10/15/16

1

Early interventions for ASD: Active ingredients and deployment in the

community

Connie Kasari. PhD

SPD Conference, SeattleNovember 5, 2016

Autism Intervention Research Network for Behavioral Health

None

Disclosures

1

Introduction

ba

Summary

3c

3 issues

10/15/16

2

1

Introduction

ba

Research topractice gap

Summary

3c

3 issues

1

Introduction

No singletreatment

ba

Research topractice gap

Summary

3c

3 issues

1

Introduction

No singletreatment

ba

Research topractice gap

Summary

3c

DeploymentSustainability

3 issues

10/15/16

3

Today’s intervention landscape

LAUSD…

Hispanic

Other

Black

White

Hispanic

White

Black

Other

2nd largest school district; 14,000 students

with ASD

District as a whole is…..75% Hispanic;

80% free & reduced lunch;9% White

Students with ASD

Disparities are real…..

Public schools should be place without disparities in services

10/15/16

4

3 Issues to consider……

There is a HUGE research to practice GAP

1

Most children with ASD have never been in a research study

AND MOST INTERVENTIONS have never been tested in any controlled study;

The ‘evidence-base’ does not reflect most children in the community (Weisz, 2004)

10/15/16

5

Joint Attention !Symbolic Play Engagement !

Regulation JASPER Early Start Denver !Model ESDM Pivotal Response Training !

PRT Floortime DIR Developmental Individual !difference Relationship-based RDI Relationship !

Development Intervention Early Achievements Proj!ImPACT PEERS Secret Agent Society Discrete Trial !

Training DTT Verbal Behavior Intervention VB SCERT !Social Communication Emotion Regulation Transaction

Supports Unstuck and On Target Focused Playtime Intervention Adapted Responsive Intervention Joint

Attention Symbolic Play Engagement Regulation JASP!Early Start Denver Model ESDM Pivotal Response !Training PRT Floortime Developmental Individual !difference Relationship-based DIR Relationship !

Development Intervention RDI Early Achi !Project ImPACT PEERS Secret Agent!

Society Discrete Trial Trainin !DTT Verbal Behavior

Proliferation of programs

What is the evidence-base for early interventions?

We can improve cognitive outcomes (DQ) in young children with comprehensive, many hour

per week interventionsDTT- preschoolers30 hrs/wk vs. <10

40

50

60

70

80

90

100

Baseline 2 years

16 pts

1 pt

ESDM-toddlers25 hrs/wk vs. fewer hrs

40

50

60

70

80

90

100

Baseline 1 year 2 years

19 pts

5 pts

Smith, Groen & Wynn, AJMR, 2000 Dawson et al, Pediatrics, 2010

DQ points

n=28 n=48

10/15/16

6

Reproducibility of findings

Problem in science

Original studies—small samples

Often do not replicate

Science, 2015

Confidence in findings?

Replications

DTT original Smith et al, 2000 n=28 p<.05Replication Sallows & Graupner, 2005 n=24 NS

ESDM original Dawson et al 2010 n=48 p<.05Replication Rogers et al 2014 n=98 NS

Follow up Studies

Rare…..

Follow up of ESDM, Estes et al, 2015 n= 39 NS

10/15/16

7

Similar outcomes in brief parent mediated interventions?

Focused Playtime Intervention

40

50

60

70

80

90

100

Baseline 1 year

18 pts

14 pts

SCERTS

40

50

60

70

80

90

100

Baseline 9 months

6 pts

2 pts

Kasari, Siller et al, Infant Behavior & Development, 2014

DQ points

Wetherby et al, Pediatrics, 2014

Intervention phaseIntervention

phase

What do we make of these findings…..

Outcome measures

Active ingredients

Primary outcome has been IQ

Cannot rule out dose of intervention

Year of study matters—”the counterfactual”

But many factors can influence IQ

10/15/16

8

While IQ is likely important; it is not a core

deficit of ASD

What are the active ingredients of early interventions?

DoseAgent of change (BCBA therapist, parent, other?)

Teaching approachContent ---what is focused on

Others…..

What are the active ingredients of early interventions?

✔DoseAgent of change (BCBA therapist, parent, other?)

✔Teaching approachContent ---what is focused on

Others…..

10/15/16

9

What are the active ingredients of early interventions?

✔DoseAgent of change (BCBA therapist, parent, other?)

✔Teaching approach✔Content ---what is focused on

Others…..

Content focusing on core impairments

Social CommunicationRestricted, repetitive behaviors

To large extent, independent of IQ

Engagement Communicativegestures & language

Play

Early Core DeficitsSocial Communication

10/15/16

10

Characteristics of ASD

Rigid, no joy in play

Object focused

Typical 17 month old

3 year old with ASD

EngagementRepetitive behavior

Social engagement, joint attention, shared

affect

Why focus on early core deficits?

10/15/16

11

Language by age 5-6 best social outcomes (Lord, 2000; Rutter, 1978)

Joint attention predicts to language (Kasari et al,2008; Kasari et al, 2012; Mundy et al., 1986; Mundy, Sigman, & Kasari,

1990)

Play skills associated with cognitive abilities(Kasari et al, 2012)

Evidence suggests that approach matters for

improvements in joint attention, play, engagement

Spontaneous, child initiations

Differences in approach to teaching play

ABA-traditionalDTT teaching symbolic play

NDBIJASPER

10/15/16

12

Joint Attention, Symbolic Play, Engagement, & Regulation

JASPER

Naturalistic Developmental Behavioral Intervention (Workgroup, 2014)

NDBI Targeted

Modular Short-term

Research Addressing Core Deficits

JASPER Core ConceptsJoint Attention

PlayEngagement

10/15/16

13

JASPER Core StrategiesEnvironmental Arrangement

Imitation & ModelingRoutines (play)Programming for JALanguage Techniques

Toy choices in close proximity

Clear play space –(rug is defined play area)

Caregiver facing child

10/15/16

14

Engagement

Play

Joint attention

Language

JASPER is a layered intervention

Highly replicable findings (therapists, teachers, parents)

é Joint engagementé Play skillsé Joint Attention

JASPER demonstrates effects above and beyond background treatment

Evidence

10/15/16

15

Parent mediated interventions….strategies layered in over time

FIRST---Goalisselectedforchildthatisdevelopmentallyontarget—PHASE1(3weeks)

Environmentalarrangements(settinguptheenvironment)(andknowingchildplaylevel)

Allowingthechildtoinitiateanactivity(followingthechild’sattentionalfocus)Establishingandplayingwithinestablishedroutines

PHASE2(3weeks)Facilitating,maintainingengagementstatesBalancebetweenimitatingandmodeling

Expansions,showingenjoyment,eye-contactJointattention:recognizing,responding,modeling

PHASE3(3- 5weeks)AllowingchildaccesstocommunicationInitiatingandexpandinglanguage

GeneralizingskillsacrossotherroutinesPractice….upto2weeksmore

Someadditionalmodulescanfloat….”unengagedmodule”;“behaviorregulation”

Information Sharing Hands on Coaching

(Kasari, Gulsrud, Paparella, Hellemann, Berry, JCCP, 2015)

RCT—Comparative EfficacyN=86

050

100150200250300350400

Entry Exit Follow up

JASPER

Parent Education

Parent Mediated Interventions

Seco

nds

Child initiated joint engagement

10/15/16

16

(Kasari, Gulsrud, Paparella, Hellemann, Berry, JCCP, 2015

Parent Mediated Intervention--Toddlers

86 children (2-3 year olds)

Comparison 2 interventions

20 sessions over 10 weeks

6 month follow up

JASPER groupé Joint engagement

é Play diversityé Play level

Group comparisonê Parenting stress

(Kasari, Lawton, Shih, Barker et al, Pediatrics, 2014

200

220

240

260

280

300

320

340

360

380

400

Entry Exit Follow up

JASPER

Group Education

RCT-Comparative EfficacyN=112

Parent Mediated InterventionChild initiated joint engagement

(Kasari, Lawton et al, Pediatrics, 2014)

Parent Mediated Intervention--Preschoolers

112 children (2-5 year olds)

Comparison 2 interventions

24 sessions over 12 weeks

3 month follow up

JASPER group

é Joint engagementé Symbolic playé Initiating JA

10/15/16

17

Good evidence of improving core deficits using JASPER

(NDBI interventions)Ingersoll, 2011, JADD RCT—improvement in JA initiations from training

imitationWetherby et al, 2014, Pediatrics---improvement of collapsed requesting, joint

attention category

Major questions going forward..

What are the downstream effects?

Can the intervention be deployed effectively?

IJA predicts to expressive language 1 and 5 years later

1. Which strategies or ‘active ingredients’ affect change in outcomes

a. 2 studies testing mechanism….both parent mediated models--- in toddlers ‘mirrored pacing’ affects joint engagement for JASPER (Gulsrud et al, 2015, JCPP) and in preschoolers parental synchrony affects ASD symptoms for PACT (Pickles et al, 2014, JCPP)

Bridging research to practiceConsider….

Active ingredients will be very important in deployment to rural, distance sites----Macedonia

10/15/16

18

1. Which strategies or ‘active ingredients’ affect change in outcomes

a. 2 studies testing mechanism….both parent mediated models--- in toddlers ‘mirrored pacing’ affects joint engagement for JASPER (Gulsrud et al, 2015, JCPP) and in preschoolers parental synchrony affects ASD symptoms for PACT (Pickles et al, 2014, JCPP)

2. Strategies, modules to build a comprehensive program

a. Clinicians more likely to adopt a module than completely changing practices (Chorpita et al, 2005; Weisz et al, 2012)

Bridging research to practiceConsider….

A single treatment is not effective for all!

2

Methodologies are needed to personalize, tailor interventions

Address for whom the intervention works, and why…..

Sequence treatments…..

10/15/16

19

Need: Diversify samples, in real world settings

Problem is that researchers tend to exclude subpopulations

Minimally verbal often excluded

NIH Workshop Who are the ‘minimally verbal’

• Clear most are not ‘nonverbal’

• Defined by number of functional words spoken

• Some can speak but rarely do or only in some contexts

• Treatment is often to do MORE of the same…..

• (or less, blaming child for lack of progress)

Sequence of treatments

Adaptations based on child response

10/15/16

20

DEFINITION: A sequence of decision rules that specify whether, how, when (timing) and

based on which measures, to alter the dosage (duration, frequency or amount), type or

delivery of treatment(s) at decision stages in the course of care.

Adaptive Intervention designs systematize clinical practice

SMART design

Sequential Multiple Assignment Randomized Trial

Characterizing Cognition in Nonverbal Children with ASD (CCNIA) Intervention

61 children aged 5 to 8 years

Minimally verbal (fewer than 20 functional words)

Had already received 2 years of intensive early intervention

ALL received JASPER plus a spoken language intervention

(EMT)

HALF randomized to also receive Speech Generating Device (iPad)

Kasari, Kaiser, Landa, Neitfeld, Mathy, Murphy, Almirall, JAACAP, 2014

10/15/16

21

Slide courtesy of Danny Almirall May 16, 2014

Minimally verbal and meaningful outcomes

2030

4050

6070

Week

Tota

l Soc

ially

Com

mun

icat

ive U

ttera

nces

0 12 24 36

●

●

●

●

●

●

●

●

●

JAE+EMT+AACJAE+EMT

Socially communicative utterances

Novel words and comments

Kasari, Kaiser, Goods, Neitfeld, Mathy, Landa, Murphy, Almirall, JAACAP, 2014

10/15/16

22

Minimally verbalSchool aged: No words in beginning; needs access to

communication

Preverbal child: preschool aged

Day 1

Day 2

Day 12

Deployment can be limitedSustainability poor

Training models and implementation fidelity

3

1. Deployment focused modelsIntervention development and testing with participants within

context (e.g., homes/schools) for which intervention is intended and by people intended to deliver the intervention (e.g.,

parents, teachers)

2. Hybrid implementation modelsFocus on implementation and outcomes

3. Community partnered participatory research models

Shared goals and responsibilities

Sustainability might be improved by…

10/15/16

23

Deployment Models

Train the trainer models; deploying to paraeducators who are doing both assessment and intervention

NYTD

LAPDFitting the intervention

into existing structure with support

!

Ya-Chih Chang, PhD

Stephanie Shire, PhD

Chang, Shire, Shih, Kasari, 2016, JADDShire, Chang, Shih, Bracaglia, Kodjoe, & Kasari, 2016, JCPP

JASPER findings similar to lab when conducted in school: Better joint engagement with others

05

101520253035404550

JASPER Wait

EntryExit

Perc

enta

ge o

f tim

e

Nonverbal gestures (high level joint attention)

0

5

10

15

20

25

30

35

40

JASPER Wait

EntryExit

Perc

enta

ge o

f chi

ldre

n us

ing

poin

ts, s

how

s, g

ives

10/15/16

24

Change in Language

0

10

20

30

40

50

60

70

JASPER Wait list

ENTRYEXIt

% o

f chi

ldre

n us

ing

lang

uage

Community partnered participatory research models

(AIR-B Network)

Sustainable Change

Community-Academic

Partnership

Modular

Implementation Sustainability

10/15/16

25

We know some things

• Can change core impairments and improve meaningful outcomes

• Teaching others matters----parents, teachers, and peers

• Evidence of some generalization and longer term outcomes

We have much to learn!• Active ingredients

• dismantling studies

• Personalizing treatments• adaptive designs• delivering interventions just when needed

• Disseminating effective interventions into the community using hybrid implementation models

Appreciation to families and children who participate in studiesFunding sources• NICHD, NINDS, NIMH, HRSA, Autism Speaks, Merck Fund

Lab…..research assistants, graduate students, post-docs Amy Dominguez, Caitlin McCracken, Andy Schlink, Maria Pizzano, Kiana Krolick, Marina Mladenovic, Marcella Mattos, Nichole Tu, Alyssa Tan, Broghan Hedges, Marta Wirga, Devyn Tharnstrom, Jamie Crisostomo, Hilary Gould, Jonathan Panganiban, KC Berry, Alison Holbrook, Chrissy Kang, Belinda Williams, Jill Locke, Mark Kretzmann, Michelle Dean

Colleagues-UCLA—

Amanda Gulsrud,PhD, Stephanie Shire, PhD, Shafali Jeste, MD

Acknowledgements

10/15/16

26

airbnetwork.orgkasarilab.org