Disparity in autism spectrum disorder prevalence among Taiwan National Health Insurance

Disparity in autism spectrum disorder prevalence among Taiwan

National Health Insurance enrollees: Age, gender and urbanization effects

Shang-Wei Hsu

, Po-Huang Chiang

, Lam-Ping Lin

, Jin-Ding Lin

*

aGraduateInstituteofHealthcareAdministration,AsiaUniversity,Taichung,Taiwan

bDepartmentofPublicHealth,ChinaMedicalUniversity,Taichung,Taiwan

cInstituteofPopulationHealthSciences,NationalHealthResearchInstitutes,Miaoli,Taiwan

dSchoolofPublicHealth,NationalDefenseMedicalCenter,Taipei,Taiwan

eChung-HuaFoundationforPersonswithIntellectualDisabilities,NewTaipeiCity,Taiwan

fCenterforEnvironmentandPopulationHealth,GriffithUniversity,Brisbane,Australia

1. Introduction

Theautismspectrumdisorders(ASDs)areasetofdevelopmentaldisabilitiescharacterizedbyatypicaldevelopmentin socialization,communication,andbehavior(AmericanPsychiatricAssociation,2000;CDC,2009;Matson,2007).ASDisnota diseasebutasyndromewithmultiplecauses,bothgeneticandnon-genetic.Thenumberofgenesinvolved,aswellastheir identity,remainunknown(Muhle,Trentacoste,&Rapin,2004).Manystudieshavediscussedthemethodologicalissues involvedinthedifferentialdiagnosisofASD(Matson,Gonzalez,&Wilkins,2009;Matson,Gonzalez,Wilkins,&Rivet,2008;

Matson,Nebel-Schwalm,&Matson,2007;Woolfenden,Sarkozy,Ridley,&Williams,2011), andvariousrevisions have recentlybeenproposedtothediagnosticcriteriaintheDSM-V.However,furthermodificationisneededinvariousareas, includingsocial imaginationanddiagnosisininfancyandadulthood.Moreover,autisminfemales maycontinuetobe misdiagnosedorunrecognizedunderthenewcriteria(Wing,Gould,&Gillberg,2011).

ARTICLE INFO

Articlehistory:

Received11September2011 Accepted13September2011

Keywords:

Autismspectrumdisorders Prevalence

Gender Age Urbanization

ABSTRACT

Thepresentstudy aimstocharacterize theprevalence ofautismspectrumdisorders (ASDs)inTaiwanwhileexaminingtheeffectsofage,gender,andurbanizationonASD occurrence.Across-sectionalstudywasconductedtoanalyzedatafrom895,639random healthinsuranceclaimantswhoclaimedmedicalservicesintheyear2007.Autismwas defined using the ICD-9-CM Diagnosis Code 299.0 (autism, current or active). The prevalence ofautistic caseswas found to be 12.3%(10,868/884,771) inthe general population,with theprevalenceamong males (19.2%)significantly higher thanthat amongfemales(6%).Withregardstoagedistribution,wefoundthattheautisticgroup (meanage=16.0years)was significantlyyounger thanthegeneralpopulation (mean age=37.2years). A logistic regression analysis found that age, gender, residence urbanization level and Bureau of National Health Insurance regional division all constituted influence factors for autistic occurrence. The results demonstrate the importanceoftakingintoaccountage,gender,andgeographicaldisparitiesinautistic prevalence in order to implement appropriate public health policies for the ASD population.

ß2011ElsevierLtd.Allrightsreserved.

*Correspondingauthorat:SchoolofPublicHealth,NationalDefenseMedicalCenter,No.161,Min-ChunEastRoad,Section6,Nei-Hu,Taipei,Taiwan.

Tel.:+886287923100x18447;fax:+886287923147.

E-mailaddress:a530706@ndmctsgh.edu.tw(J.-D.Lin).

ContentslistsavailableatSciVerseScienceDirect

Research in Autism Spectrum Disorders

J our na l ho me pa ge : ht t p: / / e e s. e l s e v i e r. c om/ R A S D / de f a ul t . a sp

1750-9467/$–seefrontmatterß2011ElsevierLtd.Allrightsreserved.

doi:10.1016/j.rasd.2011.09.006

TheobservedprevalenceofASDhasincreasedoverthepastfewdecades(Leonardetal.,2010;Lin,Hung,Lin,&Lai,2011;

Lin,Lin,&Wu,2009;Lin,Sung,etal.,2011;Lingametal.,2003;Matson&Kozlowski,2011;Nassaretal.,2009).Thisobserved increasemaybeduetovariousfactors,includingchangesindiagnosticcriteria,thedevelopmentofamoreexpansivenotion ofASD,differentmethodsusedinstudies,andgrowingawarenessamongthepublicaboutASDcombinedwithgreater availabilityofprofessionalservices.Itisalsopossiblethattheprevalenceinthepopulationhasactuallyincreased(Wing&

Potter,2002).After reviewing40 studiesofASDprevalence,Williams, Higgins,and Brayne(2006)reportedanoverall randomeffectsprevalenceestimatefortypicalautismof7.1per10,000;thecomparablevalueforallASDwas20.0per 10,000.

TheUSCenterofDiseaseControl(2009)reportedanincreasedprevalenceofASDsamongUSchildrenandunderscored theneedtoregardthesyndromeasanurgentpublichealthconcern(CDC,2009).Anincreaseinprevalencewouldsuggest directingmoreattentionandfundingtowardimprovingservicesandalleviatingenvironmentalriskfactorsforchildrenwith ASDinsteadofthelabelofASD(Linetal.,2009).OurpreviousstudyfoundthatindividualswithASDhadmorefrequent hospital admissionscompared tothegeneral population;accordingly, wesuggested that health caredecision makers developappropriatehealthpoliciesfortheseindividuals(Lin,Hung,etal.,2011;Lin,Sung,etal.,2011).Manystudieshave alsoreportedthepresenceofcomorbiditiesamongASDindividuals;however,thenatureandprevalenceoftheseconditions remainpoorlyunderstood(Bauman,2010;Lauritsen,Mors,Mortensen,&Ewald,2002).Anaccurateunderstandingofthe relationshipbetweenautismseverityandhealthcareandeducationneedsisalsonecessaryinordertoimprovethequalityof serviceforpeoplewithASD(Lin,Hung,etal.,2011;Lin,Sung,etal.,2011).Duetotheimportanceofobtainingaccurate demographicdataonASDindividualsforformulatingpublichealthpolicy,thepresentstudysoughttocharacterizeASD prevalenceinTaiwan.Wealsoexaminedtheeffectsofage,gender,andurbanizationonASDoccurrence.

2. Methods

ThepresentstudysoughttoexaminetheprevalenceofautisminTaiwanbymeansofacross-sectionalanalysisofa nationalhealthinsurancedatasetprovidedbytheNationalHealthResearchInstitute.Thedatasetincluded895,639random healthinsuranceclaimantswhoclaimedmedicalservicesin theyear2007.Amongtheclaimants,10,868peoplehada principaldiagnosisofautismspectrumdisorder(ICD-9-CMDiagnosisCode299.0:Autism,currentoractive).

Thefollowingvariableswereincludedintheanalysistoidentifythedemographiccharacteristicsassociatedwithautism prevalence:autisticpopulationnumber,age,gender,urbanizationlevelandregionaldivisionoftheclient’sresidence,as categorized by the Bureauof National Health Insurance (BNHI). The statistical methods utilized include frequencies, percentages,populationpyramids,t-tests,ANOVA,andlogisticregression toexaminetherelationshipbetweenautism prevalenceandage,gender,andurbanizationinTaiwan.

ThedefinitionofautismusedinthestudywasthatoftheUSNationalCenterforHealthStatistics(1993):autisticdisorder beginninginchildhoodismarkedbythepresenceofnotablyabnormalorimpaireddevelopmentinsocialinteractionand communicationaswellasa markedlyrestrictedrepertoireofactivityandinterest.Manifestationsofthedisordervary greatlydependingonthedevelopmentallevelandchronologicalageoftheindividual.InTaiwan,autismisoneof16officially registered disabilities (Physically and Mentally Disabled Citizens Protection Act, 1997), and persons withautism are characterizedbyimpairedsocialinteraction,problemswithverbalandnonverbalcommunication,andunusual,repetitive, orseverelylimitedactivitiesandinterests(TaiwanDepartmentofHealth,2006).

Urbanizationinthestudywasstratifiedintofourlevels–I,II,IIIandIV–rangingfromgreatertolowerdegreesof urbanization oftheresidentialarea. Theclassificationwasdefinedbased onthefollowingcriteria:populationdensity (people/km²),proportionofresidentswithatleastacollegeeducation,proportionofelderlypeopleolderthan65,proportion ofagriculturalworkersandnumberofphysiciansper100,000residents(Lin,Hsieh,Chiou,Wu,&Huang,2010).

3. Results

ThedemographiccharacteristicsoftheautisticandthegeneralpopulationsarepresentedinTable1andFig.1.The prevalence ofautistic caseswas12.3% (10,868/884,771)in thegeneralpopulation withtheprevalenceamong males (19.2%)significantlyhigherthanthatamongfemales(6%).Withregardstotheagedistribution,wefoundtheautisticgroup (meanage=16.0years)wassignificantlyyoungerthanthegeneralpopulation(meanage=37.2years).Themorepopulous metropolitanareashadahigherprevalenceofautisticcasesthantheruralareasinTaiwan.

Table 2reveals agedisparities intheautistic populationbydemographiccharacteristics.Generalizedlineartests found thatmalecases (mean age=15.1years)weresignificantlyyoungerthanfemalecases (meanage=18.4years).

Casesinmoreurbanizedareaswerealsosignificantlyyoungerthancasesinlessurbanizedareas(meanage=15.6years in level I versusmean age=19.7years in level IV). Analyses basedon client’s residence in BNHI regional division similarly indicated that autistic cases in metropolitan cities were significantly younger than those living in less metropolitanareas.

Table3providestheresultsofalogisticregressionanalysisofautismprevalenceinTaiwan.Wefoundthatage,gender, residenceurbanizationlevelandBNHIregionaldivisionwereallfactorsthataffectedautisticprevalencecaseswereless commonamongfemalesthaninmales(OR=0.339,95%CI=0.324–0.354),whereasindividualslivinginlevelIurbanized areasreportedgreaterprevalencethanindividualslivinginlevelIVurbanizedareas(OR=1.151,95%CI=1.016–1.303).In

termsofthegeographiccorrelatesofautisticoccurrence,theanalysesshowedthatclientsresiding intheTaipeiBNHI regionaldivision had a significantly higher autistic prevalence than those residing in eastern areas (OR=1.186, 95%

CI=1.035–1.359).However,thecentral(OR=0.735,95%CI=0.639–0.845)andsouthernareas(OR=0.764,95%CI=0.663–

0.880)hadlowerprevalencethantheeasternareas.Wealsofoundthatautismwasmoreprevalentamongyoungerpeople, withindividualsyoungerthanage50reportinghigherautisticprevalencethanthoseolderthan65years.Prevalencewas especially elevated among the following age groups: 0–5years (OR=18.869, 95% CI=14.898–23.899), 6–10years (OR=57.014,95%CI=45.540–71.446),11–15years(OR=60.016,95%CI=47.931–75.147),16–20years(OR=32.447,95%

CI=25.849–40.720),and21–25years(OR=19.293,95%CI=15.300–24.329).

4. Discussions

Autismis a condition that manifestsin earlychildhoodand is characterized byqualitative abnormalities in social interactions,markedlyaberrantcommunicationskills,and restricted,repetitive andstereotypedbehaviors.Theautism spectrumisbroad,andtherearesignificantdifferencesbetweenhigh-functioningautismorAsperger’ssyndromeandthe moreseveremanifestationsofthisdisorder(Levy&Perry,2011).Amajorpublichealthconcernforchildrenwithautismand Table1

Demographiccharacteristicsoftheautisticandthegeneralpopulations.

Variables Generalpopulation,N(%) Autisticpopulation,N(%) x²test

Gender 3100.4^***

Female 463,606(52.4) 2777(25.6)

Male 421,165(47.6) 8091(74.4)

Agegroup(meanSD) 37.220.8 16.011.0 t=106.3^***,20,160.3^***

0–5 37,136(4.2) 603(5.6)

6–10 60,257(6.8) 2952(27.2)

11–15 64,778(7.3) 3331(30.7)

16–20 62,193(7.0) 1696(15.6)

21–25 60,823(6.9) 887(8.2)

26–30 75,632(8.6) 494(4.6)

31–35 71,334(8.0) 283(2.6)

36–40 70,926(8.0) 196(1.8)

41–45 73,462(8.3) 130(1.2)

46–50 69,467(7.9) 101(0.9)

51–55 64,142(7.3) 48(0.4)

56–60 48,185(5.5) 34(0.3)

61–65 31,161(3.5) 36(0.3)

66 95,275(10.8) 78(0.7)

Urbanizationlevelofresidence 119.9^***

I 511,347(57.8) 6841(62.9)

II 299,537(33.9) 3280(30.2)

III 45,968(5.2) 465(4.3)

IV 27,919(3.2) 283(2.6)

BNHIregionaldivisionofresidence 362.3^***

Taipeidivision 318,979(36.1) 4785(44.0)

Northerndivision 126,913(14.3) 1454(13.4)

Centraldivision 163,157(18.4) 1584(14.6)

Southerndivision 163,157(14.1) 1182(10.9)

Kaopingdivision 131,613(14.9) 1620(14.9)

Easterndivision 19,426(2.2) 244(2.2)

*** p<0.001.

Fig.1.Populationpyramidtypesofautismandthegeneral.

otherpervasivedevelopmentaldisordersistheinconsistentdiagnosisofautism(AmericanPsychiatricAssociation,2000;

NationalCenterforHealthStatistics,1993).ThepresentpaperprovidesdataonASDprevalencebasedonTaiwanesenational healthinsuranceclaimantsandexaminestheeffectsofage,genderandurbanizationonASDoccurrence.Theresultsshow that theprevalenceofautistic caseswas12.3%inthegeneral population,withtheprevalenceamongmales(19.2%) significantlyhigherthanthatamongfemales(6%).

Incomparingourresultswiththoseobtainedbyotherstudiesofautisticprevalence,approximately1%wasclassifiedas havinganASDinasurveyof11USsitesintheAutismandDevelopmentalDisabilitiesMonitoringNetwork(CDC,2009).In thestateofNew Jersey,ASDprevalence was10.2/1000, higherinboysthanin girls(16versus4/1000)andhigherin wealthiercensustracts,perhapsduetodifferentialaccesstopediatricanddevelopmentalservices(Thomasetal.,2011).In theUK,Baron-Cohenetal.(2009)estimatedtheprevalenceofASDtobe157per10,000basedonasurveyofautism- spectrumconditionsusingtheSpecialEducationalNeeds(SEN)registerinCambridgeshire.InAsiancountries,theaverage prevalenceofASDbefore1980wasaround1.9/10,000whileitwas14.8/10,000from1980tothepresent(Sun&Allison, Table2

Agedisparityinautisticpopulationbydemographiccharacteristics(generalizedlinearmodel,GLMtest).

Variables Age(meanSD) Fvalue Scheffe’stest

Gender

1.Female 18.414.0 196.8^*** 1>2

2.Male 15.19.6

Urbanizationlevel

1.I 15.610.4 9.3^*** 4>1,2

2.II 16.111.3 3>1,2

3.III 18.413.8

4.IV 19.714.4

BNHIregionaldivision

1.Taipeidivision 14.910.2 31.7^*** 5>3,4>1,2

2.Northerndivision 15.010.1 6>1,2

3.Centraldivision 16.211.0 4>1,2

4.Southerndivision 16.611.1 3>1,2

5.Kaopingdivision 19.012.8

6.Easterndivision 17.912.8

*** p<0.001.

Table3

LogisticregressionanalysisofautisticpopulationoccurrenceinTaiwan(N=895,639).

Variables Oddsratio 95%CI pvalue

Gender

1.Female 0.339 0.324–0.354 <0.001

2.Male(ref.) Urbanizationlevel

1.I 1.151 1.016–1.303 0.0270

2.II 1.034 0.912–1.171 0.604

3.III 1.087 0.934–1.266 0.0282

4.IV(ref.) BNHIregionaldivision

1.Taipeidivision 1.186 1.035–1.359 0.011

2.Northerndivision 0.877 0.762–1.011 0.070

3.Centraldivision 0.735 0.639–0.845 <0.001

4.Southerndivision 0.764 0.663–0.880 0.002

5.Kaopingdivision 1.004 0.878–1.156 0.951

6.Easterndivision(ref.) Agegroup

1.0–5 18.869 14.898–23.899 <0.001

2.6–10 57.014 45.540–71.446 <0.001

3.11–15 60.016 47.931–75.147 <0.001

4.16–20 32.447 25.849–40.720 <0.001

5.21–25 19.293 15.300–24.329 <0.001

6.26–30 8.096 6.374–10.283 <0.001

7.31–35 4.815 3.747–6.189 <0.001

8.36–40 3.329 2.560–4.329 <0.001

9.41–45 2.126 1.605–2.816 <0.001

10.46–50 1.740 1.294–2.338 0.002

11.51–55 0.892 0.623–1.278 0.534

12.6–60 0.845 0.564–1.264 0.411

13.61–65 1.411 0.951–2.095 0.088

14.66(ref.)

2010).InSouthKorea,Kimetal.(2011)estimatedtheprevalenceofASDstobe2.64%;with1.89%,themale-to-femaleratios were2.5:1and5.1:1inthegeneral-populationsampleand0.75%inthehigh-probabilitygroup.

Age,genderandgeographicaldisparitiesexistintheautisticpopulationinTaiwan.Alogisticregressionanalysisinthe presentstudyfoundthattheinfluencefactorsofautisticoccurrenceincludeage,gender,residenceurbanizationleveland BNHIregionaldivision.ASDvariedgeographicalregion.Similarly,Roelfsemaetal.(2011)foundthatASDwasmoreprevalent inaninformation-technologyregionbasedonaschool-basedstudyofthreeregionsintheNetherlands.Ourpreviousstudy reportedtheadministrativeprevalenceofASDamongvariousagegroupsinTaiwanfrom2000to2007asfollows:2.4–7.8/

10,000amongthoseaged0–5years,5.0–17.3/10,000amongthoseaged6–11years,and2.1–10.4/10,000amongthoseaged 12–17years(Linetal.,2009).

GenderdifferencesinsymptomsrepresentingthetriadofimpairmentsofASDremainunclear.Sipes,Matson,Worley,and Kozlowski(2011)foundthatgenderdifferencesinasampleoftoddlersoccurredonlyinregardstotherestrictedinterests and repetitive behavior domain, with females with an average developmental quotient having significantly fewer endorsementsonitemsrelatedtorestrictiveandrepetitive behaviors.However, genderdifferencesinprevalencerates emergedin communityor clinicalpopulations(Worley,Matson,Sipes, &Kozlowski,2011). Baron-Cohenet al.(2011) reportedthatASDismuchmorecommoninmales,citingtheextrememalebrain(EMB)theory,whichviewsASDasan extrememanifestationofthemalebrain,asthelikeliestexplanation.OurpreviousanalysisconcludedthatintheASD disabilityleveldifferencebygender,themalecasesoccupiedmostoftheautisticnumbersinthestudy.Thereportedmale/

femaleratiowas7.3–8.4inmilddisabilitygroup,5.9–7.5inmoderatedisabilitygroup,4.1–4.7inseveredisabilitygroup,and 2.7–4.4inprofounddisabilitygroup(Lin,Hung,etal.,2011;Lin,Sung,etal.,2011).

TheawarenessofASDshasresultedinincreasingnumbersofchildrenbeingdiagnosedatyoungages(Fernell&Gillberg, 2010).MatsonandKozlowski(2011)suggestedthatdifferentresearchmethodologiesanddiagnosticcriteria,increasedASD awareness,cultural factors,and environmental contributions mightalso play roles in theincrease in ASD diagnoses.

However,Muhleetal.(2004)arguedthattheincreaseismostlikelyattributabletoheightenedawarenessandchanging diagnosticcriteriaratherthantonewenvironmentalinfluences.Williamsetal.(2006)foundthatdiagnosticcriteria,ageof thesampledindividuals, and urbanorrural locationwereall associatedwithestimated prevalence ofall ASD.Many countrieshaveadoptedthepolicyofearly-agedetectioninASDcaseidentification(Dababnah,Parish,Brown,&Hooper, 2011;LoaJonsdottir,Saemundsen,Antonsdottir,Sigurdardottir,&Olason,2011;Lung,Chiang,Lin,&Shu,2011).However, thequestionofhowearlyistooearlywithrespecttoearlyidentificationanddiagnosisofASDinyoungchildrenandinfants remainsopentodebate(Matson,Wilkins,&Gonzalez,2008).

ToimplementappropriatepublichealthpoliciesfortheASDpopulation,itisimportanttotakeintoaccountthebroader developmentalprofilesofASDchildren,theneedforrepeatedassessmentofcognitivefunctionsandfollow-upovertimeand the required medical/neurological consideration and work-up (Fernell & Gillberg, 2010). Future research should be conductedusingdiagnosticcriteriaandassessmentinstrumentsthatareasuniformaspossible(Matson&Kozlowski,2011).

AssuggestedbyKimetal.(2011),rigorousscreeningandcomprehensivepopulationcoveragearenecessarytoproducemore accurateASDprevalenceestimates,afactthatonlyunderscorestheneedforbetterdetection,assessment,andservices.

Acknowledgements

Thisstudyis basedinpartona projectfortheNationalHealthInsuranceResearch(PH-100-PP-27).The datawere providedbytheBureauofNationalHealthInsurance,DepartmentofHealthandmanagedbytheNationalHealthResearch Institutes.TheinterpretationandconclusionscontainedhereindonotrepresentthoseoftheBureauofNationalHealth Insurance,theDepartmentofHealthortheNationalHealthResearchInstitutes.

References

AmericanPsychiatricAssociation.(2000).Diagnosticandstatisticalmanualofmentaldisorders(4thed.–textrevision).Washington,DC:Author.

Baron-Cohen,S.,Lombardo,M.V.,Auyeung,B.,Ashwin,E.,Chakrabarti,B.,&Knickmeyer,R.(2011).Whyareautismspectrumconditionsmoreprevalentinmales?

PLoSBiology,9,e1001081.

Baron-Cohen,S.,Scott,F.J.,Allison,C.,Williams,J.,Bolton,P.,Matthews,F.E.,etal.(2009).Prevalenceofautism-spectrumconditions:UKschool-basedpopulation study.TheBritishJournalofPsychiatry,194,500–509.

Bauman,M.L.(2010).Medicalcomorbiditiesinautism:Challengestodiagnosisandtreatment.Neurotherapeutics,7,320–327.

CDC.(2009).Prevalenceofautismspectrumdisorders—AutismandDevelopmentalDisabilitiesMonitoringNetwork,UnitedStates,2006.MorbidityandMortality WeeklyReport:SurveillanceSummaries,58,1–20.

Dababnah,S.,Parish,S.L.,Brown,L.T.,&Hooper,S.R.(2011).Earlyscreeningforautismspectrumdisorders:Aprimerforsocialworkpractice.ChildrenandYouth ServicesReview,33,265–273.

Fernell,E.,&Gillberg,C.(2010).AutismspectrumdisorderdiagnosesinStockholmpreschoolers.ResearchinDevelopmentalDisabilities,31,680–685.

Kim,Y.S.,Leventhal,B.L.,Koh,Y.J.,Fombonne,E.,Laska,E.,Lim,E.C.,etal.(2011).Prevalenceofautismspectrumdisordersinatotalpopulationsample.American JournalofPsychiatry May9(Epubaheadofprint).

Lauritsen,M.B.,Mors,O.,Mortensen,P.B.,&Ewald,H.(2002).MedicaldisordersamonginpatientswithautisminDenmarkaccordingtoICD-8:Anationwide register-basedstudy.JournalofAutismandDevelopmentalDisorders,32,115–119.

Leonard,H.,Dixon,G.,Whitehouse,A.,Bourke,J.,Aiberti,K.,Nassar,N.,etal.(2010).Unpackingthecomplexnatureoftheautismepidemic.ResearchinAutism SpectrumDisorders,4,548–554.

Levy,A.,&Perry,A.(2011).Outcomesinadolescentsandadultswitautism:Areviewoftheliterature.ResearchinAutismSpectrumDisorders,5,1271–1282.

Lin,J.D.,Hung,W.J.,Lin,L.P.,&Lai,C.I.(2011).Utilizationandexpenditureofhospitaladmissioninpatientswithautismspectrumdisorder:NationalHealth Insuranceclaimsdatabaseanalysis.ResearchinAutismSpectrumDisorders,5,1138–1142.

Lin,J.D.,Lin,L.P.,&Wu,J.L.(2009).AdministrativeprevalenceofautismspectrumdisordersbasedonnationaldisabilityregistersinTaiwan.ResearchinAutism SpectrumDisorders,3,269–274.

Lin,J.D.,Sung,C.L.,Lin,L.P.,Hsu,S.W.,Chien,W.C.,Su,S.L.,etal.(2011).Ten-yeartrendanalysisofautismseverity:Anationwidepopulation-basedregister study.ResearchinAutismSpectrumDisorders,5,570–574.

Lin,W.,Hsieh,C.J.,Chiou,S.J.,Wu,H.Y.,&Huang,I.C.(2010).Aneffectivenessevaluationofapay-for-performanceprogramfordiabetesbasedonthepropensity scorematchingmethod.TaiwanJournalofPublicHealth,29,54–64.

Lingam,R.,Simmons,A.,Andrews,N.,Miller,E.,Stowe,J.,&Taylor,B.(2003).PrevalenceofautismandparentallyreportedtriggersinanortheastLondon population.ArchivesofDiseaseinChildhood,88,666–670.

LoaJonsdottir,S.,Saemundsen,E.,Antonsdottir,I.S.,Sigurdardottir,S.,&Olason,D.(2011).Childrendiagnosedwithautismspectrumdisorderbeforeorafterthe ageof6years.ResearchinAutismSpectrumDisorders,5,175–184.

Lung,F.W.,Chiang,T.L.,Lin,S.J.,&Shu,B.C.(2011).Autism-riskscreeninginthefirst3yearsoflifeinTaiwanBirthCohortPilotStudy.ResearchinAutismSpectrum Disorders,5,1385–1389.

Matson,J.L.(2007).Currentstatusofdifferentialdiagnosisforchildrenwithautismspectrumdisorders.ResearchinDevelopmentalDisabilities,28,109–118.

Matson,J.L.,Gonzalez,M.L.,&Wilkins,J.(2009).ValiditystudyoftheAutismSpectrumDisorders-DiagnosticforChildren(ASD-DC).ResearchinAutismSpectrum Disorders,3,196–206.

Matson,J.L.,Gonzalez,M.L.,Wilkins,J.,&Rivet,T.T.(2008).ReliabilityoftheAutismSpectrumDisorder-DiagnosticforChildren(ASD-DC).ResearchinAutism SpectrumDisorders,2,533–545.

Matson,J.L.,&Kozlowski,A.M.(2011).Theincreasingprevalenceofautismspectrumdisorders.ResearchinAutismSpectrumDisorders,5,418–425.

Matson,J.L.,Nebel-Schwalm,M.,&Matson,M.L.(2007).Areviewofmethodologicalissuesinthedifferentialdiagnosisofautismspectrumdisordersinchildren.

ResearchinAutismSpectrumDisorders,1,38–54.

Matson,J.L.,Wilkins,J.,&Gonzalez,M.(2008).Earlyidentificationanddiagnosisinautismspectrumdisordersinyoungchildrenandinfants:Howearlyistoo early?ResearchinAutismSpectrumDisorders,2,75–84.

Muhle,R.,Trentacoste,S.V.,&Rapin,I.(2004).Thegeneticsofautism.Pediatrics,113,e472–e486.

Nassar,N.,Dixon,G.,Bourke,J.,Bower,C.,Glasson,E., deKlerk,N.,etal.(2009).Autismspectrumdisordersinyoungchildren:Effectofchangesindiagnostic practices.InternationalJournalofEpidemiology,38,1245–1254.

NationalCenterforHealthStatistics.(1993).InternationalClassificationofDiseases(ICD-9-CM),9thRevision,ClinicalModification(4thed.).LosAngeles,CA:Practice ManagementInformationCorporation(PMIC).

Sipes,M.,Matson,J.L.,Worley,J.A.,&Kozlowski,A.M.(2011).Genderdifferencesinsymptomsofautismspectrumdisordersintoddlers.ResearchinAutism SpectrumDisorders,5,1465–1470.

Sun,X.,&Allison,C.(2010).AreviewoftheprevalenceofautismspectrumdisorderinAsia.ResearchinAutismSpectrumDisorders,4,156–167.

TaiwanDepartmentofHealth.(2006).Disabilitylevelandclassification.DepartmentofHealth,no.0952801549,September8,2006,Taiwan,RepublicofChina.

Thomas,P.,Zahorodny,W.,Peng,B.,Kim,S.,Jani,N.,Halperin,W.,etal.(2011).Theassociationofautismdiagnosiswithsocioeconomicstatus.AutismAugust2 (Epubaheadofprint).

PhysicallyandMentallyDisabledCitizensProtectionAct.(1997).PresidentOrderHua-Tsung-Yi-Tzu,no.8600097810,April23,1997,Taiwan,RepublicofChina.

Roelfsema,M.T.,Hoekstra,R.A.,Allison,C.,Wheelwright,S.,Brayne,C., Matthews,F.E.,etal.(2011).Areautismspectrumconditionsmoreprevalentinan information-technologyregion?Aschool-basedstudyofthreeregionsintheNetherlands.JournalofAutismandDevelopmentalDisordersJune17(Epubahead ofprint).

Williams,J.G.,Higgins,J.P.,&Brayne,C.E.(2006).Systematicreviewofprevalencestudiesofautismspectrumdisorders.ArchivesofDiseaseinChildhood,91,8–15.

Wing,L.,Gould,J.,&Gillberg,C.(2011).AutismspectrumdisordersintheDSM-V:BetterorworsethantheDSM-IV?ResearchinDevelopmentalDisabilities,32, 768–773.

Wing,L.,&Potter,D.(2002).Theepidemiologyofautisticspectrumdisorders:Istheprevalencerising?MentalRetardationandDevelopmentalDisabilitiesResearch Reviews,8,151–161.

Woolfenden,S.,Sarkozy,V.,Ridley,G.,&Williams,K.(2011).Asystematicreviewofthediagnosticstabilityofautismspectrumdisorder.ResearchinAutism SpectrumDisordersdoi:10.1016/j.rasd.2011.06.008.

Worley,J.A.,Matson,J.L.,Sipes,M.,&Kozlowski,A.M.(2011).Prevalenceofautismspectrumdisordersintoddlersreceivingearlyinterventionservices.Research inAutismSpectrumDisorders,5,920–925.