Maximizing Outcomes for Preschoolers With Developmental Language Disorders

Maximizing Outcomes for Preschoolers With Developmental Language Disorder: Testing the Effects of a Sequentially Targeted Naturalistic Intervention.

The objective of the proposed study is to evaluate the efficacy of the Enhanced Milieu Teaching-Sentence Focus (EMT-SF) intervention, implemented by caregivers and interventionists, relative to a control condition enrolling 108 30-month-old children and their caregivers. The central hypothesis is that intervention will result in better overall child language skills at 49 months of age.

A multi-site, phase 2, randomized clinical trial will be used to determine whether communication support strategies are effective for improving language outcomes in children with emergent developmental language disorder. At study entry, 108 children with emergent developmental language disorder (DLD) at 30 months of age will be randomly assigned 1:1 to either the EMT-SF treatment condition or a Business as Usual (BAU) control group. The control group is necessary to determine the efficacy of the EMT-SF intervention. The EMT-SF group is necessary to evaluate the effects of systematically teaching caregivers to use these strategies. Because all children in the study have language delays that will make them eligible to receive the early intervention services through the state early intervention program, children in both experimental conditions will receive state-provided community-based intervention according to their Individualized Family Service Plan - the current standard of care or from private speech language therapy providers. Children in the EMT-SF condition will receive an additional 18 months of interventionist plus caregiver-implemented intervention sessions. Children in both groups will be assessed at the start of the study and every 3 months until the child is 49 months old. The goal is to enroll all children at 30 months of age and provide a minimum of 60 of the planned 66 sessions of intervention to each child in the treatment condition; however, variability in age at study entry (e.g., 30 months), intervention dosage, and number of assessment data points will be addressed in the statistical analysis.

A stratified randomized clinical trial design will be used with site (Vanderbilt, Northwestern) and biological sex (boys, girls) as stratification factors and EMT-SF or business-as-usual (BAU) control as a randomization factor.

The study intervention is a behavioral intervention which will include individually teaching caregivers to use the intervention strategies from the Enhanced Milieu Teaching-Sentence Focus (EMT-SF) intervention using a manualized protocol (Teach-Model-Coach-Review). Caregivers will participate in 66 intervention sessions across 18 months, targeting vocabulary and grammar as well as the transition to decontextualized language.

Caregivers in the control group will participate in community-based intervention services and receive the same printed intervention instructions, books, and toys at the same intervals as the treatment group, but will not receive EMT-SF intervention.

Caregivers in the treatment group (EMT-SF) will receive 66 EMT-SF intervention sessions at home over 18 months. They will be taught the intervention strategies using a manualized protocol (Teach-Model-Coach-Review).

CELF-P2 core language score at 49 months. The core language score includes summing the scaled scores from the following subtests: Word Structure (raw score min = 0; max = 24), Sentence Structure (raw score min = 0; max = 22), and Expressive Vocabulary (raw score mix = 0; max = 40), and deriving a standardized composite score (min = 45; max = 155). These three subtests provide the best diagnostic accuracy between children with and without developmental language disorders. Thus, the Core Language Score is considered the most representative reflection of a child's overall language skills. Higher standard scores represent better language outcomes.

Slope of total number of different spoken word (NDW) roots from a Language Sample measured every 3 months from 30 to 49 months derived from a 20-minute language sample. Higher scores represent better language outcomes.

Change in a latent variable of number of different verbs, number of different subjects, tense/agreement productivity, index of productive syntax, clauses per utterance derived from a 20-minute language sample. Higher scores represent better language outcomes.

Expressive Vocabulary Test (EVT-3) standard score (min = 20; max = 160). Higher scores represent better outcomes.

Structured Photographic Expressive Language Test (SPELT P-2) raw score (min = 0; max = 40). Higher scores represent better outcomes.

Structured Photographic Expressive Language Test (SPELT P-2) raw score (min = 0; max = 40). Higher scores represent better outcomes.

Test of Early Grammatical Impairment (TEGI) raw score for subscales: Third Person Singular Probe subscale (min = 0; max = 10) and the Past Tense Probe subscale (min = 0; max = 18). Higher scores represent better outcomes.

Test of Early Grammatical Impairment (TEGI) raw score for subscales: Third Person Singular Probe subscale (min = 0; max = 10) and the Past Tense Probe subscale (min = 0; max = 18). Higher scores represent better outcomes.

Focus on the Outcomes of Children Under Six (FOCUS) total raw score (min = 0, max = 350). Higher scores represent better outcomes.

Total number of utterances that contain decontextualized language from a 20-minute Language Sample. Higher scores represent better language outcomes.

Information score (min = 0; max = 52) from the Renfrew Bus Story. Higher scores represent better outcomes.

Child Behavior Checklist Internalizing T score (min = 29; max = 100). Lower scores represent better outcomes.

Inclusion Criteria: Parent or caregiver (mother, father, grandparent) willing to participate in caregiver training over the full 18 months of the study Normal non-verbal cognitive abilities Receptive and expressive language delay: English as the only language spoken to the child in the home Provision of signed and dated informed consent form Attempts to imitate 10 words Exclusion Criteria: Diagnosis of a neurodevelopmental disability other than DLD (e.g., Down syndrome, ASD, intellectual disability) Caregiver report of a major medical condition (e.g., seizures, cancer, stroke, traumatic head injury, cleft lip/palate, cerebral palsy, legally blind, any genetic diagnosis associated - - Caregiver report of preterm birth (i.e., < 37 weeks gestation) Caregiver report of hearing impairment or audiological testing indicating hearing thresholds > 20dB Caregiver report or direct observation of any problems chewing, sucking through a straw, or blowing bubbles.

Transcripts from caregiver-child and examiner-child language samples will also be donated to the Child Language Data Exchange System (http://childes.psy.cmu.edu), an international repository of child language data. Access to the data will be openly available. Researchers who access the transcripts will be expected to abide by the established guidelines for use of TalkBank data http://talkbank.org/share/irb/options.html.

Data and associated documentation available to users only under a data-sharing agreement that provides for a commitment: (1) to use the data only for research purposes and not to identify any individual participant; (2) to secure the data using appropriate computer technology; (3) to destroy or return the data after analyses are completed; and (4) to cite the grant and key publications describing the database and measures in any resulting presentations and publications.