American Academy of Child and Adolescent Psychiatry, 51, 368–383.

Muthe´n, L K., & Muthe´n, B O (1998–2012) Mplus user’s guide (6th ed.) Los Angeles, California: Muthe´n & Muthe´n.

Posserud, M., Lundervold, A J., Lie, S A., & Gillberg, C (2010) The prevalence of autism spectrum disorders: Impact of diagnostic instrument and non-response bias Social Psychiatry and Psychiatric Epidemiology, 45, 319–327.

Posserud, M B., Lundervold, A J., & Gillberg, C (2006) Autistic features in a total population of 7–9-year-old children assessed by the ASSQ (autism spectrum screening questionnaire) Journal of Child Psychology and Psychiatry, and Allied Disciplines, 47, 167–175.

Ronald, A., Happe, F., Bolton, P., Butcher, L M., Price, T S., Wheelwright, S., et al (2006) Genetic heterogeneity between the three components of the autism spectrum: A twin study Journal of the American Academy of Child and Adolescent Psychiatry, 45, 691–699.

Ronald, A., Happe, F., & Plomin, R (2005) The genetic relationship between individual differences in social and nonsocial behaviours characteristic of autism Developmental Science, 8, 444–458.

Sanci, L A., Sawyer, S M., Kang, M S., Haller, D M., & Patton, G C (2005) Confidential health care for adolescents: Reconciling clinical evidence with family values The Medical Journal of Australia, 183, 410–414.

Shuster, J., Perry, A., Bebko, J., & Toplak, M E (2013) Review of factor analytic studies examining symptoms of autism spectrum disorders Journal of Autism and Developmental Disorders, 1–21.

Skuse, D H (2012) DSM-5’s conceptualization of autistic disorders Journal of the American Academy of Child and Adolescent Psychiatry, 51, 344–346.

Steiger, J H., & Lind, J C (1980) Statistically-based tests for the number of common factors In Spring Meeting of the Psychometric Society.

Tanguay, P E (2011) Autism in DSM-5 The American Journal of Psychiatry, 168, 1142–1144.

Thrall, J S., McCloskey, L., Ettner, S L., Rothman, E., Tighe, J E., & Emans, S J (2000) Confidentiality and adolescents’ use of providers for health information and for pelvic examinations Archives of Pediatric and Adolescent Medicine, 154, 885–892.

Tucker, L R., & Lewis, C (1973) Reliability coefficient for maximum likelihood factor-analysis Psychometrika, 38, 1–10.

Ullebo, A K., Breivik, K., Gillberg, C., Lundervold, A J., & Posserud, M B (2012) The factor structure of ADHD in a general population of primary school children Journal of Child Psychology and Psychiatry, and Allied Disciplines, 53, 927–936.

Woodbury-Smith, M R., Robinson, J., Wheelwright, S., & Baron-Cohen, S (2005) Screening adults for Asperger syndrome using the AQ: A preliminary study of its diagnostic validity in clinical practice Journal of Autism and Developmental Disorders, 35, 331–335.

Yu, C Y (2002) Evaluating cut-off criteria of model fit indices for latent variable models with binary and continuous outcomes University of California (Unpublished Doctoral).