NEW RESEARCH

JOURNAL

VOLUM

Children’s Yale–Brown Obsessive CompulsiveScale in Autism Spectrum Disorder:Component Structure and Correlates

of Symptom ChecklistLawrence Scahill, MSN, PhD, Anastasia Dimitropoulos, PhD,

Christopher J. McDougle, MD, Michael G. Aman, PhD, Irene D. Feurer, PhD,James T. McCracken, MD, Elaine Tierney, MD, Jie Pu, PhD, Susan White, PhD,Luc Lecavalier, PhD, Victoria Hallett, PhD, Karen Bearss, PhD, Bryan King, MD,

L. Eugene Arnold, MD, MEd, Benedetto Vitiello, MD

Objective: Repetitive behaviors in autism spectrum disorders (ASD) range from motor ste-reotypy to immersion in restricted interests. The modified Children’s Yale–Brown ObsessiveCompulsive Scale for children with autism spectrum disorder (CYBOCS-ASD) includes aSymptom Checklist (behavior present or absent) and 5 severity scales (Time Spent, Interference,Distress, Resistance and Control). Method: We assembled CYBOCS-ASD data from 3Research Units on Pediatric Psychopharmacology Autism Network trials to explore thecomponent structure of repetitive behaviors in children with ASD. Raters trained to reliabilityconducted the CYBOCS-ASD in 272 medication-free subjects. Fifteen Symptom Checklist itemswere endorsed for less than 5% of the sample and were dropped. Principal component analysiswas used to explore the clustering of 23 checklist items. Component scores computed for eachsubject were correlated with other measures. We also examined the distribution of severityscales. Results: The subjects (229 boys and 43 girls; mean age ¼ 7.8 � 2.6 years) met criteriafor an ASD; half were intellectually disabled. The PCA resulted in a 5-component solution toclassify repetitive behaviors (34.4% of the variance): hoarding and ritualistic behavior; sensoryand arranging behavior; sameness and self-injurious behavior; stereotypy; restricted interests.Sensory and arranging and stereotypy components were associated with lower adaptivefunctioning (Pearson r ¼ 0.2–0.3; p < .003). The resistance scale showed little variation, withmore than 60% of the sample with the highest score. Conclusions: Rarely endorsed items canbe dropped from the Checklist. The resistance item does not appear to be relevant for childrenwith ASD. J. Am. Acad. Child Adolesc. Psychiatry, 2014;53(1):97–107. Key Words: autismspectrum disorder, clinical trials, outcome measures, repetitive behavior

utism spectrum disorder (ASD) is achronic condition of early childhood

A onset that is defined by impaired social

interaction and social communication as wellas repetitive behavior and restricted interests.1

Current prevalence estimates for ASD rangefrom 0.6% to 1.3% of school-aged children in theUnited States.2,3 Two medications, risperidone

Supplemental material cited in this article is available online.

OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATR

E 53 NUMBER 1 JANUARY 2014

and aripiprazole, are currently approved by theUS Food and Drug Administration for treatingirritability (tantrums, aggression, and self-injury)in children with more narrowly defined autisticdisorder. Empirically supported treatments forimpaired social interaction or repetitive behaviorare less well established and are hampered byincomplete consensus on outcome measurement.

Repetitive behavior is a defining feature ofother psychiatric disorders including Tourettesyndrome (TS) and obsessive-compulsive disor-der (OCD). Tics in TS range from eye blinking,head jerking, and throat clearing to more complex

Y

www.jaacap.org 97

SCAHILL et al.

touching habits and blurting out words or partsof words.4 In OCD, children may show hand-washing, checking or elaborate rituals to pre-vent harm, or touching in patterns to achieve asense of completion.5,6

Children with ASD may exhibit stereotypicmotor behavior such as hand flapping, rocking,or spinning objects, or more complex behaviorssuch as repeatedly watching particular videosegments.7,8 Several investigators have described“higher”- and “lower”-order repetitive behav-iors.9-12 In this model, higher-order behaviorsreflect insistence on routines and more purpose-ful behavior (watching the same video), whereaslower-order behaviors include motor stereotypiesor sensory-motor behavior such as spinningan object. Other investigators propose circum-scribed interests as a third domain.11 The behav-ioral correlates of circumscribed interests mightinclude reading books on a certain subject ortalking about that topic (e.g., mechanical devices)to the exclusion of other topics.13 In contrast tothe unwanted repetitive behaviors in TS andOCD, children with ASD may not struggleagainst their repetitive behaviors. Indeed, formany children with ASD, the repetitive behavioris a preferred activity. For some children, frus-tration, protest, and tantrums may occur whenthe child is asked to stop the behavior.7,13

In preparation for multisite trials in ASD, theResearch Units on Pediatric Psychopharmacol-ogy (RUPP) Autism Network modified theChildren’s Yale–Brown Obsessive CompulsiveScale (CYBOCS).7 The original CYBOCS is aclinician-rated interview designed to measurecurrent symptom severity in youth with OCD.14

It contains separate checklists for obsessions andcompulsions, 5 severity items for obsessions and5 for compulsions. The severity items are scoredfrom 0 (not present) to 4 (extreme), yielding anObsessions score of 0 to 20, a Compulsions scoreof 0 to 20, and a total of 0 to 40. Given thechallenge of ascertaining obsessions in in-dividuals with ASD, the modified CYBOCS forASD retains only the 5 compulsion items.7 Wealso added items to the compulsions checklistand made minor adjustments to the severityanchor points for use in children with ASD. Therevised instrumnt has demonstrated reliabilityand validity and has also demonstrated sensi-tivity to change.7,15

The purpose of this study is to extend ourprevious work on adapting the original CYBOCSas an outcome measure in pediatric clinical trials

JOURN

98 www.jaacap.org

in ASD and other developmental disorders. Tothis end, we examined the component structureof the compulsions checklist and the distributionof severity scores on the CYBOCS-ASD in asample of well-characterized children with ASD.

METHODParticipants and SettingA total of 291 subjects participated in 1 of 3 multisiterandomized trials conducted by the RUPP AutismNetwork.16-19 Baseline CYBOCS-ASD checklists weremissing or illegible for 19 subjects. Thus, the studysample included 272 subjects (229 boys and 43 girls;aged 4 through 17 years). The institutional reviewboard at each site approved the studies and writteninformed consent was obtained from parents or legalguardians before data collection. Assent was obtainedfrom children who were able to understand the detailsof their involvement.

Eligible subjects at baseline were healthy, had amental age of 18 months or higher, were free of apsychiatric condition requiring another treatment andwere medication free for at least 2 weeks (1 month forfluoxetine or antipsychotic medications). In the firsttrial (RUPP I), 101 children and adolescents withautistic disorder (aged 5–17 years) accompanied byserious behavioral problems (Aberrant BehaviorChecklist [ABC] Irritability subscale �18) were ra-ndomly assigned to risperidone or placebo for 8 weeks(RUPP Autism Network, 2002). RUPP II compared3 doses of methylphenidate to placebo in 66 children(aged 5–14 years) with autistic disorder, Asperger’sdisorder, or pervasive developmental disorder nototherwise specified (PDD-NOS) plus hyperactivity in a4-week crossover trial (RUPP Autism Network, 2005).17

In the third trial (RUPP III), 124 children (aged 4.5–13years) with ASD and serious behavioral problems(ABC Irritability subscale �18) were randomlyassigned to risperidone only or risperidone plus parenttraining for 6 months.18,19

ProceduresThe clinical assessments were conducted by experi-enced multidisciplinary teams and included completemedical, developmental, and psychiatric histories, testsof IQ and adaptive functioning, as well as parent in-terviews and questionnaires.16-19 The ASD diagnosiswas based on clinical interview and observation andwere supported by the Autism Diagnostic Interview–

Revised.20

MeasuresIntellectual Functioning. We used 1 of several standardtests according to the child’s ability: Wechsler Intelli-gence Scale for Children–III; Leiter InternationalPerformance Scale–Revised; Mullen Scales of Early

AL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

VOLUME 53 NUMBER 1 JANUARY 2014

TABLE 1 Clinical Characteristics of 272 Subjects in 3Research Units on Pediatric Psychopharmacology (RUPP)Autism Network Trials

Gender, male, n (%) 229 (84.2)IQ <70,a n (%) 141(55.7)Functional language (on ADI-R verbal item),n (%)

198 (73.6)

Diagnosis, n (%)Autistic disorder 209 (76.8)Asperger’s disorder 14 (5.1)PDD-NOS 49 (18.0)

Clinical Global Impression Severity, n (%)Moderate 75 (27.7)Marked 135(49.8)Severe 59 (21.8)Extreme 2 (0.7)

Age, y, mean (SD) 7.8 (2.6)CYBOCS-ASD total, m (SD) 14.8 (3.2)

Time spent 2.9 (0.94)Interference 2.4 (0.91)Distress 2.7 (0.86)Resistance 3.4 (0.79)Control 3.3 (0.75)

Child and Adolescent Symptom Inventory(CASI), m (SD)Anxiety scale 14.7 (9.5)PDD scale 22.2 (7.6)

Vineland Adaptive Behavior Scales,m (SD)Communication 54.9 (20.9)Socialization 55.6 (15.9)Daily Living 46.1(20.4)

Aberrant Behavior Checklist, m (SD)Irritability 25.4 (9.3)Social withdrawal 15.0 (8.9)Stereotypy 8.7 (5.4)Hyperactivity 34.1 (8.7)Inappropriate speech 5.9 (3.8)

Note: ADI-R ¼ Autism Diagnostic IntervieweRevised; CASI ¼ Childand Adolescent Symptom Inventory; CYBOCS-ASD ¼ Children’sYaleeBrown Obsessive Compulsive Scale for Children with AutismSpectrum Disorder; PDD-NOS ¼ pervasive developmental disor-der not otherwise specified.aIQ was missing for 19 subjects because of poor cooperation with

testing.

CYBOCS-ASD COMPONENT STRUCTURE

Learning; Slosson Intelligence Test; or Stanford-Binet,version 5. Given the use of multiple tests, we classi-fied subjects as intellectually disabled (IQ < 70) or inthe normal range (IQ > 70).

Vineland Adaptive Behavior Scales, Survey Edition(Vineland; Sparrow, Balla, and Cicchetti, 1984). TheVineland is an age and gender standardized measureof adaptive functioning (population mean of 100 � 15)across 3 domains: Communication, Socialization, andDaily Living Skills. Vineland items are scored 0 (behaviornot performed), 1 (performed sometimes), or 2 (per-formed on a regular basis).21

Aberrant Behavior Checklist (ABC)22. This reliableand valid, 58-item, informant-based scale (scored 0–3)contains 5 factor-analyzed subscales: Irritability, SocialWithdrawal, Stereotypy, Hyperactivity, and Inappro-priate Speech.

Clinical Global Impression Scale–Severity (CGI-S)23. TheCGI-S is a commonly used, clinician-rated scale assess-ing overall impairment. Raters were trained to reliabilityon this 7-point scale that ranges from nonsymptomatic(1) to extreme (7). In these 3 trials; a score of 4 (moder-ate) or greater was required for entry.

Child and Adolescent Symptom Inventory (CASI)24.This 132-item, DSM-IV–referenced, parent-rated scalewas used to screen for psychiatric disorders. Items arescored from 0 (never) to 3 (very often), and can bescored dimensionally as the sum of the scores forthe specific diagnostic subscale (e.g., attention-deficit/hyperactivity disorder [ADHD], depression). TheCASI-Anxiety scale is a reliable, 20-item subscale thatcontains items of generalized anxiety, separation anxi-ety, social phobia, panic and specific phobia.25TheCASI-PDD scale includes 12 items on social deficits,communication problems, and repetitive behaviors.

Children’s Yale–Brown Obsessive Compulsive Scale Modi-fied for Autism Spectrum Disorder (CYBOCS-ASD)7. The5-item CYBOCS-ASD is a clinician-rated measuredesigned to document the current severity of repetitivebehavior in children with ASD. The semistructuredinterview uses the parent as the primary informant, butchildren are encouraged to participate to the extentpossible.7 The interview includes a Symptom Checklistof possible repetitive behaviors grouped into 8 cate-gories (e.g., Washing Rituals, Checking Behaviors,Hoarding). Raters were trained to mark currentsymptoms (past week) on the checklist. The absence ofa mark indicated that the behavior was not present.Each category also permitted the interviewer to writein “other” behaviors presumed relevant to that overallcategory. For example, in the section on RepeatingRituals, a query about spinning objects might promptthe parent to report “he flips a pencil in front of hiseyes.” Once current repetitive behaviors are identified,the interviewer inquires about Time Spent, Interferencein everyday life, Distress (if prevented from performingthe behavior), Resistance (child’s own effort to limit thebehaviors), and degree of Control over the behavior

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATR

VOLUME 53 NUMBER 1 JANUARY 2014

when instructed by parents to stop the behavior. Eachseverity item is anchored to the past week and scoredfrom 0 (not present) to 4 (severe) for a total score of0 to 20.

Rater training included instruction on how to ascer-tain and document checklist items, identify currenttarget symptoms, and establish current severity. Twoto 3 video recordings were used to illustrate interviewmethods and scoring. Raters were then asked to score 3videos independently. Reliable raters had to be within15% of the gold standard rater on the total score.

Y

www.jaacap.org 99

SCAHILL et al.

Classification of Added ItemsAll handwritten items added to the checklist for eachchild were entered verbatim and reviewed by a 2-personpanel (A.D., L.S.). Similar items were collapsed to re-duce item count and avoid duplication. The resultingitem list included 39 items (Table 2). Most items werefrom the original checklist, some were combinations oforiginal items (e.g., excessive or ritualized hand washingwas merged with excessive or ritualized showering,

TABLE 2 Children’s Yale–Brown Obsessive Compulsive ScaleASD) Symptom Checklist Items Before Principal Component A

Checklist Category Che

Washing/cleaning Excessive or ritualized handExcessive cleaning of items (eRepetitive water playOther measures to prevent/re

Checking Checking locks, toys, other itChecking associated with geChecking that did not/will noChecking that did not/will noChecking that nothing terribleChecking that did not make mChecking tied to somatic obs

Repeating Rereading (e.g., telephone bNeed to repeat routine activiTouching in patternsBody stereotypya

Object stereotypya

EcholaliaHand, finger stereotypya

Repetitive behavior involvingRestricted interests (i.e., extenwatching same video over

Counting Counting objects, certain numOrdering/arranging Lining up objects or evening

Ordering, arranging (insistenfurniture in assigned place)

Hoarding/saving Difficulty throwing things awaExcessive games/superstitiousbehavior

Excessive games/superstitiou

Rituals involving another person Need to involve another persRepetitive requests or demanDwelling on a preferred topicInsistent on routines in everyd

Miscellaneous Mental ritualsb

Need to tell, ask, or confessMeasures to prevent harm toRitualized eating behaviorsRepetitive sexual behaviorExcessive list makingb

Rituals involving blinking or sTrichotillomania, skin pickingSelf-injurious behaviorPicaa,b

Note: aCompiled items are combinations of the original items or newly createbItems endorsed by less than 5% of subjects dropped for the Principal Com

JOURN

100 www.jaacap.org

bathing, teeth brushing, grooming), and others werebased on the handwritten items (e.g., watching the samevideo segment over and over).

Analytic PlanFirst, we examined the clinical characteristics of theavailable sample from the RUPP Autism Network trialsand the frequency of symptom endorsement on theCYBOCS checklist. Endorsement was calculated as the

for Children With Autism Spectrum Disorder (CYBOCS-nalyses

cklist Items Endorsement, n (%)

washing, showering, bathinga 22 (8.1).g., personal clothing)b 13 (4.8)

92 (33.8)move contaminantsb 7 (2.6)ems, etc. 49 (18)tting washed, dressed, or undressedb 11 (4)t harm othersb 1 (0.4)t harm selfb 0 (0)did/will happenb 7 (2.6)istakeb 6 (2.2)

essionsb 3 (1.1)ook, train schedule) or rewriting 35 (12.9)ties 38 (14)

27 (9.9)168 (61.8)75 (27.6)

133 (48.9)40 (14.7)

sensationa 43 (15.8)ded play with preferred object,and over) a

79 (29)

bers, words, etc. 22 (8.1)up 132 (48.5)t on placement of toys/objects/a

19 (7)

y, saving bits of paper, string 66 (24.3)s behaviorb 9 (3.3)

on (usually a parent) in ritual 129 (47.4)dsb 2 (0.7)(not echolalia)a,b 12 (4.4)ay situationsa 56 (20.6)

2 (0.7)29 (10.7)

self; or terrible consequencesb 0 (0)66 (24.3)41 (15.1)2 (0.7)

taring 32 (11.8)48 (17.6)54 (19.9)1 (0.4)

d items based on the “other” items reported.ponent Analysis (PCA).

AL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

VOLUME 53 NUMBER 1 JANUARY 2014

CYBOCS-ASD COMPONENT STRUCTURE

number of subjects with that symptom over the totalnumber of subjects.

Principal components analysis (PCA) was used toexamine the dimensionality of CYBOCS checklistitems. Given that the original CYBOCS checklist wasdeveloped for OCD, the PCA proceeded without as-sumptions about the latent structure of the binary(present or absent) checklist items in children withASD. To decrease the likelihood of extracting spuriouscomponents, we dropped items endorsed by less than5% of the sample.26 The number of componentsextracted was determined to avoid retaining too few(underextraction) or too many (overextraction) com-ponents and to verify sample adequacy.27,28 Ratherthan relying on any single criterion (e.g., eigenvalues>1) for determining the number of components toextract, we considered multiple criteria as follows: thecumulative proportion of item covariance extracted(and the related Scree plot); whether the retainedcomponents were well saturated; and whether thecomponents were well identified.28 An orthogonal(varimax) rotation was applied initially and the reli-ability of the structure was verified using alternateorthogonal (equamax) and oblique (promax) rotations.Rotated item loadings of �0.3 (moderate effect) wereconsidered salient for interpreting the components,which were subsequently confirmed by content experts(L.S., A.D.). Scores on components were computed foreach participant on the basis of all item loadings.

In the third set of analysis, we examined theconstruct validity of the identified components. Theseanalyses evaluated the association between componentscores and scores on CYBOCS severity scales (TimeSpent, Interference, Distress, Resistance, and Control)using Spearman correlation. We also examined theassociation of component scores with selected sub-scales on the CASI, ABC subscales, and Vinelandscores with Pearson correlation. Discriminant functionanalyses were conducted to determine whether iden-tified components could differentiate participantsbased on trial-specific treatment targets (hyperactivityin the methylphenidate trial or serious behavioralproblems in the risperidone trials), intelligence (IQ �70or IQ <70), or verbal ability (verbal or nonverbal itemon the Autism Diagnostic Interview–Revised).

The final set of analysis examined the distributionsof the CYBOCS-ASD dimensions (Time Spent, Inter-ference, Distress, Resistance, and Control) and com-puted the internal consistency (Cronbach’s a) for all5 dimensions and with each dimension removed.Pearson correlations were calculated for total CYBOCS-ASD scores with ABC subscales, selected CASI sub-scales, and Vineland score to estimate divergentvalidity.

RESULTSTable 1 summarizes the clinical characteristics ofthe sample. Nineteen subjects were unable to

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATR

VOLUME 53 NUMBER 1 JANUARY 2014

cooperate with testing and did not have an IQscore. Of those with IQ data, 141 (51.8%) wereclassified as intellectually disabled (IQ < 70).Based on parent-report, the sample was 72.4%white, 12.5% African American, 7.4% Hispanic,5.1% Asian, and 2.6% “other.”

Fifteen checklist items were endorsed for lessthan 5% of subjects and were dropped (Table 2).As shown in Table 3, the rate of endorsement forthe remaining 24 items ranged in frequency from7% (ordering, putting objects into proper place) to62% (body stereotypy).

Principal Components AnalysisTwenty-four items were initially included in thePCA; however, 1 item (excessive ritualizedhand washing, showering) did not reach ourthreshold (�0.30) on any component and wasdropped. Twenty-three checklist items wereincluded in a second PCA (16 original items and7 new items). The median endorsement rate forthese 23 items was 18%. This frequency is 11points above the minimum and 20% of the55-point range (7%–62%), suggesting that pair-wise disparities in marginal frequencies werenot extreme and mitigating concerns regardingbinary item data. A 5-component solution pro-vided the best classification of repetitive be-haviors (accounting for 34.4% of the variance).Both 4- and 6-component solutions were con-sidered. The 4-component solution accountedfor less variance, and the 6-component solutionresulted in a greater number of items loading onmore than 1 component.

Table 3 displays the varimax rotatedcomponent structure, which was consistentacross the alternative equamax and promaxrotations. The 5-component model yielded adifferent structure than the original CYBOCScategories. Component I included Hoardingand Ritualistic Behavior (e.g., counting, ritual-ized eating, and checking); Component IIinvolved Sensory/Motor (e.g., echolalia, repet-itive water play) and Arranging Behaviors (e.g.,ordering or lining up objects); ComponentIII included Insistence on Sameness andSelf-Injurious Behaviors (SIB); Component IVcomprised Stereotypy; and Component V ca-ptured Restricted Interests. Object Stereotypy(e.g., repetitive manipulation of a pencil, straw,piece of string) loaded on Components III(Sameness) and IV (Stereotypy). Given theminimal difference on loadings of object ste-reotypy for Components III and IV, we placed

Y

www.jaacap.org 101

TABLE 3 Varimax Rotated Component Structure of Repetitive Behaviors on the Children’s Yale–Brown ObsessiveCompulsive Scale for Children With Autism Spectrum Disorder Symptom Checklist (N ¼ 272)

Checklist Items Frequency

I Hoarding andRitualisticBehavior

II Sensory andArranging

IIISameness/

SIBIV

StereotypyV RestrictedInterest

Difficulty throwing things away, savingbits of paper, string

24.3 0.536a �0.057 �0.054 �.204 .322

Counts objects, certain numbers,words, etc.

8.1 0.527 0.027 0.022 .048 �.076

Checks locks, toys, books/items, etc. 18.0 0.520 �0.173 0.247 .129 .081Ritualized eating behaviors 24.3 0.513 0.047 �0.025 �.033 �.209Trichotillomania (hair pulling), skinpicking

17.6 0.389 0.111 0.110 .034 .321

Ordering, arranging (placing toys/objects/furniture in assigned place) a

7.0 �0.213 0.639 0.210 �0.043 0.043

Echolalia 48.9 0.293 0.581 �0.076 �0.101 0.001Repetitive behavior involvingsensationa

15.8 �0.048 0.473 �0.094 �0.118 0.261

Repetitive water play 33.8 0.088 0.454 �0.027 0.262 �0.224Repetitive sexual behavior 15.1 �0.076 0.350 0.373 0.231 0.156Rereading or rewriting 12.9 0.119 �0.066 0.607 �0.133 �0.234Insisting on routinesa 20.6 �0.011 0.004 0.563 �0.104 0.214Self-injurious behavior 19.9 0.206 0.280 0.442 0.111 �0.011Body stereotypya 61.8 �0.078 0.247 L0.371 0.125 �0.101Object stereotypya 27.6 �0.140 0.102 0.340 0.383 0.040Need to repeat routine activities 14.0 0.213 0.087 �0.031 0.483 0.123Hand/finger stereotypya 14.7 0.053 �0.083 �0.183 0.349 0.054Rituals involving blinking or staring 11.8 0.307 �0.032 0.092 0.366 0.193Need to involve another person inritual

47.4 0.379 0.183 0.009 L0.591 0.093

Need to tell, ask, or confess 10.7 0.052 �0.060 0.046 L0.463 0.188Touching in patterns 9.9 0.073 �0.031 0.077 �0.031 0.597Restricted interests (i.e., trains,videos, etc.)a

29.0 0.008 0.157 0.094 0.082 0.348

Need for symmetry or evening up 48.5 0.332 0.122 0.310 0.033 L0.538

Note: Loadings with an absolute value �0.30 are in boldface type. Underlined items load similarly on 2 components. The 5 components accounted for34.4% of the variance.aCompiled items.

SCAHILL et al.

this item on both components. A few otheritems showed similar loadings on more than 1Component (Repetitive Sexual Behavior loadedon Component II (Sensory and Arranging) andComponent III (Sameness and SIB). Hair-pulling loaded on Components I and V. Com-ponents I and III, blinking and staring ritualsalso loaded on Components I and IV. Bycontrast, there was a higher loading for hoard-ing on Component I compared to Component V(Restricted Interests), suggesting that hoardingbelongs in Component I. It is possible thatfascination with a specific object (ComponentV) drives the reluctance to discard the object,rather than indicating hoarding as a generalproblem.

JOURN

102 www.jaacap.org

Association of the 5 Components With OtherMeasuresBased on parent report, many subjects exhibitedrepetitive behavior in more than 1 component.Standardized scores (mean ¼ 0, SD ¼ 1) on eachof the 5 components were computed for eachparticipant. A higher component score for a givensubject reflects a greater likelihood of exhibitingthat type of repetitive behavior. Table 4 presentscorrelations between component scores and ageas well as scores on several clinical measures.

Subjects ranked with higher scores on theCYBOCS-ASD Distress dimension were higheron Component I (Ritualized behavior) (Spearmanrho ¼ 0.31, p < .003) (Table 4). Scores on the CASIAnxiety scale were associated with scores on

AL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

VOLUME 53 NUMBER 1 JANUARY 2014

TABLE 4 Correlations Between Principal Component Scores and Scores on Clinical Measures

I Hoarding/RitualisticBehavior

II Sensory andArranging

IIISameness/ SIB IV Stereotypy V Preoccupation

Age 0.03 0.05 �0.02 �0.12 0.23CYBOCS-ASD Total Severitya 0.28c 0.27c 0.04 0.09 �0.07

Time spent 0.23c 0.25c �0.02 0.13 �0.01Interference 0.18 0.26c 0.02 0.10 �0.05Distress 0.31c 0.16 0.02 0.06 �0.15Resistance 0.10 0.11 0.05 �0.05 0.02Control 0.18 0.10 0.06 0.03 0.02

CASIAnxiety scale 0.30c 0.06 �0.05 �0.06 0.02PDD scale 0.03 0.33c 0.01 0.18 �0.01

Vineland Adaptive BehaviorScale b

Communication 0.09 �0.26c 0.03 �0.31c �0.06Socialization �0.01 �0.21c �0.07 �0.15 �0.11Daily living 0.02 �0.21c �0.01 �0.17 �0.10

Aberrant Behavior ChecklistIrritability 0.21c 0.14 0.07 0.20c �0.02Social withdrawal 0.08 0.21c 0.01 0.14 �0.04Stereotypy �0.05 0.28c �0.16 0.24c �0.01Hyperactivity 0.07 0.07 �0.06 0.04 �0.07Inappropriate speech 0.17 0.13 �0.04 �0.13 �0.02

Note: CASI ¼ Child and Adolescent Symptom Inventory; CYBOCS-ASD ¼ Children’s YaleeBrown Obsessive Compulsive Scale for children with autismspectrum disorder; PDD ¼ pervasive developmental disorder; SIB ¼ self-injurious behavior.aSpearman, all others Pearson p < .003.bHigher Vineland scores reflect better adaptive functioning, hence the negative correlations.cp < .003

CYBOCS-ASD COMPONENT STRUCTURE

Component I (Pearson r ¼ 0.30, p < .003). Subjectswith higher scores on Component II scoredhigher on the CASI-PDD scale (r ¼ 0.33, p < .003).Although significant, these small to mediumcorrelations suggest that components are notstrongly associated with clinical measures.

Discriminant function analyses demonstratedthat the 5 component scores did not discriminateon trial-specific targets (hyperactivity in themethylphenidate trial or serious behavioral prob-lems in risperidone trials; l ¼ 0.964, p ¼ .081).However, component scores differentiated signif-icantly by IQ (l ¼ 0.826, p<.001). Participants withlower IQ had lower scores on Component 1(Hoarding and Ritualistic Behavior) (p ¼ .005) andhigher scores on Component II (Sensory and RuleBased Behavior) and Component IV (Stereotypy)(p < .001 for both). Similarly, these 3 componentscores differentiated among nonverbal versusverbal subjects (based on the Autism DiagnosticInterview–Revised [ADI-R]) (l ¼ 0.854, p < .001).Participants who were nonverbal had lower scoreson Component I (p ¼ .005) and higher scoreson Component II (p ¼ .021) and Component IV

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATR

VOLUME 53 NUMBER 1 JANUARY 2014

(p < .001). Component III (Sameness and SIB) andComponent V (Restricted Interest) were not sig-nificant in any discriminant analyses.

Distribution of Scores on CYBOCS-ASD DimensionsThe Resistance and Control dimensions eachhad a modal score of 4 compared to a modalscore of 3 for the other 3 dimensions (Table 5).The distribution of the Resistance item is partic-ularly distorted, with more than 60% of subjectsrated as extreme (score of 4).

The internal consistency for the 5-itemCYBOCS-ASD Total score was .81. When eachitem was removed, Cronbach’s a ranged from0.76 to 0.79 (see Table S1, available online). TheCYBOCS-ASD total score showed modest cor-relations with measures of adaptive functioningand maladaptive behavior. For example, corre-lations with the ABC subscales for Irritability,Social Withdrawal, Stereotypy, and Hyperac-tivity ranged from 0.25 to 0.32 (p < .01). Simi-larly, correlations between the CYBOCS-ASDtotal score and the Vineland (higher scoresreflect higher functioning) were also modest

Y

www.jaacap.org 103

TABLE 5 Item Mean, Standard Deviation, and Frequency Distribution of Children’s Yale–Brown Obsessive CompulsiveScale for Children With Autism Spectrum Disorder (CYBOCS-ASD) Severity Dimensions (N ¼ 272)

Time Spent Interference Distress Resistance Control

Mean (SD) 2.86 (0.94) 2.43 (0.91) 2.73 (0.86) 3.45 (0.79) 3.32 (0.75)Frequency distribution, n (%)

0 (None) 2 (0.7) 7 (2.6) 6 (2.2) 2 (0.7) 2 (0.7)1 (Mild) 24 (8.8) 33 (12.1) 18 (6.6) 3 (1.1) 3 (1.1)2 (Moderate) 58 (21.3) 92 (33.8) 57 (21.0) 30 (11.0) 26 (9.6)3 (severe) 115 (42.3) 115 (42.3) 154 (56.6) 73 (26.8) 116 (42.6)4 (Extreme) 73 (26.8) 25 (9.2) 37 (13.6) 164 (60.3) 125 (46.0)

SCAHILL et al.

(�0.26 to �0.30, p < .01 for all 3 domains) (seeTable S2, available online). These small to me-dium correlations suggest that the CYBOCS-ASDis measuring a separate construct from theseother instruments.

DISCUSSIONIn its current form, the CYBOCS-ASD is areliable and valid measure of repetitivebehavior in youth with ASD.7 The results of thecurrent study show that several originalCYBOCS checklist items are not relevant forchildren with ASD and can be dropped. A totalof 23 checklist items endorsed by at least 5%of this sample fell into 5 components. These5 clinically plausible components explainednearly 35% of the variance. For example,Component IV (Stereotypy) includes motorbehaviors such as repetitive object manipula-tion, hand and finger stereotypy, blinking andstaring rituals, and repetition of routine activ-ities. In this sample, subjects often exhibitedbehaviors in more than 1 component, suggest-ing that typologies of repetitive behavior inASD may not be strictly drawn. By contrast,however, the discriminant function analysisshowed that children with verbal capacity (asreported on the ADI-R) and those with IQ � 70were more likely to exhibit counting, checking,and ritualized eating behaviors (Component I)than were nonverbal children or those withintellectual disability. Lower-functioning chil-dren were more likely to engage in orderingand arranging, hand and finger stereotypy, andobject manipulation (Components II and IV).Collectively, these findings support the formatof the CYBOCS-ASD, which yields a total scorebased on the impact of the symptoms presentrather than subscale scores for types of repetitivebehavior. A reliable and valid clinician-ratedmeasure with a single total score has obvious

JOURN

104 www.jaacap.org

statistical advantages as an endpoint in clinicaltrials.

The components identified in this analysisshow differences and similarities to previous in-vestigations. The International Classification ofDisease–Tenth edition (ICD-10) describes 4 cate-gories of repetitive behavior in ASD, as follows:preoccupations with part-objects; stereotypedmotor mannerisms; circumscribed patterns ofinterest; and adherence to specific routines orrituals.29 Several prior factor analyses based onthe 10 to 12 repetitive behavior items from theADI-R reported 2 factors, namely, stereotypy andinsistence on sameness.30 This is consistent withthe traditional concepts of ”lower”- and “higher”-order behaviors.9 However, circumscribed in-terests were not included in most of these ADI-Rfactor analyses. One study that did include cir-cumscribed interest showed that it emerged as athird factor, which is consistent with ICD-10.11

The limited number of repetitive behaviors onthe ADI-R may have hindered the identificationof more than 2 factors.

Honey et al.12 examined the factor structure ofthe 26-item, parent-rated Repetitive BehaviorQuestionnaire (RBQ)9 in 180 children and ado-lescents with ASD who were aged 3 to 16 years.Participants with self-injurious behavior (SIB)were excluded from the analysis. Two factorsemerged, namely, Sensory/Motor behaviors andInsistence on Sameness/ Circumscribed Interests.This classification merges the 4 ICD-10 categoriesinto 2. Preoccupations with part-objects and ste-reotyped motor mannerisms formed a singleSensory/Motor factor; circumscribed patterns ofinterest and adherence to routines became a sin-gle Sameness/Circumscribed Interest factor. This2-factor solution may have missed the distinctionbetween sameness and circumscribed interestsreported by Lam et al. and the findings in thepresent study.11

AL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

VOLUME 53 NUMBER 1 JANUARY 2014

CYBOCS-ASD COMPONENT STRUCTURE

Five factors have been identified for theparent-rated Repetitive Behavior Scale–Revised(RBSR) developed by Bodfish et al.10,31,32 Thesefactors do not precisely correspond with ICD-10.For example, it includes an SIB factor that is notlisted in ICD-10. As in ICD-10, the RBSR hasseparate factors for stereotypic behavior, insis-tence on sameness, and circumscribed interests.The fifth factor included ordering and arrangingand the need to touch or tap things is similar toour Component II, but is not clearly enumeratedin ICD-10.

The 5-component structure identified in thecurrent study included a larger set of behaviorsthan the ADI-R. Unlike the study of the RBQby Honey et al.,12 our study included SIB.Component III in our analysis includes insistenceon routines and SIB, which does not fit neatlywith the concept of higher- and lower-order re-petitive behaviors9 but which has been reportedby others.33 Although SIB is multidetermined, itmay be that some children resort to SIB whenroutines are not followed. Consistent with theconcept of lower-order repetitive behaviors,Components II (Sensory Motor) and IV (Stereo-typy) in this study were associated withnonverbal and intellectually disabled subjects.However, these 2 components emerged asdistinct in our analysis, rather than the singlefactor reported by others.

Taken together, repetitive behaviors exhibitedby children with ASD appear to cluster in pat-terns that are collapsed in some analyses andmore finely delineated in others. Other seemingdiscrepancies in factor analysis may be due todifferences in sample characteristics, sample size,as well as differences in the coverage of behav-iors. Children with ASD often exhibit behaviorsfrom more than 1 category, suggesting that re-petitive behaviors are not factor specific. Theassociation of motor stereotypy and sensory-motor behavior with intellectual disability andthe association of more complex ritualisticbehavior with IQ � 70 are replicated findings.10,11

The implications for treatment are not clear.Consistent with the results of our previous

report, this study raises questions about the rele-vance of the Resistance item in the ASD popula-tion.7 In OCD, resistance may be a sign of health.Patients who resist the urge to carry out ritualizedbehavior are considered less severe on thisdimension than those who do not resist.34 Giventhat individuals with ASD often engage in repeti-tive behavior as a preferred activity, resistance is

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATR

VOLUME 53 NUMBER 1 JANUARY 2014

less informative. Indeed, more than 60% of thesubjects in our sample had a Resistance score of 4,reflecting no resistance. Whether the Resistanceitem score changes with treatment is an unan-swered empirical question. If the Resistance item isdropped, additional severity items such as in-tensity of the behavior or the unusual appearanceof the behavior could be developed and tested toretain the scoring range from 0 to 20.

Our results underscore other differences inrepetitive behavior in ASD compared to OCD.Repetitive behaviors in OCD such as excessivehand washing and rituals to prevent harmobserved in children with OCD were rarelyendorsed in this sample of children with ASD. Inpractice, attention to these distinctions may helpwith differential diagnosis. The phenomenolog-ical dissimilarities in OCD and ASD may explainthe differential response to the selective serotoninreuptake inhibitors.35

The results of this study provide an incre-mental refinement on measuring repetitivebehavior in children with ASD. However, thereare several limitations of this study. First, this wasa sample of convenience, consisting of childrenselected for participation in multisite clinical tri-als. The sample may not be representative of allchildren with ASD. Second, the symptom check-list included a wide range of repetitive behaviorsgenerated by an expert panel and compulsionsfrom the original CYBOCS. Nonetheless, we mayhave missed some repetitive behaviors in chil-dren with ASD. Future research could explorewhether the addition of relevant repetitive be-haviors affects the component structure. Third,we used the simple exploratory model (PCA) inthis study, without assumptions about the latentstructure of the binary (present or absent)checklist items. Future studies could proceedwith confirmatory analysis to verify this compo-nent structure in youth with ASD. Fourth, theweights used to generate component scores,which were then correlated with other measures,are sample specific. These associations maychange in magnitude in a different sample. &

Y

Accepted October 14, 2013.

Drs. Scahill and Bearss are with Emory University. Dr. Dimitropoulosis with Case Western Reserve University. Dr. McDougle is withHarvard University. Drs. Aman, Lecavalier and Arnold are with OhioState University. Dr. Feurer is with Vanderbilt University School ofMedicine. Dr. McCracken is with University of California at LosAngeles. Dr. Tierney is with Johns Hopkins University. Dr. Pu is withthe University of Arizona. Dr. White is with Virginia Polytechnic

www.jaacap.org 105

10

SCAHILL et al.

University. Dr. Hallett is with Kings College in London. Dr. King iswith the University of Washington. Dr. Vitiello is with the NationalInstitute of Mental Health (NIMH).

This work was funded by NIMH by the following Research Units onPediatric Psychopharmacology (RUPP) grants: Yale, U10MH66764;Indiana University, U10MH66766; and Ohio State University,U10MH66768. This publication was also supported by theYale CTSA, UL1 RR024139, IU CTSA UL1 RR025761, OSU CTSAUL1 RR025755 from the National Center for Research Resources(NCRR).

Drs. Feurer, Pu, Dimitropoulos, and Scahill served as the statisticalexperts for this research.

Drs. Dimitropoulos, Feurer, Pu, and Scahill had full access to all datain the study and take responsibility for the integrity of the data and theaccuracy of the data analysis.

The opinions and assertions contained in this report are the privateviews of the authors and are not to be construed as official or asreflecting the views of the Department of Health and Human Services,the National Institutes of Health, or NIMH.

The authors acknowledge Wayne Goodman, MD, of Mt SinaiHospital; Lawrence Prince, MD of Brown University Ann Wagner,PhD, and Louse Ritz, MBA, of NIMH; James Dziura, PhD, Yan-hong Deng, MPH, and Allison Gavaletz, BA, of Yale; andNaomi Swiezy, PhD, and Jennifer E. Mullett, RN, of IndianaUniversity.

JOURNAL

6 www.jaacap.org

Disclosure: Dr. Scahill has served as a consultant to Bracket, BioMarin,and Coronado; has received research support from Shire, Roche, andPfizer, and has served on the speakers’ bureau of the Tourette SyndromAssociation. Dr. Aman has received research contracts, consulted with,or served on advisory boards of Biomarin Pharmaceuticals, Bristol-MyersSquibb, Confluence Pharmaceutica, Coronado Bioscience, ForestResearch, Hoffman LaRoche, Johnson and Johnson, Novartis, Pfizer,and Supernus Pharmaceutica. Dr. McCracken has served as a consul-tant to Roche, Novartis, BioMarin, and PharmaNet; and has receivedresearch support from Seaside Therapeutics, Roche, and Otsuka.Dr. Tierney has served as a consultant to BioMarin. Dr. King has servedas a consultant to and has received research support from SeasideTherapeutics and Roche. Dr. Arnold has served on the advisory boardsof Biomarin, Novartis, Noven, Roche, and Seaside Therapeutics; hasreceived research funding from CureMark, Forest, Eli Lilly and Co., andShire; has received travel support from Noven; and has served as aconsultant to Tris Pharma. Drs. Dimitropoulos, McDougle, Feurer, Pu,White, Lecavalier, Hallett, Bearss, and Vitiello report no biomedicalfinancial interests or potential conflicts of interest.

Correspondence to Lawrence Scahill, MSN, PhD, Marcus Center;1920 Briarcliff Road, Atlanta, GA 30329; e-mail: lawrence.scahill@emory.edu

0890-8567/$36.00/ª2014 American Academy of Child andAdolescent Psychiatry

http://dx.doi.org/10.1016/j.jaac.2013.09.018

REFERENCES

1. American Psychiatric Association. DSM5: Diagnostic and Statisti-

cal Manual of Mental Disorders. Fifth Edition. Washington, DC:American Psychiatric Association; 2013.

2. Centers for Disease Control and Prevention. Prevalence of autismspectrum disorders: Autism and Developmental DisabilitiesMonitoring Network, 14 sites, United States, 2008. Morbid MortalWkly Rep Surveill Summ. 2012;61:1-18.

3. Fombonne E. The epidemiology of pervasive developmental dis-orders. Pediatr Res. 2009;65:591-598.

4. Leckman JF, Bloch MH, Scahill L, King RA. Tourette syndrome:the self under siege. J Child Neurology. 2006;21:642-649.

5. Mataix-Cols D, Nakatani E, Micali N, Heyman I. Structure ofobsessive-compulsive symptoms in pediatric OCD. J Am AcadChild Adolesc Psychiatry. 2008;47:773-778.

6. Rosario-Campos MC, Miguel EC, Quatrano S, et al. The Dimen-sional Yale–Brown Obsessive-Compulsive Scale (DY-BOCS): aninstrument for assessing obsessive-compulsive symptom di-mensions. Mol Psychiatry. 2006;11:495-504.

7. Scahill L, McDougle CJ, Williams SK, et al. The Children’s Yale–Brown Obsessive Compulsive Scales modified for pervasivedevelopmental disorders. J Am Acad Child Adolesc Psychiatry.2006;45:1114-1123.

8. Zandt F, Prior M, Kyrios M. Repetitive behavior in children withhigh functioning autism and obsessive-compulsive disorder.J Autism Dev Disord. 2007;37:251-259.

9. Turner M. Annotation: Repetitive behaviour in autism: a reviewof psychological research. J Child Psychol Psychiatry. 1999;40:839-849.

10. Bodfish JW, Symons FJ, Parker D, Lewis MH. Varieties of repeti-tive behavior in autism:comparisons to mental retardation.J Autism Dev Disord. 2000;30:237-243.

11. Lam KSL, Bodfish JW, Piven J. Evidence for three subtypes ofrepetitive behavior in autism that differ in familiality and associ-ation with other symptoms. J Child Psychol Psychiatry. 2008;49:1193-1200.

12. Honey E, McConachie H, Turner M, Rodgers J. Validation of theRepetitive Behaviour Questionnaire for use with children withautism spectrumdisorder. Res Autism Spectr Disord. 2012;6:355-364.

13. Turner-Brown LM, Lam KSL, Holtzclaw TN, Dichter GS,Bodfish JW. Phenomenology and measurement of circum-scribed interests in autism spectrum disorders. Autism. 2011;15:437-456.

14. Scahill L, Riddle MA, McSwiggin-Hardin M, et al. Children’s Yale–Brown Obsessive Compulsive Scale: reliability and validity. J AmAcad Child Adolesc Psychiatry. 1997;36:844-852.

15. McDougle CJ, Scahill L, Aman MG, et al. Risperidone for the coresymptom domains of autism: results from the RUPP AutismNetwork Study. Am J Psychiatry. 2005;162:1142-1148.

16. Research Units on Pediatric Psychopharmacology AutismNetwork. Risperidone in children with autism for serious behav-ioral problems. N Engl J Med. 2002;347:314-321.

17. Research Units on Pediatric Psychopharmacology (RUPP) AutismNetwork. Randomized, controlled, crossover trial of methylphe-nidate in pervasive developmental disorder. Arch Gen Psychiatry.2005;62:1266-1274.

18. Aman MG, McDougle CJ, Scahill L, et al. Medication and parenttraining in children with pervasive developmental disorders andserious behavioral problems: results from a randomized clinicaltrial. J Am Acad Child Adolesc Psychiatry. 2009;48:1143-1154.

19. Scahill L, McDougle CJ, Aman MG, et al. Effects of risperidoneand parent training on adaptive functioning in children witha pervasive developmental disorders and serious behavioralproblems. J Am Acad Child Adolesc Psychiatry. 2012;51:136-146.

20. Lord C, Pickles A, McLennan J, et al. Diagnosing autism: analysesof data from the Autism Diagnostic Interview. J Autism DevDisord. 1997;27:501-517.

21. Sparrow SS, Balla DA, Cicchetti DV. Vineland Adaptive BehaviorScales: Survey Form Manual. Circle Pines, MN: American Guid-ance Service; 1984.

22. Aman MG, Singh NN, Stewart AW, Field CJ. Psychometric char-acteristics of the Aberrant Behavior Checklist. Am J Ment Defic.1985;89:492-502.

23. Guy W. ECDEU Assessment Manual for Psychopharmacology.Washington, DC: Department of Health, Education and Welfare,National Institute of Mental Health; 1976.

24. Gadow KD, Sprafkin J. Child Symptom Inventory–4 Screeningand Norms Manual. Stony Brook, NY: Checkmate Plus; 2002.

25. Hallett V, Lecavalier L, Sukhodolsky DG, et al. Exploring themanifestations of anxiety in children with autism spectrum dis-orders. J Autism Dev Disord. 2013;43:2341-2352.

26. Gorsuch RL. Factor Analysis. Hillsdale, NJ: Erlbaum; 1983.27. Wood JM, Tataryn DJ, Gorsuch RL. Effects of under- and over-

extraction on principal axis factor analysis with varimax rotation.Psychol Methods. 1996;1:354-365.

OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

VOLUME 53 NUMBER 1 JANUARY 2014

CYBOCS-ASD COMPONENT STRUCTURE

28. Velicer WF, Fava JL. Effects of variable and subject samplingon factor pattern recovery. Psychol Methods. 1998;3:231-251.

29. World Health Organization. The ICD-10 Classification of Mentaland Behavioural Disorders: Diagnostic Criteria. Geneva: WorldHealth Organization; 2007.

30. Honey E, Rodgers J, McConachie H. Measurement of restrictedand repetitive behaviour in children with autism spectrum dis-order: selecting a questionnaire or interview. Res Autism SpectrDisord. 2012;6:757-776.

31. Lam KSL, Aman MG. The Repetitive Behavior Scale–Revised:independent validation in individuals with autism spectrumdisorders. J Autism Dev Disord. 2007;37:855-866.

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATR

VOLUME 53 NUMBER 1 JANUARY 2014

32. Bishop SL, Hus V, Duncan A, et al. Subcategories of restricted andrepetitive behaviors in children with autism spectrum disorders.J Autism Dev Disord. 2013;43:1287-1297.

33. Duerden EG, Oatley HK, Mak-Fan MK, et al. Risk factors associatedwith self-injurious behaviors in children andadolescentswith autismspectrum disorders. J Autism Dev Disord. 2012;42:2460-2470.

34. Goodman WK, Price LH, Rasmussen SA, et al. Yale–BrownObsessive Compulsive Scale. I. Development, use, and reliability.Arch Gen Psychiatry. 1989;46:1006-1011.

35. King BH, Hollander E, Sikich L, et al; for the STAART Psycho-pharmacology Network. Lack of efficacy of citalopram in childrenwith autism spectrum disorders and high levels of repetitivebehavior. Arch Gen Psychiatry. 2009;66:583-590.

Y

www.jaacap.org 107

TABLE S1 Item Analysis and Average Severity Scores onChildren’s Yale–Brown Obsessive Compulsive Scale forChildren With Autism Spectrum Disorder (CYBOCS-ASD)(N ¼ 272)

Item Mean (SD)a Value With Item

Removeda

Time spent 2.86 (0.94) 0.76Interference 2.43 (0.91) 0.76Distress 2.73 (0.86) 0.76Resistance 3.45 (0.79) 0.79Control 3.32 (0.75) 0.77

Note: aOverall Cronbach’s a ¼ 0.81.

TABLE S2 Pearson Correlations Between Children’sYale–Brown Obsessive Compulsive Scale for ChildrenWith Autism Spectrum Disorder (CYBOCS-ASD) TotalScores for Full Sample (N ¼ 272)

r Mean (SD)

Child and Adolescent SymptomInventory (CASI)Anxiety scale 0.11 14.7 (9.5)PDD scale 0.32* 22.2 (7.6)

Vineland Adaptive Behavior ScalesCommunication �0.27* 54.9 (20.9)Socialization �0.30* 46.0 (20.2)Daily living �0.26* 48.1 (17.4)

Aberrant Behavior ChecklistIrritability 0.32* 25.4 (9.3)Social withdrawal 0.31* 15.0 (8.9)Stereotypy 0.31* 8.7 (5.4)Hyperactivity 0.25* 34.1 (8.7)Inappropriate speech 0.14 5.9 (3.8)

Note: CASI ¼ Child and Adolescent Symptom Inventory; PDD ¼pervasive developmental disorder.*p � .001.

SCAHILL et al.

JOURNAL OF THE AMERICAN ACADEMY OF CHILD & ADOLESCENT PSYCHIATRY

107.e1 www.jaacap.org VOLUME 53 NUMBER 1 JANUARY 2014