Adapting Enhanced Milieu Teaching for Young Children With. Communication Impairment. Ann P. Kaiser Vanderbilt University CEC 2015

Transcription

1 Adapting Enhanced Milieu Teaching for Young Children With Communication Impairment Ann P. Kaiser Vanderbilt University 1

2 Today s Talk Building a new genera>on of communica>on interven>ons Enhanced Milieu Teaching (EMT) Brief overview Summary of research evidence EMT ac>ve ingredients What children bring to EMT Addi>ons to core EMT Adap>ng EMT to fit learner characteris>cs Profiles of four popula>ons Adapta>ons to maximize social communica>on outcomes Research on adap>ons The Interven>on Tool Box: Tools for Adap>ng EMT Summary and Conclusions 2

3 3

4 What is Enhanced Milieu Teaching? EMT is a naturalistic, conversation- based intervention that uses child interests and initiations as opportunities to model and prompt language in everyday contexts. EMT can be used throughout the day as part of the everyday interactions. EMT is an evidence- based intervention with 20 years of research. EMT is an effective intervention. 4

5 EMT is effective Increases child use of language targets Vocabulary (Kaiser et al, l993; Scherer & Kaiser, 2010; Wright et al, 2013, Roberts & Kaiser, 2012; Kaiser & Roberts, 2012) Early syntac>c forms (Kaiser & Hester, l994) Moderately complex syntax (Warren & Kaiser, l986) Increases child frequency of communica>on (Warren et al, l994; Kaiser et al, l993) Generaliza>on across se\ngs, people, and language concepts (Warren & Bambara, l989; Goldstein & Mouse>s, l989; Kaiser & Roberts, 2012) Maintenance of newly learned targets (Warren & Kaiser, l986; Kaiser & Roberts, 2012) More effec>ve than drill- prac>ce methods (Yoder, Kaiser et Alpert, l991, Kaiser, Yoder, et al,1996) 5

6 EMT Active Ingredients Environmental arrangement to promote communica1on Play and engage Follow child s lead in play and ac1vity Respond to child communica1on Model target language in context Expand child communica1on** Use Time Delays to prompt requests or ini1a1ons Use Milieu Teaching Prompts to promote prac1ce Teach across sebngs, ac1vi1es and partners ** In 2 randomized trials, expansion has been the ingredient most highly correlated with child outcomes (Kaiser & Roberts, 2012; Roberts & Kaiser, under review) 6

7 EMT Example 7

8 What Children Bring to EMT Access to Input Intelligibility Fluency Person Object Ac1vity Mode Engagement Strategies Baseline Communi- ca1on Learning Strategies Rate Form Func1ons Transparency to partners Imita>on Auditory memory Efficiency 8

9 EMT Modi?ications to Fit What Children Bring Provide alterna1ve mode Signs SGD Teach partners mode Mode Engagement Strategies Teach play Increase person engagement Teach coordinated joint anen1on Baseline Communi- ca1on Learning Strategies Teach joint anen1on skills Support partner comprehension Teach imita1on Add discrete trials Increase dosage 9

10 Modi?ications of EMT JASPER + EMT [J- EMT] Teaches joint anen1on, symbolic play, regula1on JASPER + EMT + AAC [J- EMT+ SGD ; Words + Signs] Teaches joint anen1on, symbolic play, regula1on Includes speech genera1ng device or signs for input and output Phonological Emphasis + EMT [PE- EMT] Models speech targets Recasts for speech + Discrete trial training [Rescue protocol; preteaching protocol] - Reduce prompt complexity, number of prompts [Simplify] + Increase Dosage [Dosage] + Support Partners to use mode and EMT [Partner] 10

11 Population Speci?ic Modi?ications Popula1on Toddlers with Rec/Express Delay Modifica1ons Mode Engagement Learning Strategy Baseline Communica> on None None Support partner as teacher Down syndrome + Sign or SGD Teach play, object engagement, sustained ajen>on +Dosage Support partner comprehension and as teacher Clek + Speech targets None +Recast + Speech prac>ce Minimally Verbal ASD + SGD Teach play, engagement +Dosage +Rescue or Pre- teach Protocol: imita>on, recep>ve language Teach joint ajen>on skills Support partner as teacher 11

12 The Effects of a Parent- Implemented Language Interven>on for Children With Language Impairment Megan Y. Roberts, PhD, CCC- SLP Ann P. Kaiser, PhD

13 Toddlers with Receptive/Expressive Delays Communica1on Challenges Adapta1ons Problem behaviors Low rates of talking Low lexical diversity Increase ajen>on to posi>ve behavior, plan rou>nes, teach communica>ve alterna>ves Use responsiveness strategies to increase rate Model expanded vocabulary before and during early syntax targets 13

14 Kaiser, Camarata, & Roberts (2011) IES R324A090181; Roberts & Kaiser, 2015 Toddlers with Receptive/Expressive Delays Study Component Design Interven>on Measures Descrip1on Randomized Clinical Trial 45 Interven>on, 43 Control EMT with Play Skills 28 sessions ( 4 workshops, 14 clinic, 10 home across rou>nes) Parent + Therapist Pre, 6 wks, 12 wks, 18 wks (end of interven>on) Standardized, observa>onal, parent report Par>cipants Average age: 31 months Average Bayley Cogni>ve Score: 85 Gender: 83% male PLS- 4: 70 14

15 CEC 2015 Parent + Therapist EMT 15

16 Outcomes Intervention vs. Control d =0.3 d =0.3 d =0.3 d = Standard Score Expressive language (PLS- 4) Receptive language (PLS- 4) Expressive vocabulary (EOWPVT- 3) Receptive Vocabulary (PPVT- 4) Intervention Control 16

17 Outcomes Intervention vs Control: Number of Different Words Number of Different Words d = d = d = MCDI T: 264 C: 215 D =0.4 0 Start Month 1 Month 2 Month 3 Treatment Control 17

18 COMMUNICATION INTERVENTIONS FOR MINIMALLY VERBAL CHILDREN WITH AUTISM Kasari, Kaiser, Goods, Nieneld, Mathy, Landa, Murphy, & Almirall ( 2014). Clinical Trials Number: NCT This study was funded by Au>sm Speaks #5666, Characterizing Cogni>on in Nonverbal Individuals with Au>sm (CCNIA).

19 Children with Autism Study Component Design Interven>on Measures Descrip1on Randomized Clinical Trial; Mul>ple Baseline AAC, Verbal only EMT + Joint Ajen>on and Symbolic Play 48 sessions in the clinic (24 therapist only, 24 parent + therapist) Pre, Post, 6 months Standardized, observa>onal, parent report Par>cipants Average age: 6 years, 6 months Average Leiter: 61 Gender: 74% male PPVT: 32 Kasari, Kaiser, Landa et al, 2011 Au>sm Speaks

20 Children with Autism Communica1on Challenges Difficulty with joint engagement Adapta1ons Model and teach joint engagement behavior Few play skills and brief dura>on of play Reques>ng rather than commen>ng Model and teach play skills Model commen>ng, limit reques>ng Interfering behavior Very low rate spoken language Determine which behaviors are communica>ve; respond differen>ally Add SGD 20

21 Children with Autism Study Component Design Interven>on Measures Par>cipants Descrip1on Randomized Clinical Trial EMT + Joint Ajen>on and Symbolic Play (J- EMT) 48 sessions in the clinic (24 therapist only, 24 parent + therapist) with/ without SGD Pre, Post, 6 months Standardized, observa>onal, parent report 61 children with ASD Average age: 6 years, 6 months Average Leiter: Gender: 74% male PPVT: 32 Mn words at pre: Kasari, Kaiser, Landa et al, 2014 Au>sm Speaks 5566

22 Intervention Variations J- EMT Spoken Language Only J- EMT + SGD Speech Genera>ng Device - Dynavox or ipad Model using spoken language and SGD At least 50% of ujerances, 70% of expansions Child could speak or use SGD to respond and communicate 22

23 Use of SGD SGD available to the child Programmed pages for toys sets Used communica>vely with the child 50% of adult ujerance 70% of adult expansions Child could respond to prompts with either SGD or spoken language Embedded in JASPER- EMT interac>ons 23

24 Results 70% of whole group met criterion for response to treatment at week 12 Greater percentage of par>cipants in the JASP + EMT+ SGD group (77%) were early treatment responders than in the JASP +SGD group (62%) Par>cipants in the JASP + EMT +SGD group had : more Social Communica>ve Ujerances (SCU), greater Number of Different Word Roots (NDW), more comments (COM) than par>cipants in JASP+ EMT group Both groups shows gains over >me in SCU and NDW; only the +EMT+SGD group showed gains in COM JASP 24

25 Results At 12 Weeks Baseline 12 weeks Treatment Responders TSCU TND W TCO M TSCU TND W TCOM JASP+ EMT % JASP + EMT + SGD % (difference) % Effect Size P value NS NS NS NS Social communicative utterances (TSCU), Number of different word roots(tndw ) and number of comments (TCOM )were derived from a naturalistic language sample with a blinded clinician 25

26 Primary aim results for the primary outcome (TSCU).! JASP+EMT+SGD JASP+EMT Total Socially Communicative Utterances !!!!!! Week Open plo\ng characters denote observed means; closed denote model- es>mated means. Error bars denote 95% confidence intervals for the model- es>mated means.

27 Ongoing Studies ACE: Minimally Verbal Children with Au1sm year old children with ASD Fewer than 20 words RCT SMART design Discrete Trial Training (DTT) JASPER- EMT Adap>ve treatment Slow responders to either interven>on received tailored combina>on of DTT +JASPER- EMT Both with ipads Kasari, Kaiser, Smith, & Lord in progress; NIH Au>sm Center of Excellence C. Kasari,PI Minimally Verbal Preschoolers with Au1sm year old children with ASD Fewer than 20 words RCT Treatment vs Control JASPER- EMT with DDT pre- teaching core skills Individualized based on baseline imita>on and recep>ve skills With ipads Parent training Kaiser & Hampton, in progress HRSA 27

28 EFFECTS OF NATURALISTIC SIGN INTERVENTION ON EXPRESSIVE LANGUAGE OF TODDLERS WITH DOWN SYNDROME. Wright, C.A., Kaiser, A.P., Reikowsky, D.I., & Roberts, M.Y. (2013). Journal of Speech, Language, and Hearing Research, 56,

29 Children with Down Syndrome Communica1on Challenges Low rate of symbol infused joint ajen>on Poor ar>cula>on skills Poor auditory memory/ strong visual skills Poor generaliza>on across partners, se\ngs Adapta1ons Model communica>on in joint engagement episodes Teach sign + word as mode Model words + sign Teach with mul>ple partners, se\ngs, ac>vi>es 29

30 Children with Down Syndrome Study Component Design Interven>on Measures Par>cipants Descrip1on Mul>ple Baseline Single Subject EMT Words + Signs 24 sessions at home Therapist + Parent Pre, Post, ever 3 months Standardized, observa>onal, parent report Use of signs Gender: 1 male, 2 female Average age: 25 months (2.83) Average Mullen: 69 (8.04) Average PLS- Total Standard Score: (5.32) 30

31 Intervention Variation EMT Words + Signs Simplify and reduce promp>ng Parent training aker responding to prompts was established with therapist 31

32 EMT Words + Signs for Young Children 3 Toddlers with DS mos Multiple Baseline Design Taught by SLP in Clinic Generalization to home activities with parents Phase 2, teaching parents Wright et al, under review with DS Parent Outcomes % Matched Turns % Targets % Expansions % Correct Time Delay % Correct Prompting % Target JA models Child Outcomes Number of Symbols Used Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Pre Post Ryan 41% 90% 1% 74% 0% 48% 0% 83% 0% 94% 0% 82% Erin 36% 82% 7% 64% 3% 57% 0% 100% 0% 100% 0% 66% Jay 34% 95% 2% 75% 0% 60% 0% 100% 0% 100% 0% 80%

33 3-4.5 year olds with DS Parent + Therapist JASPER- EMT 48 sessions; 4 days/ week Childcare/preschool + Home ipads RCT treatment vs control Kaiser, Kasari & Wright in progress, John Merck Founda>on CEC 2015 J- EMT Words + ipads 33

34 EFFECTS OF EMT+PE ON THE LANGUAGE SKILLS OF YOUNG CHILDREN WITH CLEFT PALATE Kaiser, Scherer, Frey & Roberts (submijed) NIDCD 1R21DCOO

35 Children with Repaired Cleft Communica1on Challenges Low intelligibility Low rate of communica>on Oken shy, nonresponsive to promp>ng Adapta1ons Recast unintelligible ujerances, model phonological targets Use responsiveness strategies to increase rate of communica>ng Increase promp>ng aker sessions 35

36 Children with Repaired Cleft Study Component Design Interven>on Measures Descrip1on Pilot Randomized Clinical Trial 7 Interven>on, 9 Control PE- EMT 48 sessions in the clinic Therapist only Pre, Mid, Post, 3 months, 6 months Standardized, observa>onal, parent report Par>cipants Average age: 25 months Average Bayley Cogni>ve Score: 101 Gender: 69% male PLS- 4: 100 Scherer & Kaiser, 2010 NIDCD 1R21DC A1 36

37 Intervention Variation Phonological Emphasis PE- EMT Choose word targets with target sounds EMT for words Recast for phonological correctness Simplify prompt sequence 37

38 Children with Repaired Cleft Number of Different Words ES: d =.72 p = T0 T1 T2 Intervention Control 38

39 Children with Repaired Cleft 100% Percentage of Consonants Correct 80% 60% 40% 52% 34% 58% 57% 71% 78% 20% 0% T0 T1 T2 Intervention Control 39

40 Tools for Practice Skills Needed for Effective Intervention Fluent in the use of EMT* Skills for training parents and partners* Skilled in the addi>onal components added to EMT JASPER hjp:// jasper Sign Speech genera>ng device: use, management, assessment, instruc>on Speech recas>ng Discrete trial training (Smith, 2010) 40 * Informa>on available at hjp://kc.vanderbilt.edu/kidtalk/

41 Tools for Practice Assessment & Progress Monitoring Structured Play Assessment * Language Sample* Transcribed Coded for gesture Words, MLU, rate of ini>a>ons, rate of communica>on, consonant produc>on Speech assessments Arizona, PEEPS or language sample with consonants transcribed Baseline EMT session* Responsiveness to comments, TD, Prompts; Prompted and spontaneous verbal imita>on Use of targets Imita>on probe * Recep>ve language probe : recep>ve object and picture labeling Toy preference assessment (ongoing) * Informa>on available at hjp:// kc.vanderbilt.edu/kidtalk/ 41

42 Tools for Practice Progress Monitoring is Essential Every child presents unique challenges in implemen>ng EMT How child is responding to the interven>on is the test of whether the fit is right Adap>ve treatments are the 4 th genera>on of language interven>on Choose the best treatment based on assessment, baseline Monitor progress against benchmarks If needed, make adapta>ons Quick tools for monitoring: IGDI hjp:// Trackers for session data for therapist and child * 42

43 Tools for Practice Fidelity and Dosage Matter Is the interven>on being delivered at fidelity? Is the dosage of components within in the interven>on sessions sufficient? Models, expansions, prompts Is child responding to the ac>ve ingredients? Are sessions frequent enough, long enough? Do other partners need to be trained to increase dosage? 43

44 EMT is evolving with new individualized adapta>ons The core of the interven>on is always the social communica>ve connec>on between the child and partner The most important immediate outcome is increased communica>on Fine tuning interven>on to fit child characteris>cs can improve outcomes when combined with the core EMT and components are delivered at fidelity CEC 2015 Last words 44

45 References Kaiser, A.P., & Roberts, M.Y. (2013). Parent- implemented enhanced milieu teaching with preschool children with intellectual disabili>es. Journal of Speech, Language, and Hearing Research, 56, Kaiser, A.P., & Wright, C.A. (2013). Enhanced milieu teaching: Incorpora>ng AAC into naturalis>c teaching with young children and their partners. Perspec<ves on Augmenta<ve and Alterna<ve Communica<on, 22, Kaiser, A.P. & Roberts, M.Y. (2013). Parents as communica>on partners: An evidence based strategy for improving parent support for language and communica>on in everyday se\ngs. Perspec<ves on Language Learning and Educa<on, 20(3), Kasari, C., Kaiser, A.P., Goods, K., Nieneld, J., Mathy, J., Landa, R., Murphy, S., & Almirall, D. (2014). Communica>on interven>ons for minimally verbal children with au>sm: Sequen>al mul>ple assignment randomized trial. Journal of the American Academy of Child & Adolescent Psychiatry. Advance online publica>on. doi: /j.jaac Wright, C.A., Kaiser, A.P., Reikowsky, D.I., & Roberts, M.Y. (2013). Effects of naturalis>c sign interven>on on expressive language of toddlers with Down Syndrome. Journal of Speech, Language, and Hearing Research, 56, Roberts, M., & Kaiser, A. (2012). Assessing the effects of a parent- implemented language interven>on for children with language impairments using empirical benchmarks: A pilot study. Journal of Speech, Language, and Hearing Research, 55, Roberts, M.Y. & Kaiser, A. P. (In press) Early interven>on for toddlers with language delays: A randomized controlled trial. Pediatrics. Scherer, N., and Kaiser, A.P. (2010). Enhanced milieu teaching with phonological emphasis: Applica>on for children with CLP in treatment of sound disorders in children. Chapter to appear in Williams, L., McLeod, S., & McCauley, R. (Eds.), Interven<ons for Speech Sound Disorders in Children. Bal>more: Brookes Publishing. 45

46 Appreciation! KidTalk Research Team at Vanderbilt Jennifer Nieneld, Stephanie Jordan, Suzanne Thrower, Courtney Wright, Lauren Hampton, Kelly Windsor, Julie Bryant, Lizzy Fuller, Jodi Heidlage, Kim McCulla, Morgan Lueck Families and children who par>cipated in our studies Our collaborators Connie Kasari (UCLA) Danny Almirall (Univ of Michigan), Rebecca Landa (Kennedy Kreiger, Johns Hopkins Univ) Tristam Smith (Univ of Rochester) Nancy Scherer( ASU) Jennifer Frey ( GWU) Megan Roberts ( Northwestern Univ) Juliann Woods (FSU) Our funding agencies: IES, OSEP, NICHD, NINDS, John Merck Founda>on, HRSA For more informa>on Ann.Kaiser@Vanderbilt.edu This talk will be posted at hjp://kc.vanderbilt.edu/kidtalk/ Follow us on Facebook: Vanderbilt Kidtalk 46