Engaging Autism: Implications for Successful School Adaptation Connie Kasari, PhD University of California, Los Angeles AIR-B --Autism Intervention Research Network for Behavioral Health
Dec 30, 2015
Engaging Autism: Implications for Successful School Adaptation
Connie Kasari, PhD
University of California, Los Angeles
AIR-B --Autism Intervention Research Network for
Behavioral Health
Today’s Talk
1. Active ingredients of interventions Factors that matter—why the intervention
works Research chipping away at these issues
2. Focus on core deficits DSM V---still the same core deficits—social
and communication 3. Intervention studies that are based in
school settings
Active Ingredients
Approach ABA most common Many types and interpretations
Dose Intensity (hours per week for how long?) Density?
Agent of Change—parent, teacher, therapist, etc
Content Context
Active Ingredients
Approach ABA most common Many types and interpretations
Dose Intensity (hours per week for how long?) Density?
Agent of Change—parent, teacher, therapist, etc
Content Context
CONTEXT—Why Schools?
Kids spend the most part of the day in school Limited evidence that school programs utilize
evidence based practices Schools often use eclectic approaches
Eclectic is good---when informed not random Often random; driven by outside forces; convenient;
untested For mainstreamed children, interventions may be
absent Parents spend a lot of time driving children to
therapies outside of school Often for attention they are not getting in school
Critical need to bring general education into the conversation
Conducting research in schools
Not easy…… Schools have additional layers of
complexity State mandated curricula District or building level procedures Multiple interruptions and pressures that
are not in any manual Suspicion about researchers, and research
in general
Evidence based interventions in schools
Important to remember (Weisz, 2004)
Vast majority of children have never been tested in any outcome study
Of the many treatments available, only a fraction have ever been tested in research
Although particular programs have their disciples, most therapists/teachers do not adhere to any one treatment but create their own based on previous training, supervision and experiences.
Researchers have their own tradition of moving interventions into schools
Efficacy research In the lab studies---highly controlled with
specific types of children Partial effectiveness
Researchers in natural setting (home or school) Effectiveness research
School staff who implement research under close supervision
Deployment Community partnered research methods
Issues we need to consider to bring interventions to scale in schools……
Need dismantling studies Figure out the active ingredients—what is important to
an intervention (Comprehensive interventions necessary, but not all aspects
are important)
An active ingredient can lead to a module---teachers more likely to add a module than completely change practices (e.g. Chorpita, 2004; Weisz et al, 2011)
Partial effectiveness research from the beginning to determine active ingredients Reduces time from lab to school
Examples from our work—Focusing on Core Deficits
Lab based efficacy studies Joint attention and play in preschool children
Partial Effectiveness Studies Studies conducted in schools by research staff
Peer interaction studies Effectiveness studies
Teach staff to deliver Teacher/paraprofessional mediated studies
Community Partnered research—the future
11
General Theme: Engagement as Critical Intervention Target
Issues around engagement consistent across age….. What are behavioral signs of engagement?
Shared attention and affect Joint attention Social play with others Conversation
EXAMPLE 1Comparative Efficacy Study: Focus on Core Deficits
Joint AttentionInitiations: Point to share,
Show
Symbolic Play JASPER Model
JASPER model
All children were in same preschool program ABA based, 30 hours per week Hospital based school program 58, 3-4-year-old children
Randomly assigned children to 1 of 3 conditions Joint attention, Symbolic play, Control Short term (6 weeks), every day intervention Expert therapists (children seen outside of class)
Goal to improve core social communication skills and predict to language a year later
Changes obtained in joint attention, play and language outcome one year later
Kasari, Freeman & Paparella, 2006, JCPPKasari, Paparella, Freeman, & Jahromi, 2008, JCCP
What We Learned
Treatment protocols evolve as you learn more about how they work……
Learn about active ingredients, potential mechanisms for why the intervention works…..
Also learn what might not work….. Clinical significance
Effectiveness Trial in Preschools
Much adaptation may be necessary to bring treatment to real world contexts
Challenges are the classroom environment Some teachers do not work directly with children If they do, sustaining focus in the midst of distractions Collecting data, not a preferred task
Two examples in preschool environment UCLA study; Norway trial
Targeted JASPER Intervention with Teachers as the Mediators (pilot with 16 teachers)
Lawton & Kasari, in press, JCCP
Joint Attention Intervention:Replication with 58 children and teacher mediated (Kaale, Smith,
Sponheim, 2011)
What We Learned
Buy in critical Important to teach teachers what ‘change
processes’ they needed to effect; not just techniques or materials used
Important so they can apply to the next child who may be quite different from the first
Also important to establish where the same strategies can benefit all children
Example 2: Partial Effectiveness Trial of Peer Interventions in Mainstream Schools Partial Effectiveness ---testing the
intervention in the context with real world participants from the beginning
School based comparative efficacy study 60 HFA first to fifth graders (30 different
schools in Los Angeles) Testing common interventions—peer
mediated and child assisted
Peer Related School Intervention StudyPeer Intervention Study in Schools
Peer Mediated ApproachChild Assisted Approach
Kasari, Rotheram-Fuller, Locke, & Gulsrud, 2011, JCPP
Summary of UCLA Peer Study
CHILD(1:1)
PEER(3 peers)
NO Treatment
CHILD+PEER
• PEER Mediated Interventions > CHILD Assisted Interventions
• Primary Outcome• Social Network
Salience (d=.79)
6 WEEK TREATMENT (12 SESSIONS)
12 WEEK FOLLOW UP
Kasari, Rotheram-Fuller, Locke, & Gulsrud, 2011, JCPP
Alejandro (4)
Giovanni (6)
Lucas (2)
Leah (7)
Nora (2)
Olivia (9)
Alicia (4)
Adam (3)
Elijah (6)
Charlotte (8)
Cory (7)Larry (5)
Leah (4)Ella (7)
Sam (4)
Miguel (4)
Tomas (4)
Magnolia (3) Nola (1)
Isolate: Nicholas (3), Nolan (4)
4.5
5.5
2
7.5
5
8
Summary of UCLA Peer Study
CHILD(1:1)
PEER(3 peers)
NO Treatment
CHILD+PEER
• Other Findings favoring Peer Mediated Interventions:
• Number of Received Friend Nominations (d=74)
• Less isolated on playground (growth curves over tx)
• Improved rating of social skills (by Teachers) (d=.44)
6 WEEK TREATMENT (12 SESSIONS)
12 WEEK FOLLOW UP
Other Findings
What about children who are doing well (socially connected)? 20% of children had a reciprocal friendship These same children had higher social network status They were NOT any more engaged on the playground?
Playground a difficult environment—requires specific intervention
1:1 assistant as solution in school setting In this study, children with a 1:1 were less engaged
Deployment Focused Model
Idea would be to bring treatment research into practice settings early (not the last phase)
Consider sequential process in the setting, from the beginning 1. treatment that can work in everyday
practice 2. assess treatment outcome in practice 3. examine moderators and mediators in
context
Considerations in bringing interventions into practice settings
Practitioner concerns about relevance of EBT To their situations Their children Their families
Alliance and buy in critical….. Researchers must understand the
context (the particular schools) in which they work…..
Example 3: Involving School Staff
Using transitions to facilitate peer interactions, language and behavior regulationAnd particularly to work in the playground setting
What We Still Don’t Know but are Attempting to Find Out
Expansion to populations we know less about…..
Low income, underserved families Minimally verbal
Treatment experienced ‘tx resistant’ children Observations of minimally verbal 5 to 8 year olds
in class….41% of time unengaged; 18% jointly engaged; more time on break than academically engaged
Need for effective school based interventions that academically and socially challenge children
Conclusions---Next steps
Natural time course of treatments from research to practice (too long!)
Schools are where children with ASD spend the most time and this is where interventions should take place
Researchers need to collaborate with school staff to move the needle forward in bringing evidence based interventions to scale
We need to measure child outcomes of school interventions—what works, what doesn’t
Next steps are to deploy interventions into the community that can be sustained
Acknowledgements
CollaboratorsGail Fox AdamsYa-Chih ChangLauren ElderAmy FullerKelly Stickles GoodsAmanda GulsrudNancy HuynhEric IshijimaMark KretzmannKelley KruegerJill LockeCharlotte MucchettiStephanie Patterson
• Funding• Autism Speaks• NIH• HRSA; Autism Intervention
Network for Behavioral Health—AIR-B
• Private donors