Dr. Katherine Holman

Intervention targeting development of sociallysynchronous engagement in toddlers withautism spectrum disorder: a randomized

controlled trial

Rebecca J. Landa,1 Katherine C. Holman,2 Allison H. O’Neill,3 andElizabeth A. Stuart4

1Kennedy Krieger Institute, Center for Autism and Related Disorders, Johns Hopkins University School of Medicine,Department of Psychiatry and Behavioral Sciences, Baltimore, MD, USA; 2Towson University, Department of SpecialEducation, Towson, MD, USA; 3University of Maryland School of Public Health, Department of Epidemiology andBiostatistics, College Park, MD, USA; 4Johns Hopkins Bloomberg School of Public Health, Departments of Mental

Health and Biostatistics, Baltimore, MD, USA

Background: Social and communication impairments are core deficits and prognostic indicators ofautism. We evaluated the impact of supplementing a comprehensive intervention with a curriculumtargeting socially synchronous behavior on social outcomes of toddlers with autism spectrum disorders(ASD). Methods: Fifty toddlers with ASD, ages 21 to 33 months, were randomized to one of two six-month interventions: Interpersonal Synchrony or Non-Interpersonal Synchrony. The interventionsprovided identical intensity (10 hours per week in classroom), student-to-teacher ratio, schedule, home-based parent training (1.5 hours per month), parent education (38 hours), and instructional strategies,except the Interpersonal Synchrony condition provided a supplementary curriculum targeting sociallyengaged imitation, joint attention, and affect sharing; measures of these were primary outcomes.Assessments were conducted pre-intervention, immediately post-intervention, and, to assess mainte-nance, at six-month follow-up. Random effects models were used to examine differences betweengroups over time. Secondary analyses examined gains in expressive language and nonverbal cognition,and time effects during the intervention and follow-up periods. Results: A significant treatment effectwas found for socially engaged imitation (p = .02), with more than doubling (17% to 42%) of imitated actspaired with eye contact in the Interpersonal Synchrony group after the intervention. This skill wasgeneralized to unfamiliar contexts and maintained through follow-up. Similar gains were observed forinitiation of joint attention and shared positive affect, but between-group differences did not reachstatistical significance. A significant time effect was found for all outcomes (p < .001); greatest changeoccurred during the intervention period, particularly in the Interpersonal Synchrony group. Conclu-sions: This is the first ASD randomized trial involving toddlers to identify an active ingredient forenhancing socially engaged imitation. Adding social engagement targets to intervention improves short-term outcome at no additional cost to the intervention. The social, language, and cognitive gains in ourparticipants provide evidence for plasticity of these developmental systems in toddlers with ASD.http://www.clinicaltrials.gov/ct2/show/NCT00106210?term = landa&rank = 3. Keywords: Autisticdisorder, intervention. Abbreviations: IS: Interpersonal Synchrony; IJA: initiation of joint attention;SEI: socially engaged imitation; SPA: shared positive affect; MSEL: Mullen Scales of Early Learning; VR:visual reception; EL: expressive language; ICC: intraclass correlation coefficient; AEPS: Assessment,Evaluation and Programming System.

Autism spectrum disorders (ASDs) are defined by thepresence of social and communication impairmentsthat appear early in life and persist into adulthood(Howlin, Goode, Hutton, & Rutter, 2004; Landa,Holman, & Garrett-Mayer, 2007). Yet many behav-iorally based interventions are associated withimprovement in cognitive, behavioral, and languageimpairments for children with ASD (Rogers &Vismara, 2008), suggesting plasticity in theseaspects of development, particularly when interven-tion is begun early (Drew et al., 2002; Harris &Handelman, 2000). Despite evidence that ASD can

be diagnosed reliably in 2-year-olds (Chawarska,Klin, Paul, & Volkmar, 2007), only a few interventionstudies, mostly with small sample sizes, have beenpublished involving this age group, and mosthave relied on descriptive or quasi-experimentalapproaches. The present study examined the effectsof a hypothesized active ingredient targeting sociallysynchronous behavior within a behavioral inter-vention on outcome measures of socially engagedimitation, joint attention, and affect sharing in2-year-olds with ASD. These core social deficitsinterfere with children’s ability to establish andmaintain synchronous, reciprocal engagement withothers and thus, compromise prognosis.Conflict of interest statement: No conflicts declared.

Journal of Child Psychology and Psychiatry 52:1 (2011), pp 13–21 doi:10.1111/j.1469-7610.2010.02288.x

! 2010 The Authors. Journal of Child Psychology and Psychiatry! 2010 Association for Child and Adolescent Mental Health.Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main St, Malden, MA 02148, USA

Social and communication impairments thatdistinguish ASD from typical development anddevelopmental delay include abnormality in affectexpression, initiation of communication, reciprocityin interaction, social/affective signaling, joint atten-tion, symbolic behavior, motor imitation, languageunderstanding, and conventional use of gestures(Charman et al., 1997; Landa et al., 2007; Rogers,Hepburn, Stackhouse, & Wehner, 2003; Wetherby,Watt, Morgan, & Shumway, 2007). These impair-ments, which emerge early in typical development,formaweak foundation for the ability to establish andmaintain synchronous, reciprocal engagement withothers. Thus, young children with ASDwho elicit onlyinfrequent, brief epochs of engagement with otherslimit opportunities for language and social learning.For this reason, thepresent study focusedoneffects ofdirectly targeting socially engaged imitation, jointattention, and affect sharing.

While social impairments are central to autism,only three peer-reviewed intervention studies identi-fied social improvements in 2-year-olds with ASD. Intheir descriptive study of 20 two-year-olds enrolled ina classroom-based, multi-instructional methodintervention, Stahmer, Ingersoll, and Koegel (2004)identified improvements in some children’s toleranceof proximity to peers, response to others’ initiations,andengagement in reciprocal interactionafter ameanof 9.7months of intervention. In their description of aparent-mediated intervention for young children withASD, Chandler, Christie, Newson, and Prevezer(2002) reported improved social interest in all 18participants, but details were not provided about themagnitude of change nor how it was measured. Inanother parent-mediated intervention, Schertz andOdom (2007) reported improvement in two of threechildren’s response to others’ joint attention bids andinitiation of joint attention.

Two randomized controlled trials (RCTs) involvingvery young children with ASD reported lack of socialimprovement. Drew and colleagues (2002) found nosocial gains after about one year of intervention on aparent interview measure in two groups of nearly2-year-olds enrolled in a parent training (plus localservices) or a local services only group (Drew et al.,2002). Dawson and colleagues (2009) reported anegative change in Vineland Adaptive BehaviorScales Socialization domain standard score (Spar-row, Balla, & Cicchetti, 1984) and lack of improve-ment in severity scores from the Autism DiagnosticObservation Schedule (ADOS; Lord, Rutter, DiLa-vore, & Risi, 1999) after two years of intervention in23-month-olds receiving either the Early DenverModel (20 hours per week of 1:1 therapist-mediatedin-home intervention) or local services.

The mixed evidence for social improvement inyoung children with ASD highlights the need forresearch to determine whether toddlers with ASDcan show gains in core social deficit domains such associally engaged imitation, joint attention and affect

sharing, and to examine the stability of these gainspost-intervention. Perhaps other aspects of devel-opment, such as language and nonverbal cognition,are more plastic than social functioning in earlydevelopment of children with ASD. This possibility issupported by reports of expressive language gains inboth RCTs cited above (Dawson et al., 2009; Drew etal., 2002) and of modest gains in nonverbal cognitionreported by Dawson and colleagues (2009).

The preliminary evidence of social and languageimprovement following early intervention for ASDhighlights the importance of designing early thera-peutic opportunities to practice joint processing ofself–other experiences within social communicativeexchanges. According to Mundy, Sullivan, andMastergeorge (2009), joint processing of one’s ownand others’ actions and objects of attention gives wayto an automaticity that is likely fundamental tosymbolic and social cognitive learning. Thus, earlyintervention that directly targets the development ofsocially engaged imitation, joint attention and affectsharing could have potent effects on later language,play and social outcomes (Bono, Daley, & Sigman,2004; Sigman & Ruskin, 1999; Toth, Munson,Meltzoff, & Dawson, 2006).

The present RCT examined the effects of a hypoth-esized active intervention ingredient targeting socialdevelopment in toddlers with autism. We comparedtwo groups receiving identical intervention except onegroup received a supplemental social curriculum.Theprimary question addressed was: is there a differencein outcome measures of socially engaged imitation,joint attention, and shared positive affect in childrenreceiving this supplementary curriculum? Secondaryquestions addressedwere: (1) do children who receivethe supplementary social curriculum show differen-tial growth in expressive language and nonverbalcognitive functioning; and (2) do children maintainthe gains established during intervention throughouta six-month follow-up period?

Methods

This study was registered with a Data Safety andMonitoring Board at the National Institutes of Healthand was approved by the Johns Hopkins MedicalInstitutional Review Board; all families gave writteninformed consent for their child’s participation.

Participants

Fifty 2-year-olds diagnosed with ASD were randomizedto one of the two intervention conditions. A supple-mentary social curriculum was implemented in theInterpersonal Synchrony (IS) condition, but not withinthe comparison Non-Interpersonal Synchrony (Non-IS)condition. Analyses were conducted on a final sample of48 children (IS n = 24; Non-IS n = 24). One child waswithdrawn by his parents to enroll in a program forchildren with typical development and one child wasmissing a primary baseline variable (Figure 1).

14 Rebecca J. Landa et al.

! 2010 The AuthorsJournal of Child Psychology and Psychiatry ! 2010 Association for Child and Adolescent Mental Health.

Participants were recruited through Kennedy KriegerInstitute’s Center for Autism and Related Disorders,advocacy groups, conferences, Infants and Toddlersprograms, physician’s offices, and word-of-mouth. Eli-gibility criteria included: meeting criteria on the ADOSfor ASD or autism and receiving a diagnosis of ASD byan expert clinician; chronological age between 21 and33 months; non-verbal mental age of at least 8 months(per Mullen Scales of Early Learning [MSEL; Mullen,1995] Visual Reception [VR] scale); having no siblingswith ASD (to avoid confounds introduced by someparents having more skills in treatment delivery withinthe home); primary language spoken within the homewas English; and no known etiology for ASD. Inter-raterreliability of diagnosis was examined for 9 (18%) cases;agreement was 100%. Type and intensity of otherinterventions received by participants were docu-mented via parent report during monthly home visits.

Prior to randomization, children were placed intomatched pairs based on MSEL Receptive Language andVR T scores and ADOS Social Interaction algorithmscores. Within each pair, one child was randomlyselected for the IS group. Table 1 shows that there wereno group differences in socioeconomic status(Hollingshead, 1975) scores (p = .741), proportion ofnon-Caucasian participants (p = .94), or amount ofoutside speech–language treatment during the inter-vention (p = .268) or follow-up period (p = .696). Therewas no correlation between number of hours of outside

speech–language treatment and change in MSELReceptive or Expressive Language T score from pre- topost-test (r = –.233, r = –.236, respectively).

Intervention procedures

Intervention was provided within a Kennedy Kriegerclassroom four days per week for 2.5 hours per day for

Assessed for eligibility (n = 69)

Excluded(n = 19)

Not meeting inclusion criteria (n = 19) Refused to participate (n = 0) Other reasons (n = 0)

Analyzed (n = 24) Excluded from analysis (n = 0)

Lost to follow-up(n = 1; referenced below) Discontinued intervention (n = 1)

Allocated to intervention (IS) (n = 25)

Received allocated intervention (n = 24) Did not receive allocated intervention (n = 1)

Lost to follow-up(n = 0) Discontinued intervention (n = 0)

Allocated to intervention (Non-IS) (n = 25)

Received allocated intervention (n = 25) Did not receive allocated intervention (n = 0)

Analyzed (n = 24) Excluded from analysis (n = 1; missing a baseline measure)

Allocation

Analysis

Follow-Up

Enrollment

Randomized

Figure 1 Consort table flowchart

Table 1 Subject characteristics at start of intervention forInterpersonal Synchrony (IS) and Non-Interpersonal Syn-chrony (Non-IS) groups

Characteristic

IS Group(n = 24)

Non-IS Group(n = 24)

p-valueDemographic, n (%)

Male 20 (83.3) 20 (80) 1.0Caucasian 19 (79.2) 19 (79.2) 1.0Chronological ageat start of tx(mos.), M (sd)

28.6 (2.6) 28.8 (2.8) .721

Hollingshead SESscore, M (sd)

54.7 (8.7) 53.3 (10.3) .663

# hrs of SpeechLanguage tx(pre to post)

24.45 (19.38) 21.38 (16.20) .561

# hrs of SpeechLanguage tx(post to follow-up)

28.07 (27.01) 26.26 (18.82) .822

RCT of social intervention for toddlers with ASD 15


six months. The IS and Non-IS groups received a mean(standard deviation) of 205.66 (18.63) and 196.21(28.4) hours of classroom-based intervention, respec-tively which was not significantly different (p = .17).Student to interventionist ratio was 5:3. One inter-ventionist was employed in both classrooms and wasinstructed not to discuss differences with the otherinterventionists.

Instructional strategies for both conditions repre-sented a continuum of adult-imposed structure rangingfrom discrete trial teaching (Lovaas, 1987) to pivotalresponse training (Koegel et al., 1989) to routines-basedinteractions. Low-tech augmentative communicationsystems were used as needed. Visual cues (Carr,Binkoff, Kologinsky, & Eddy, 1978) and visually basedorganizational strategies (Lord, Bristol, & Schopler,1993) were provided.

Within both conditions, highly motivating interven-tion tasks, materials, and natural consequences weredesigned to elicit frequent child-initiated intentionalcommunication and diverse object play, where inter-ventionists followed the child’s attentional lead andprovided expansions of language and play behavior.Interventionists made relevant cues (e.g., for languagereferents) salient through environmental arrangements,use of brief utterances, and use of linguistic mappingduring child attention and engagement with objects andpeople.

The curriculum used within both conditions was theAssessment, Evaluation, and Programming System forInfants and Children (AEPS; Bricker, 2002), a compre-hensive developmental curriculum that providesguidelines for complexity of intervention goals based onchildren’s individual developmental profile. While theAEPS included goals for joint attention and imitation,the IS group received many more orchestrated oppor-tunities to respond to and initiate joint attention, imi-tate others during social interaction, and share positiveaffect, with interventionists enticing and modelingsocial targets, prompting as needed. More discretebreakdown of social targets was presented within the ISthan the AEPS curriculum. The IS physical learningenvironment presented frequent ecologically validopportunities for initiating and responding to jointattention to proximal and distal objects, people, andevents (e.g., placing thematically relevant objects orpictures on walls; creating socially enticing surpriseevents), and for sharing positive affect (e.g., activitiesinvolving imitation of peers and adults performing sillyand engaging actions with objects).

Identical parent education classes (38 hours) focus-ing on strategies for enhancing child development(using simple language, imitating child’s actions withtoys, establishing joint action routines), coping,behavior management strategies, and advocacy wereprovided in both groups. Home-based parent trainingsessions (1.5 hours per month for six months) focusedon strategies for improving communication, adaptive,and self-regulation skills within daily living activities.These strategies were modeled by the interventionist,then parents were coached while implementing thestrategies with their children. Parents observed theclassroom twice weekly. Both groups received the sameamount of parent training, as all cancelled sessionswere rescheduled.

Measures

Fidelity. Sixty four videotapes from a random set ofsessions were coded for frequency of the following 10 ISteaching behaviors: language paired with pointing ges-ture; hand-over-hand pointing; showing; modelingfacial expression; labeling facial expression; imitation ofchild action; verbal label of imitated action; hand-over-hand imitation; modeling gesture; and modeling posi-tive affect. Interventionists were videotaped on averagetwice during each intervention session and were blindas to whether videotaping was being conducted forpurposes of coding children’s behavior or fidelity. Twotrained and reliable staff, one being blind to interven-tion condition, coded 45 minutes of instructionalactivities. Significantly more IS instruction was pro-vided by IS than Non-IS interventionists (p’s = .047 to<.001). Inter-observer reliability was assessed for 27%(n = 17) of coded videotapes; intraclass correlationcoefficients (ICC) ranged from .79 to .97.

Outcome. Children were assessed at baseline (pre-test), termination of the intervention (post-test), and ata six-month follow-up within the Autism Center’s ChildDevelopment Laboratory by a clinician blind to groupmembership and unfamiliar to the child.

Primary outcome variables. Initiation of jointattention (IJA) and shared positive affect (SPA) weremeasured using the Communication and SymbolicBehavior Scales Developmental Profile (Wetherby &Prizant, 2002), which presents communication temp-tations and collects a play sample using a standardizedtoy set. ICCs for frequency of IJA and SPA calculated on20% (n = 28) of videotaped sessions were .952 and .999,respectively.

Socially engaged imitation (SEI) was defined as pro-portion of imitations paired with eye contact to theexaminer. SEI was coded blind to group membershipand time point from videotapes of a structured imitationtask, modified from that described by Rogers andcolleagues (2003). Examiners modeled 15 novel facial,manual, and object manipulation movements, pro-viding two response opportunities per movement, aschildren were instructed to ‘Do this’. ICC calculated on15 (12%) videotapes was .956.

Secondary outcome variables. The MSEL Expres-sive Language (EL) and VR (estimate of nonverbal cog-nition) T scores (mean = 50; sd = 10) were secondaryoutcome variables. The MSEL is a developmental teststandardized for ages 3 to 69 months.

Data analysis

T-tests and chi-square tests assessed group compara-bility at baseline. Random effects models (hierarchicallinear or multilevel models; Laird & Ware, 1982) wereused to model each outcome over time. This longitudi-nal model, with a child-level random effect, permitteduse of all observations available for each child whileaccounting for the correlation of observations withineach child. The model included a factor for time(allowing growth during intervention and follow-up



periods to differ), a group (IS/Non-IS) indicator, and agroup by time interaction as predictors. This modelallowed us to test for between-group differences atpost-test and follow-up and between their growth rates.To account for chance differences between groups atpre-test and increase precision, the model also con-trolled for pre-test values of each outcome variable,ADOS algorithm combined Communication and SocialInteraction scores (a measure of autism severity), andMSEL Receptive Language T scores. Models wereestimated using xtmixed in Stata 10 (StataCorp,2007). Effect sizes were calculated using Cohen’s d(Cohen, 1988), using the pre-test pooled standarddeviation. Effect sizes reflect the regression adjustment,which controls for pre-test characteristics of the chil-dren. Significance level for all comparisons was set atp = .05.

The effects estimated are intent-to-treat effects,which estimate the effect of assignment to the ISintervention, regardless of the amount of interventionactually received. We analyzed data from all individualsoriginally assigned to an intervention group, with theexception of one child in the Non-IS group who wasmissing a pre-intervention IJA value and one childlost to follow up in the IS group. Two children weremissing an outcome variable at post-test or follow-up.Standard errors reflect occasional missing outcomevalues.

Results

Pre-intervention equivalence of groups

T-tests and chi-square tests revealed no between-group differences on any of the dependent variablesat pre-test (p’s = .239 to .924), indicating well-mat-ched groups (see Table 2).

Main effects of intervention on primary outcomevariables

The joint tests of differences involved an overallbetween-group comparison for each primary out-come (SEI, IJA, SPA). A statistically significantoverall main effect was found for SEI (chi-square= 9.98, df = 3, p = .02; Table 2), but not for IJA(chi-square = 3.65, df = 3,p = .30) orSPA (chi-square= 2.61, df = 3, p = .46). Subsequent between-groupcomparisons at post-test and follow-up revealedsignificantly more SEI performed by the IS than Non-IS group at follow-up (p = .01; d = .86). Analysis ofbetween-group differences in slopes (rates of growth)within each assessment period revealed differencesin SEI growth during the intervention period (p = .04;Table 3) but not during follow-up (p = .24). Between-group comparisons of IJA and SPA at post-test andfollow-up revealed a trend toward significance formore frequent IJA at post-test (p = .08, d = .89) andfollow-up (p = .07, d = 1.56) in the IS group. Lack ofsignificance on the joint test for IJA could be relatedto the significantly larger variance within the ISgroup for IJA at follow-up (p < .001). T

able

2Outcomedifference

sbetwee

nIS

andnon-ISgroups

Outcome

Pre-tes

tPost-tes

tFollow-u

pJointtest

of

interven

tion

groupdifference

sJointtest

of

timeeffect

Non-IS

ISDiff.

Non-IS

ISDiff.

Non-IS

ISDiff.

Mea

n(SD)

Mea

n(SD)

Effec

tsize

(p-value)

Mea

n(SD)

Mea

n(SD)

Effec

tsize

(p-value)

Mea

n(SD)

Mea

n(SD)

Effec

tsize

(p-value)

p-value

p-value

Primary

outcomes

SEI

.25(.24)

.17(.19)

).29(.30)

.35(.23)

.42(.24)

.48(.11)

.28(.21)

.44(.22)

.86(.01)*

.02*

<.001*

IJA

2.79(3.62)

2.29(3.16)

).04(.89)

5.00(7.91)

7.70(9.33)

.89(.08)

4.42(4.82)

8.83(13.22)

1.56(.07)

.30

<.001*

SPA

3.54(3.56)

2.42(2.93)

).41(.31)

5.33(6.64)

6.87(7.55)

.42(.41)

4.96(5.48)

7.57(9.76)

.81(.27)

.46

.01*

Sec

ondary

Outcomes

ELT

25.92(8.12)

23.92(5.50)

).11(.71)

31.92(13.67)

34.08(14.59)

.49(.18)

31.36(12.12)

34.52(12.33)

.57(.24)

.44

<.001*

VR

T31.12(9.86)

27.50(8.27)

).27(.26)

32.24(14.07)

36.75(14.54)

.57(.07)

30.28(16.62)

34.44(16.67)

.46(.33)

.16

<.001*

Effec

tsizesin

Tables2and3wereca

lculatedforea

choutcomebystandard

izingth

eregres

sion-adjusted

difference

betwee

nIS

andNon-ISgroupsbyth

epooled(ISandNon-IS)standard

dev

iationatpre-tes

t:SEI=0.21;IJA=3.33;SPA=3.19;ELT=6.82;VRT=8.93.Reg

ress

ioninco

rpora

tesadjustmen

tforpre-tes

tva

lues

ofea

choutcomeva

riable

andoth

erpredictors

as

detailed

inth

etext.

*Signifies

met

criteria

forsign

ifica

nce

at

£.05leve

landbolded

pva

lues

sign

ifystatisticaltren

din

Tables2and3.



Secondary analyses

To further inform the sparse intervention literatureon language, cognitive, and social growth in toddlerswith ASD, we performed two sets of secondaryanalyses involving: (a) examination of secondaryoutcomes of expressive language and nonverbalcognition; and (b) examination of growth overtime (slopes) on the entire sample and then withingroups.

Effects of intervention on secondary outcome vari-ables. Joint tests of differences did not revealoverall between-group differences on EL or VR Tscores (chi-square = 2.71, df = 3, p = .44 for EL;chi-square = 5.24, df = 3, p = .16 for VR), despitemoderate effect sizes at post-test of .49 and .57,respectively, and at follow-up of .57 and .46,respectively (Table 2). A marginal treatment effectwas found for VR T score, with a trend toward higherlevels of nonverbal cognition at post-test for the ISgroup (p = .07; Table 2), a result of greater growthduring the intervention period for the IS group (p =.02; Table 3). However, these VR trends should beinterpreted with caution owing to the lack of signifi-cance of the joint test of differences (p = .16).

Growth over time. Examining growth over timepermits an understanding of the timing, rate, anddirection (increasing, decreasing) of change, provid-ing insight into whether intervention gains weresustained after termination of the intervention.Separate joint tests of difference were conducted foreach outcome to examine overall growth trends forthe entire sample from pre-test to follow-up. Signif-icant time effects were found for the full sample forall variables (p’s = .00 to .01; Table 2), indicatingimproving performance over time.

Next, we examined growth trends of the two groupsseparately to better understand whether growth foreach group during the intervention phase was main-tained during follow-up. Table 3 examines whetherslopes within the intervention and follow-up periodsdiffered for each group. The IS group showed signifi-cant growth over time on every outcome (p’s < .01)while the Non-IS group showed significant growthonly on the EL T score (p = .01). Further analysescompared growth during the intervention period togrowthwithin the follow-up periodwithin each group.For all outcomes except for SPA (p = .07), the IS grouphad significantly more rapid growth in the interven-tion period than in the follow-up period (p’s < .001).The effect sizes for the IS group’s growth during theintervention period were all over 1 and all reachedsignificance (p < .001). In contrast, effect sizes for theIS group’s growth during the follow-up period, whilenot receivingour intervention, ranged from–.34 forVRT score to .34 for IJA, with none being statisticallysignificant. The Non-IS group showed significantgrowth during the intervention period only for EL T

Table

3Explora

tory

analyse

s:within-groupandbetwee

n-groupco

mpariso

nsofgrowth

rate

(slopes

)

Outcome

Non-ISGroup

ISGroup

Difference

betwee

nIS

and

Non-IS

Pre-tes

tto

post-tes

tPost-tes

tto

follow-u

pChange

inslopes

Ove

ralltime

effect

Pre-tes

tto

post-tes

tPost-tes

tto

follow-u

pChange

inslopes

Ove

ralltime

effect

Pre-tes

tto

post-tes

tPost-tes

tto

follow-u

p

Effec

tsize

(p-value)

Effec

tsize

(p-value)

Effec

tsize

(p-value)

p-value

Effec

tsize

(p-value)

Effec

tsize

(p-value)

Effec

tsize

(p-value)

p-value

Effec

tsize

(p-value)

Effec

tsize

(p-value)

Primary

outcomes

SEI

.48(.06)

).33(.17)

).81(.05)*

.13

1.19(<

.01)*

.05(.78)

)1.14(.01)*

<.001*

.76(.04)*

.43(.24)

IJA

.66(.10)

).34(.42)

)1.00(.05)*

.13

1.59(<

.01)*

.34(.41)

)1.25(.02)*

<.001*

.93(.11)

.68(.25)

SPA

.56(.19)

).18(.68)

).74(.25)

.40

1.39(<

.01)*

.22(.62)

)1.18(.07)

.01*

.83(.17)

.41(.52)

Sec

ondary

outcomes

ELT

.89(.02)*

).10(.73)

).99(.02)*

.01*

1.49(<

.01)*

).01(.97)

)1.50(<.001)*

.00*

.60(.13)

.09(.83)

VR

T.20(.44)

).23(.37)

).43(.25)

.51

1.04(<

.01)*

).34(.20)

)1.37(<.001)*

.00*

.84(.02)*

).10(.78)



score (p = .02), with no significant growth detected onany variable during follow-up.

Discussion

To our knowledge, this is the first RCT involving2-year-olds designed to assess effects of an interven-tion ingredient on social development. We comparedchange in self-initiated socially engaged behavior intwo groups receiving comparable treatments, differ-ing only in the presence or absence of a supplemen-tary social curriculum targeting SEI, IJA, and SPA,which are core deficits of autism. The test of SEI, IJA,and SPA outcome was stringent, reflecting general-ization of these behaviors to a novel interaction con-text (location and materials), with an unfamiliarexaminer, under unfamiliar circumstances (activity).Significant treatment effects were found for SEI atfollow-up, with a trend toward significance for IJA atpost-test and follow-up. Moderate effect sizes for SEIand IJA provide preliminary evidence for the efficacyof this intervention where a supplementary curricu-lum for social initiation and socially synchronizedengagement in toddlers is overlaid on an existinggroup-based comprehensive intervention. A trendtoward significance was found for greater improve-ment in the secondary variable of nonverbal cognitionwithin the IS group during the intervention period.Trends toward significance for IJA and nonverbalcognition should be interpretedwith caution since thejoint tests for intervention differences were not sig-nificant for these outcomes. Secondary analyses alsorevealed social, cognitive, and language gains in bothgroups, particularly during the six-month interven-tion period and within the IS group.

Our findings provide insights into the plasticity ofsocial and communication development in toddlerswith ASD. The supplementary curriculum activateddevelopment of socially engaged imitation (a greaterthan twofold increase from pre- to post-test) thatgeneralized across people, location, and interactioncontext. While others have reported improvements inimitation performance following targeted instruction(Ingersoll & Schreibman, 2006), this is the first studyto identify a targeted treatment response involvinggains in the integration of imitation with sociallydirected gaze. Such ability enables toddlers with ASDto liken themselves to peers while also signaling theirinteractive engagement. This could foster a cascade ofevents in which toddlers with ASD draw increasingsocial, play, and language learning opportunities tothemselves. Future research is needed to determinewhether toddlers with ASD who develop SEI learnmore efficiently or implicitly within other domains ofdevelopment since imitation is a vital avenue throughwhich language and play learning occurs (Ingersoll &Schreibman, 2006; Speidel & Nelson, 1989).

The finding that both groupsmade significant gainsover time in primary as well as secondary outcomes is

of interest. The absence of more robust between-group differences is likely attributable to thesimilarity between the rival interventions. Instruc-tional strategies for both interventions emphasizedself-initiated communication, child choice, motivat-ing learning materials and activities, joint actionroutines, natural reinforcers, and play-based learn-ing activities. Some of these strategies are vitalingredients for development of flexible and generativelanguage in typically developing children (Barsalou,Breazeal, & Smith, 2007; Booth &Waxman, 2002). Inaddition, the AEPS curriculum includes imitationand joint attention goals, albeit in less detail than inthe IS curriculum. The high level of child–otherengagement within the ecologically valid learningcontexts of both interventions may have been suffi-cient to activate a high level of affect sharing and jointattention even without the additional infusion ofinstruction targeting joint attention and affect shar-ing. Indeed, Kasari, Paparella, Freeman, and Jahromi(2008) reported significant gains in joint attentionbehavior in preschoolers with ASD regardless ofwhether they received instruction focused on sym-bolic play or joint attention. Kasari et al.’s (2008) andour findings indicate that development of IJA andaffect sharing may be stimulated through target-ing interactions involving a joint focus of attentionwithin motivating activities where children’s self-generated initiation is targeted. Our finding thatslopes became less steep in the IS group or decreasedin the Non-IS group during follow-up suggests theneed for ongoing intervention targeting social devel-opment, and possibly that the community-basedinterventions did not sufficiently address theseaspects of development. It also indicates that the ISgroup did not lose the skills acquired during theintervention. Further research is needed to under-stand more about the intervention ingredients thatstimulate social development in toddlers with ASD,and to identify characteristics of children who showlimited social improvement.

The gains in our participants’ expressive languageand nonverbal cognitive functioning are encourag-ing. Since we did not include an ‘assess and monitor’control group, we cannot conclusively infer thatgains in these secondary outcome variables wereeffected by the overall intervention program commonto both groups. However, there is indirect evidencefor the intervention’s developmental impact. First,the most rapid growth in development occurredduring the six-month active intervention period.

The second source of indirect evidence is found inother published reports. In Dawson and colleagues’(2009) RCT involving 2-year-olds with ASD receivingeither 25 hours per week of 1:1 home-based EarlyDenver Model intervention or referral to communityintervention providers, gains in MSEL EL and VR Tscores for the Early Denver and community inter-vention groups, after one year of intervention, werenearly identical to gains observed in our IS and Non-



IS groups, respectively, from pre-test through follow-up (measured six months after the termination oftreatment) on these same variables. Although Daw-son and colleagues did not include social variablescomparable to those reported herein, the expressivelanguage and nonverbal cognition results from theirstudy and ours indicate that toddlers with ASDare able to make substantial gains in early inter-vention using well-defined responsive instructionalstrategies and a developmentally based curriculum.Furthermore, our data indicate that such gains maybe effected in a relatively brief period of time.Numerous intervention studies involving preschool-ers with ASD have also reported gains in nonverbalcognition and language (reviewed by Rogers &Vismara, 2008); these aspects of development maybe particularly plastic and responsive to early inter-vention.

The present study contributes to a small body ofliterature focused on core deficits of communicationand social functioning in young children with aut-ism. Given the pervasive nature of imitation, jointattention and affect sharing deficits in most youngchildren with ASD (Kasari, Freeman, & Paparella,2006; Landa et al., 2007; Sullivan et al., 2007), andgiven that early social development is a major pre-dictor of outcome for children with ASD (Charmanet al., 2003; Sigman & Ruskin, 1999), there is a vitalneed to extend the research on which we report here.Future research should include a focus on whetherand how an intervention such as the one reportedhere might impact parent–child interaction, andthus, mediate intervention effects.

Conclusions

This is the first RCT to identify a developmentalsocial curriculum as an active ingredient forimproving socially engaged imitation in toddlers withASD. This effect was obtained by layering a supple-mentary social curriculum onto an intervention thatutilized a well-defined developmental curriculum

and integrated several evidence-based instruc-tional strategies. Thus, an incremental benefit wasachieved at no extra cost. The results highlight theneed for future investigation of scalable early inter-vention models targeting core deficits of ASD andthat examine factors related to generalizationand maintenance of early social, communication,and cognitive skills.

Acknowledgements

We wish to thank the children and parents whoparticipated in this study. We also thank the superbstaff who provided the intervention, administrativesupport, and data processing: Jessica Henkin, Kat-rina Martin, Dana Herman, James Mancini, AshleyEdmunds, Sarah Beal, Amy Knecht, Amy FalkSmith, Marguerite Adams, Allison Nelson, Dr. Mi-chelle Sullivan, Dr. Julie Cleary, and Andrea Gollo-her. We also thank Dr. Laura Schreibman fortraining the staff on Pivotal Response Training andfor her ongoing quality assurance assessmentsduring the study. Thanks also to Ashley Faherty,Dr. David Mandell, and Dr. Jeffery Wood for theirassistance in edits to this manuscript.

Funding/Support: Rebecca Landa (PI) thanksthe National Institute of Mental Health (154-MH066417; Studies to Advance Autism Researchand Treatment) and HRSA (R40 MC 15594) for thefunding to conduct this study and prepare thismanuscript.

The corresponding author, Dr. Rebecca Landa,had full access to all of the data in the study andtakes responsibility for the integrity of the data andthe accuracy of the data analysis.

Correspondence to

Rebecca Landa, 3901 Greenspring Avenue, Balti-more, Maryland 21211, USA; Tel: 443-923-680; Fax:443-923-7560; Email: [email protected]

Key points

• Autism intervention studies have focused largely on children age 3 and older. No randomized controlledtrials (RCTs) have shown evidence for improvement in core social deficits of autism in toddlers with ASD.

• This is the first RCT involving 2-year-olds with ASD to assess the efficacy of an intervention ingredienttargeting core social deficits of autism.

• Our findings indicate that toddlers with ASD improve in socially engaged imitation in response to atargeted curriculum and intervention activities. Significant improvements in expressive language, cog-nitive, and social functioning occurred in both intervention groups.

• Early intervention for ASD should target social abilities as part of a comprehensive intervention program.

References

Barsalou, L., Breazeal, C., & Smith, L. (2007). Cognition ascoordinated non-cognition. Cognitive Processing, 8, 79–91.

Bono, M., Daley, T., & Sigman, M. (2004). Relations amongjoint attention, amount of intervention, and languagegain in autism. Journal of Autism and DevelopmentalDisorders, 34, 495–505.



Booth, A., & Waxman, S. (2002). Object names and objectfunctions serve as cues to categories for infants. Devel-opmental Psychology, 38, 948–957.

Bricker, D. (2002). Assessment, Evaluation, and Program-ming System for Infants and Children: Birth to ThreeYears (2nd edn). Baltimore, MD: Brookes.

Carr, E.G., Binkoff, J.A., Kologinsky, E., & Eddy, M.(1978). Acquisition of sign language by autistic children.I: Expressive labeling. Journal of Applied Behavior Anal-ysis, 11, 489–501.

Chandler, S., Christie, P., Newson,E.,&Prevezer,W. (2002).Developing a diagnostic and intervention package for 2- to3-year-olds with autism: Outcomes of the frameworks forcommunication approach. Autism, 6, 47–69.

Charman, T., Baron-Cohen, S., Swettenham, J., Baird, G.,Drew, A., & Cox, A. (2003). Predicting language outcomein infants with autism and pervasive developmentaldisorder. International Journal of Language and Commu-nication and Disorders, 38, 265–285.

Charman, T., Swettenham, J., Baron-Cohen, S., Cox, A.,Baird, G., & Drew, A. (1997). Infants with autism:An investigation of empathy, pretend play, joint attention,and imitation. Developmental Psychology, 33, 781–789.

Chawarska, K., Klin, A., Paul, R., & Volkmar, F. (2007).Autism spectrum disorder in the second year: Stabilityand change in syndrome expression. Journal of ChildPsychology and Psychiatry, 48, 128–138.

Cohen, J. (1988). Statistical power analysis for the behav-ioral sciences (2nd edn). Hillsdale, NJ: Lawrence Erl-baum Associates.

Dawson, G., Rogers, S., Munson, J., Smith, M., Winter, J.,Greenson, J., Donaldson, A., & Varley, J. (2009). Ran-domized, controlled trial of an intervention for toddlerswith autism: The Early Start Denver model. Pediatrics,125, 17–23.

Drew, A., Baird, G., Baron-Cohen, S., Cox, A., Slonims, V.,Wheelwright, S., Swettenham, J., Berry, B., & Charman,T. (2002). A pilot randomized control trial of a parenttraining intervention for pre-school children with aut-ism: Preliminary findings and methodological chal-lenges. European Child and Adolescent Psychiatry, 11,266–272.

Harris, S., & Handelman, J. (2000). Age and IQ at intake aspredictors of placement for young children with autism:A four- to six-year follow-up. Journal of Autism andDevelopmental Disorders, 30, 137–142.

Hollingshead,A.B. (1975).Fourfactor indexofsocialposition.NewHaven, CT: Yale University, Department of Sociology.

Howlin, P., Goode, S., Hutton, J., & Rutter, M. (2004).Adult outcome for children with autism. Journal of ChildPsychology and Psychiatry, 45, 212–229.

Ingersoll, B., & Schreibman, L. (2006). Teaching reciprocalimitation skills to young children with autism using anaturalistic behavioral approach: Effects on language,pretend play, and joint attention. Journal of Autism andDevelopmental Disorders, 36, 487–505.

Kasari, C., Freeman, S., & Paparella, T. (2006). Jointattention and symbolic play in young children withautism: A randomized controlled intervention study.Journal of Child Psychology and Psychiatry, 47, 611–620.

Kasari, C., Paparella, T., Freeman, S., & Jahromi, L.B.(2008). Language outcome in autism: Randomized com-parison of joint attention and play interventions. Journalof Consulting and Clinical Psychology, 76, 125–137.

Koegel, R., Schreibman, L., Good, A., Cerniglia, L., Mur-phy, C., & Koegel, L. (1989). How to teach pivotalbehaviors to children with autism: A training manual.Santa Barbara: University of California.

Laird, N.M., & Ware, J.H. (1982). Random effects modelsfor longitudinal data. Biometrika, 39, 963–974.

Landa, R.J., Holman, K.C., & Garrett-Mayer, E. (2007).Social and communication development in toddlers withearly and later diagnosis of autism spectrum disorders.Archives of General Psychiatry, 64, 853–64.

Lord, C., Bristol, M., & Schopler, E. (1993). Early inter-vention for children with autism and related disorders.In E. Schopler, E. Van Bourgondien, & M. Bristol (Eds.),Preschool issues in autism (pp. 199–221). New York:Plenum Press.

Lord, C., Rutter, M., DiLavore, P., & Risi, S. (1999). AutismDiagnostic Observation Schedule. Los Angeles, CA:Western Psychological Services.

Lovaas, O. (1987). Behavioral treatment and normal edu-cational and intellectual functioning in young autisticchildren. Journal of Consulting and Clinical Psychology,55, 3–9.

Mullen, E.M. (1995). Mullen: Scales of Early Learning (AGSedn). Circle Pines, MN: American Guidance Service.

Mundy, P., Sullivan, L., & Mastergeorge, A.M. (2009). Aparallel and distributed-processing model of joint atten-tion, social cognition and autism. Autism Research:Official Journal of the International Society for AutismResearch, 2, 2–21.

Rogers, S.J., Hepburn, S.L., Stackhouse, T., & Wehner, E.(2003). Imitation performance in toddlers with autismand those with other developmental disorders. Journal ofChild Psychology and Psychiatry, 44, 763–781.

Rogers, S.J., & Vismara, L.A. (2008). Evidence-basedcomprehensive treatments for early autism. Journal ofChild and Adolescent Psychiatry, 37, 8–38.

Schertz, H., & Odom, S. (2007). Promoting joint attentionin toddlers with autism: A parent-mediated developmen-tal model. Journal of Autism and Developmental Disor-ders, 37, 1562–1575.

Sigman, M., & Ruskin, E., (1999). Continuity and changein the social competence of children with autism, Downsyndrome, and developmental delays. Monographs ofthe Society for Research in Child Development, 64,v-114.

Sparrow, S.S., Balla, D.A., & Cicchetti, D.V. (1984). Vine-land adaptive behavior scales: Interview edition surveyform manual. Minnesota: American Guidance Service.

Speidel, G.E., & Nelson, K.E. (1998). The many faces ofimitation in language learning. New York: Springer-Verlag.

Stahmer, A., Ingersoll, B., & Koegel, R. (2004). Inclusiveprogramming for toddlers with autism spectrum disor-ders: Outcomes from the Children’s Toddler School.Journal of Positive Behavior Interventions, 6, 67–82.

StataCorp. (2007). Stata Statistical Software (Release 10)[computer software]. College Station, TX: StataCorp LP.

Toth, K., Munson, J., Meltzoff, A., & Dawson, G. (2006).Early predictors of communication development inyoung children with autism spectrum disorder: Jointattention, imitation, and toy play. Journal of Autism andDevelopmental Disorders, 36, 993–1005.

Wetherby, A.M., & Prizant, B.M. (2002). CSBS DP Manual:Communication and Symbolic Behavior Scales Develop-mental Profile. Baltimore, MD: Paul H Brookes.

Wetherby, A.M., Watt, N., Morgan, L., & Shumway, S.(2007). Social communication profiles of children withautism spectrum disorders late in the second year of life.Journal of Autism and Developmental Disorders, 37,960–975.

Manuscript accepted 24 May 2010



Copyright of Journal of Child Psychology & Psychiatry is the property of Wiley-Blackwell and its content maynot be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express writtenpermission. However, users may print, download, or email articles for individual use.

Dr. Katherine Holman

Education

Transcript of Dr. Katherine Holman