Language Delay in Young Children Assessment and Intervention

Language Delay in Young Children: Assessment and Intervention Leslie Rescorla Bryn Mawr College lrescorl@brynmawr. edu Philadelphia Early Intervention Staff November, 2006

OVERVIEW w Part I • Normal language development • Screening for language delay • Outcomes of children with SLS & late talkers w Part II • Assessment and differential diagnosis • Case illustrations • Intervention approaches

Normal Language Acquisition w Language acquisition is a rapid process • 12 months: a few words • mama, dada, dog • 18 months: many children have 50 words and some are combining words • more juice, allgone car • 24 months: expect 50 words and combinations • mean: 150 -200 words on LDS, 2 -3 word phrases • telegraphic (this my cup, baby eat cookie) • 30 months: some morphemes • plurals, ing, in/on, possessive

Rapid Changes from 2. 5 - 4 Years w Short sentences by age 3 • using complete sentences (SVO) • can ask questions, use negatives • many morphemes: articles, -ed, -s, ’s w By age 4, producing complex sentences • • using clauses, multi-verb sentences auxiliary and copula forms present can carry on conversations can talk about past, present, and future

Language Delay w Language delay: a heterogeneous condition • not a unitary disease like chicken pox w Is language delay primary vs. secondary? • language problem secondary to other disorder? • rule out hearing impairment, mental retardation, and autism/PDD -- all associated w/ language problems • language delay is primary • toddlers: often called late talkers (LTs) • preschoolers: usually called SLI w Subtype • Expressive, receptive, or both?

Language Delay: Four Domains w Phonological: sound system of language • problem in phoneme perception, categorization • problem discriminating, processing rapidly? • problem in phoneme production • lack reduplicated babbling by 7 -10 months • small repertoire, few sounds - especially consonants w Lexical/semantic: vocabulary system • vocabulary delay: fewer than 50 words by 2 • most late talkers have <25 words by 24 -30 mos.

Four Domains, continued w Grammar: syntax & morphology • Few/no phrases @ 24 mos • 3 -4 word sentences only by 4 years • poor on rec & exp w/ compound, complex sentences • delay in morpheme acquisition • slow to acquire morphemes, especially verbal ones • telegraphic speech - omission in obligatory contexts w Pragmatics: communicative use of language • problems comprehending & indicating intent? • gesture, nonverbal, vocalization • problems in dialogue, conversation

Language Screening w Language delay is a public health problem • concerns about language delay prompt most referrals of young children • language delay associated with many other problems (MR, PDD, hearing impairment) • early language is a major risk factor for later learning and psychiatric problems • early identification & intervention desirable w Language Development Survey (LDS) • screening tool for language delay in toddlers • co-normed with Child Behavior Checklist

Language Development Survey w Language Development Survey (LDS) • checklist completed by parents of children 18 -35 months • 310 vocabulary words arranged by semantic category • Vocabulary score : sum of words used spontaneously • Does child combine words into phrases? • If yes, parent writes gives five longest & best phrases • mean number of words calculated for the five phrases w LDS norms • Vocabulary score norms - separately by gender – 18 -23, 24 -29, and 30 -35 months age groups • Mean Phrase Length – 24 -29 & 30 -35 months age groups with genders combined

Psychometric Properties of LDS w Reliability • Test-retest reliability: . 97 -. 99 • Cronbach’s alpha: . 99 w Validity • Correlations with expressive language tests: . 66 -. 87 • Sensitivity generally >80%, specificity > 85% • Predictive validity from age 2 (Rescorla, 2002) (N=59) • age 7 vocabulary =. 63 • age 8 grammar =. 41 • age 6 phonological awareness =. 44 • age 8 listening comprehension =. 43 • age 8 & age 9 reading =. 39 &. 34

Language Delay w Categorical vs. dimensional views • Categorical: SLI is discrete illness • at extreme: genetically based defect prevents mastery of syntax and morphology systems • Dimensional: language spectrum • delay, disorder are at tail of normal distribution • less severe - outgrow it earlier, fewer areas • more severe – persist longer, more areas affected w Late talkers vs. kids with SLI • categorical: LTs not SLI (most outgrow it) • dimensional: less severe on same spectrum

Language Endowment Spectrum NLI SLI LTs severe LI chronic LI Average language skills >average language skills superior, language skills

Outcomes of Children with SLI w Tomblin kindergarten LI study to 2 nd grade • 581 children identified in school-based study • 231 LI & 373 controls followed to 2 nd grade – did not require NV>V discrepancy (not SLI) • >1. 5 SD difference in language and reading • 52% of SLIs vs. 9% of controls had RD w Bishop age 4 SLI follow-up study • 44% of sample had “good outcome” • average in language & reading by 5 -6 and at age 8 • BUT: worse at 15 than controls on many language & reading measures • 56% “poor outcome” worse at all ages

Our Late Talker Research w Late Talkers • Bayley MDI > 85 • Reynell Receptive Language score within 3 months of CA • Reynell Expressive Language score at least 6 months below CA • < 50 words or no word combinations on LDS w Comparison Children • Bayley MDI > 85 • Reynell Receptive Language score within 3 months of CA • Reynell Expressive Language score within 3 months of CA • > 50 words and word combinations on LDS

Age 17 Follow-Up Groups INTAKE MEASURES Late Talkers Typical Developers (N=26) (N=23) Age in months 26. 62 25. 83 Hollingshead SES 53. 69 55. 74 Bayley Nonverbal 14. 15 15. 17 Reynell Receptive z . 15 1. 03 *** Reynell Expressive z -1. 62 . 36 *** LDS Total Vocabulary 24. 54 235. 17 *** No significant differences on intake measures within LT and TD groups between those seen and not seen at age 17 = no selective attrition

Language Outcomes Ages 5 -8

Phonological Outcomes Ages 5 -7

Reading Outcomes Ages 6 -9

Age 17 Outcomes w Measure LT TD d 15. 5** 113. 4** 11. 8** 109. 2 . 80. 86. 90. 39 w Vocabulary 13. 5 w Grammar 105. 1 w Verbal Memory 10. 0 w Reading 105. 4 w

Correlations Among Age 17 Measures Grammar Verbal Memory Rdg/Writing Vocabulary Grammar Verbal Memory . 65* . 39** . 63** . 70** . 57**. 38**

Our Outcome Conclusions • Most LTs have normal language skills by 5 • LTs are consistently inferior to comparison children in vocabulary, grammar, and verbal memory skills through age 17. • Late talkers weaker in decoding at 8 and 9 • LTs worse in comprehension at 13 • Age 17 vocabulary, grammar, verbal memory, and reading significantly intercorrelate • LDS at 2 predicts outcome quite well

Assessment of Language Skills w Communicative intent (PDD? S-P LD? ) • assess use of gestures, facial expression, grunts to communicate, range of intents expressed w Play & Imitation (PDD? , MR? ) • assess spontaneous pretend play, ability to imitate modeled play schemes, representational skills w Receptive language (R, E, or both delayed? ) • ability to comprehend phrases, words, gestures w Expressive language • use of phrases, words, vocalizations, babbling • phonetic repertoire, oral-motor skills

Assessment of Cognitive & Social/Emotional Development w Nonverbal ability (MR, develop. delay? ) • puzzles, blocks, visual-motor skills, problemsolving, object permanence skills) w Social relatedness (PDD? S-P LD? ) • eye contact, joint attention, reciprocity, sociability w Attention, Self-regulation (ADD? ) • concentration, attention, impulse control, frustration tolerance, & temper management

Assessment of Behavioral & Emotional Problems w Assess adjustment broadly • best to collect data from different informants • tap broad range of emotional/behavioral problems • use standardized, normed measures w CBCL/1. 5 - 5 & C-TRF • Normed in general population sample • 99 problem items (0, 1, 2 scale) (not true, somewhat or sometimes true, very true or often true • write in concerns & best things about the child • norms for empirically based syndromes & DSM scales

Relations Between Language Problems & Behavior Problems w Children > age 5 • Many studies show a link between language and behavior problems w Children at age 3 • language problems & behavior problems link found in two general population samples, but reduced when children w/ low IQ excluded w Children < age 3 • results mixed: association varies with size, diversity, yield, and referral status of sample

Case Example: Kenny Randall • Background information reported – Kenny Randall - age 30 months – nursery school teacher concerned because he did not interact much with other children and often talked to himself • Assessment process – parents completed the CBCL/1. 5 -5 – two teachers completed the C-TRF – ASEBA software scored the forms and produced the profiles

Cross-Informant Comparison • Informants have different views of the child • Informants who see the child in different contexts have different perspectives on the • ASEBA software provides systematic comparisons between informants • informants are compared on items, syndromes, DSM-oriented scales, Internalizing, Externalizing, and Total Problems scores • level of agreement between informants = correlations, which can be compared to those for similar pairs of raters in the normative sample

Case Example: Sam Harkin w Background information reported • • w Age 2 -3, youngest of four children Walked early, good language comprehension Communicative with gestures and noises Requests labels of objects and pictures Assessment process • 2 hour office visit • play activities with psychologist and SLP • LDS and CBCL completed by parents

Assessment Session Findings w Clinical observations • • • Social, interactive, responsive, playful child Comprehension of many requests, labels Attentive, cooperative, persistent Enthusiastic, appropriate, reciprocal play Good problem-solving skills with toys Learned new skills with manipulative toys Easily engaged in pretend play No intelligible words, some jargon/grunts Joint attention, spontaneous sharing interests with parents and with examiners

LDS/CBCL Findings w LDS findings • No words reported - possibly says “dada” • Below 15 th percentile for vocabulary • 39 words for boys, 83 words for girl • Below 20 th percentile for phrase length • <2. 35 for boys and girls 24 -29 months w CBCL results • In normal range on all CBCL syndromes & DSM-oriented scales • Only concern noted is expressive language delay

Case Example: Melissa Kane w Background information reported • • w Age 2 -10, oldest of two children Walked early, few phrases by 24 months Reported as active and difficult since infancy Reported to be irritable and overreactive Assessment process • 1 hour observation at daycare • 2 hour office visit • play and testing by early childhood team • CBCL/LDS completed by parents, C-TRFs completed by teachers

Assessment Findings w Daycare observations • Short attention span, active, fast-moving, intrusive • Noncompliant, controlling, demanding w Office observations • Good cognitive and play skills • Impulsive, oppositional, active, demanding w LDS/CBCL/TRF findings • LDS: 40% percentile for Vocabulary, 20 th percentile for Mean Phrase Length – slight expressive delay • Elevated scores from most informants on: • Emotionally Reactive, Attention Problems, and Aggressive Behavior syndromes • DSM-oriented Attention Deficit Hyperactivity Problems and Oppositional Defiant Problems scales

When To Provide Intervention? w Link intervention decisions to diagnosis • hearing impairment --intervene immediately to improve hearing and to foster language • MR/DD - intervene immediately and broadly to foster development in all areas • PDD - intervene immediately to foster social relatedness, language, play, & flexibility (Kenny) • SLI combined type - intervene early to foster receptive language • SLI expressive only - consult and watch till 2, intervene by 2 -6 if still not talking (Sam)

Intervention: Behavioral w Behavioral techniques • • • highly structured, data-based, decontextualized target sounds, words, phrases: drill to mastery set up discriminative stimuli for responding use shaping & prompt fading, build generalization use primary and secondary reinforcers • widely used with PDD & kids with severe behavioral impairments • may need if spontaneous communication, normal imitation, joint attention, & reciprocity are weak • can be effective in shaping communicative behaviors & increasing responses - but “unnatural”

Intervention: Play-Based w Intervention principles • model & promote language in natural interaction • build capacity for joint attention and communication through play with adult partner • stimulate words, phrases linked to child’s play activities and focus of attention • foster social connection, use mutual enjoyment to build imitative skills and reinforce communication • works best when kids have some appropriate play skills, social relatedness, & imitation skills • works best when receptive language skills and communicative intent are better • leads to more spontaneous use, when it works

Linking Assessment to Intervention • if MR: gear therapy to MA-appropriate skills • if PDD: focus on eye contact, receptive language, imitation, communicative intent, basic words • if ADD: use methods to foster & sustain attention, reduce impulsivity while shaping language skills • if socially withdrawn & anxious: use play-based, social communication methods to foster relaxed interaction, encouragement of intent • if receptive skills delayed: focus on reliable responding to nouns, verbs, adjectives & preps

Conclusions w Early identification & treatment • although many children with early language delay show spontaneous remission, many don’t • best to screen at 2, monitor till 30 months, and intervene by 3 if not making good progress w Use differential diagnosis framework • determine child’s profile of skills and deficits • determine if MR, PDD, R & E delay • assess general behavioral/emotional functioning w Link intervention to assessment results • choose degree of structure, intensity to fit the case • use more “natural” methods IF they will work