Childhood Apraxia of Speech
Conditions
Keywords
speech sound disorder
Brief summary
The study will test two modifications to speech therapy for 40 school-age children with childhood apraxia of speech to determine how to improve treatment outcomes. The study will compare treatment that includes real-time visual feedback of the tongue during speech using ultrasound vs traditional therapy that does not include ultrasound visual feedback. Additionally, some children will be treated with a traditional schedule of 2 sessions per week, whereas others will be provided with treatment that begins with intensive training (10 hours of therapy in one week) and progresses to a more distributed treatment schedule.
Detailed description
Childhood apraxia of speech is a developmental speech sound disorder that may lead to persisting speech errors, often despite years of treatment. Such impairments may lead to social, academic, and vocational limitations. Thus, there is a need to explore alternate treatment approaches. This study will explore how to improve speech sound production in school-age children with CAS by modifying a standard speech therapy program. Two adaptations to speech therapy will be tested in a 2 x 2 randomized group design. In Aim 1, a standard treatment schedule of 2 one-hour sessions per week will be compared against a treatment sequence beginning with an intensive therapy schedule (10 hrs of treatment in one week), which will then transition to a more distributed practice schedule. This treatment modification is intended to minimize erred practice between training sessions in the early stages of learning, then foster generalization through increased time between practice sessions. In Aim 2, a standard treatment that includes only verbal feedback to the client during speech practice will be compared against a treatment sequence that initially includes real-time ultrasound visual feedback of the tongue during speech, which will be faded over the course of treatment. Ultrasound visual feedback is designed to train articulatory movements. It may enhance children's understanding of the articulatory goals of speech movement patterns by comparing executed tongue movements with intended movements. Prior case reports and single subject experimental designs have shown that speech sound production may be enhanced by including ultrasound visual feedback, although no prior randomized group studies have been conducted. Beside these modifications, the other aspects of treatment will be held constant. Outcomes will be evaluated by tracking changes in percent consonants correct from a large speech sample, scored by individuals who are blind to treatment status. The four groups will be compared to determine the extent to which speech sound therapy can be enhanced through a treatment sequence that begins with intensive practice and/or with ultrasound visual feedback.
Interventions
BEHAVIORALSpeech Motor Chaining without Biofeedback
These procedures target sound sequences (consonant-vowel, consonant-consonant, or vowel-consonant). Sessions begin with Pre-practice to elicit the target sounds, with verbal cueing and shaping strategies. The Practice component then includes chaining that is response-contingent. Participants practice in blocks of 6 consecutive trials beginning at the syllable level. If 5/6 are correct, the participant advances to monosyllabic word practice, then multisyllabic word practice, phrase practice, and sentence practice (with the target syllable embedded within each level of complexity). If fewer than 5/6 trials are correct, a different syllable with the target sound pattern is practiced next. Verbal feedback is faded from 5 of 6 trials at the syllable level to only 3 of 6 trials at the sentence level. For more description, see http://speechproductionlab.syr.edu/Resources%20for%20Researchers.html
BEHAVIORALBiofeedback
Real-time images of the tongue are made available using ultrasound placed beneath the chin. Participants practice speech movements and can be cued to modify their tongue shape or position to achieve clearer speech. Practice structure is similar to the Speech Motor Chaining procedures, but with the addition of a visual reference.
BEHAVIORALDistributed Practice
2 sessions per week for 10 weeks
BEHAVIORALMassed --> Distributed Practice
Week 1: 10 hours of treatment Week 2: 3 hours of treatment Week 3: 3 hours of treatment Week 4: 2 hours of treatment Week 5: 2 hours of treatment
Sponsors
National Institute on Deafness and Other Communication Disorders (NIDCD)
Syracuse University
Study design
Allocation
RANDOMIZED
Intervention model
FACTORIAL
Primary purpose
TREATMENT
Masking
SINGLE (Outcomes Assessor)
Masking description
Individuals who conduct the phonetic transcriptions that will be used for the primary outcome data will be blinded as to whether the recordings were collected before or after treatment, and also blinded as to the group assignment of the participant.
Intervention model description
2 x 2 design Scheduling: Mass--\>Distributed vs. Distributed only Visual feedback: Sequenced Ultrasound Biofeedback vs. No biofeedback
Eligibility
Sex/Gender
ALL
Age
9 Years to 17 Years
Healthy volunteers
No
Inclusion criteria
* Native English speakers who hear English as the dominant language in their home setting. * Must pass a hearing screening. * Score at or better than -2 standard deviations from the mean on the Matrix Reasoning Task of the Wechsler Abbreviated Scale of Intelligence - 2nd Edition (WASI-2; t-score ≥ 30), Peabody Picture Vocabulary Test - 4th Edition (PPVT-4; standard score ≥ 70), and the Following Directions subtest of the Clinical Evaluation of Language Fundamentals - 5th Edition (CELF-5; scaled score ≥ 3). * Goldman-Fristoe Test of Articulation - 3rd Edition (GFTA-3) percentile ≤ 5th). * A diagnosis of CAS will be verified based on a polysyllable picture naming task, diadochokinetic task (puh-tuh-kuh), and syllable repetition task.
Exclusion criteria
* Parent report or direct evaluation reveals oral-facial structural abnormalities (e.g., cleft palate). * Parental report of neurobehavioral disorders (e.g., autism spectrum disorders, ADHD, obsessive-compulsive disorder), or vision problems that are corrected with glasses/contacts. * Fail a hearing screening, or failure to meet criteria listed above as Inclusionary
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Speech Sound Accuracy | 10 weeks from the start of treatment | Percent consonants correct for target sounds: The primary outcome measure was percent correct for each participant's target sound-positions in untreated phrase. Participants imitated 20 pre-recorded phrases, each containing the target sound pattern 2 times per stimulus (e.g., for /l/ onset, leave the location), resulting in 40 attempts at each sound-position. For each session, 3 transcribers independently transcribed in Phon software (Hedlund & Rose, 2022) and accuracy was averaged across transcribers. Instances where IPA symbols for the Actual transcription differed from the Target transcription was scored as incorrect. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Biofeedback, Massed->Distributed, Face-to-Face Sequenced biofeedback Mass Practice--\> Distributed Scheduling
Biofeedback: Real-time images of the tongue are made available using ultrasound placed beneath the chin. Participants practice speech movements and can be cued to modify their tongue shape or position to achieve clearer speech. Practice structure is similar to the Speech Motor Chaining procedures, but with the addition of a visual reference.
Massed --\> Distributed Practice: Week 1: 10 hours of treatment Week 2: 3 hours of treatment Week 3: 3 hours of treatment Week 4: 2 hours of treatment Week 5: 2 hours of treatment
Modality: Face-to-Face | 9 |
| No Biofeedback, Distributed, Face-to-Face Speech Motor Chaining with no biofeedback. 2 sessions/wk for 10 weeks
Speech Motor Chaining without Biofeedback: These procedures target sound sequences (consonant-vowel, consonant-consonant, or vowel-consonant). Sessions begin with Pre-practice to elicit the target sounds, with verbal cueing and shaping strategies. The Practice component then includes chaining that is response-contingent. Participants practice in blocks of 6 consecutive trials beginning at the syllable level. If 5/6 are correct, the participant advances to monosyllabic word practice, then multisyllabic word practice, phrase practice, and sentence practice (with the target syllable embedded within each level of complexity). If fewer than 5/6 trials are correct, a different syllable with the target sound pattern is practiced next. Verbal feedback is faded from 5 of 6 trials at the syllable level to only 3 of 6 trials at the sentence level.
Distributed Practice: 2 sessions per week for 10 weeks
Modality: Face-to-Face | 9 |
| Biofeedback, Distributed, Face-to-Face Sequenced biofeedback, 2 sessions/wk for 10 weeks
Biofeedback: Real-time images of the tongue are made available using ultrasound placed beneath the chin. Participants practice speech movements and can be cued to modify their tongue shape or position to achieve clearer speech. Practice structure is similar to the Speech Motor Chaining procedures, but with the addition of a visual reference.
Distributed Practice: 2 sessions per week for 10 weeks
Modality: Face-to-Face | 10 |
| No Biofeedback, Massed-> Distributed, Face-to-Face Speech Motor Chaining with no biofeedback. Mass Practice--\> Distributed Scheduling
Speech Motor Chaining without Biofeedback: These procedures target sound sequences (consonant-vowel, consonant-consonant, or vowel-consonant). Sessions begin with Pre-practice to elicit the target sounds, with verbal cueing and shaping strategies. The Practice component then includes chaining that is response-contingent. Participants practice in blocks of 6 consecutive trials beginning at the syllable level. If 5/6 are correct, the participant advances to monosyllabic word practice, then multisyllabic word practice, phrase practice, and sentence practice (with the target syllable embedded within each level of complexity). If fewer than 5/6 trials are correct, a different syllable with the target sound pattern is practiced next. Verbal feedback is faded from 5 of 6 trials at the syllable level to only 3 of 6 trials at the sentence level. For more description, see http://speechproductionlab.syr.edu/Resources%20for%20Researchers.html
Massed --\> Distributed Practice: Week 1: 10 hours of treatment Week 2: 3 hours of treatment Week 3: 3 hours of treatment Week 4: 2 hours of treatment Week 5: 2 hours of treatment
Modality: Face-to-Face | 9 |
| No Biofeeedback, Distributed, Telepractice Speech Motor Chaining with no biofeedback. 2 sessions/wk for 10 weeks
Speech Motor Chaining without Biofeedback: These procedures target sound sequences (consonant-vowel, consonant-consonant, or vowel-consonant). Sessions begin with Pre-practice to elicit the target sounds, with verbal cueing and shaping strategies. The Practice component then includes chaining that is response-contingent. Participants practice in blocks of 6 consecutive trials beginning at the syllable level. If 5/6 are correct, the participant advances to monosyllabic word practice, then multisyllabic word practice, phrase practice, and sentence practice (with the target syllable embedded within each level of complexity). If fewer than 5/6 trials are correct, a different syllable with the target sound pattern is practiced next. Verbal feedback is faded from 5 of 6 trials at the syllable level to only 3 of 6 trials at the sentence level.
Distributed Practice: 2 sessions per week for 10 weeks
Modality: Telepractice | 10 |
| No Biofeedback, Massed->Distributed, Telepractice Speech Motor Chaining with no biofeedback. Mass Practice--\> Distributed Scheduling
Speech Motor Chaining without Biofeedback: These procedures target sound sequences (consonant-vowel, consonant-consonant, or vowel-consonant). Sessions begin with Pre-practice to elicit the target sounds, with verbal cueing and shaping strategies. The Practice component then includes chaining that is response-contingent. Participants practice in blocks of 6 consecutive trials beginning at the syllable level. If 5/6 are correct, the participant advances to monosyllabic word practice, then multisyllabic word practice, phrase practice, and sentence practice (with the target syllable embedded within each level of complexity). If fewer than 5/6 trials are correct, a different syllable with the target sound pattern is practiced next. Verbal feedback is faded from 5 of 6 trials at the syllable level to only 3 of 6 trials at the sentence level. For more description, see http://speechproductionlab.syr.edu/Resources%20for%20Researchers.html
Massed --\> Distributed Practice: Week 1: 10 hours of treatment Week 2: 3 hours of treatment Week 3: 3 hours of treatment Week 4: 2 hours of treatment Week 5: 2 hours of treatment
Modality: Telepractice | 9 |
| Total | 56 |
Baseline characteristics
| Characteristic | Biofeedback, Massed->Distributed, Face-to-Face | Total | No Biofeedback, Massed->Distributed, Telepractice | No Biofeeedback, Distributed, Telepractice | No Biofeedback, Massed-> Distributed, Face-to-Face | Biofeedback, Distributed, Face-to-Face | No Biofeedback, Distributed, Face-to-Face |
|---|---|---|---|---|---|---|---|
| Age, Continuous | 10.31 years STANDARD_DEVIATION 1 | 10.96 years STANDARD_DEVIATION 1.9 | 11.16 years STANDARD_DEVIATION 2.3 | 11.05 years STANDARD_DEVIATION 2.4 | 10.0 years STANDARD_DEVIATION 1.1 | 12.2 years STANDARD_DEVIATION 2.1 | 10.85 years STANDARD_DEVIATION 1.7 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 3 Participants | 9 Participants | 3 Participants | 0 Participants | 0 Participants | 2 Participants | 1 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 6 Participants | 46 Participants | 6 Participants | 9 Participants | 9 Participants | 8 Participants | 8 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 1 Participants | 0 Participants | 1 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Asian | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Black or African American | 0 Participants | 1 Participants | 0 Participants | 0 Participants | 0 Participants | 1 Participants | 0 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 9 Participants | 2 Participants | 3 Participants | 2 Participants | 1 Participants | 1 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 1 Participants | 0 Participants | 1 Participants | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) White | 9 Participants | 45 Participants | 7 Participants | 6 Participants | 7 Participants | 8 Participants | 8 Participants |
| Region of Enrollment United States | 9 participants | 56 participants | 9 participants | 10 participants | 9 participants | 10 participants | 9 participants |
| Sex: Female, Male Female | 3 Participants | 17 Participants | 2 Participants | 4 Participants | 3 Participants | 2 Participants | 3 Participants |
| Sex: Female, Male Male | 6 Participants | 39 Participants | 7 Participants | 6 Participants | 6 Participants | 8 Participants | 6 Participants |
| Speech Sound Accuracy | 20.2 Percent correct STANDARD_DEVIATION 7.3 | 24.3 Percent correct STANDARD_DEVIATION 7.8 | 35.5 Percent correct STANDARD_DEVIATION 7.8 | 28.9 Percent correct STANDARD_DEVIATION 8.1 | 23.2 Percent correct STANDARD_DEVIATION 7.4 | 17.0 Percent correct STANDARD_DEVIATION 4.3 | 21.2 Percent correct STANDARD_DEVIATION 12.1 |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk | EG003 affected / at risk | EG004 affected / at risk | EG005 affected / at risk |
|---|---|---|---|---|---|---|
| deaths Total, all-cause mortality | 0 / 9 | 0 / 9 | 0 / 10 | 0 / 9 | 0 / 10 | 0 / 9 |
| other Total, other adverse events | 0 / 9 | 0 / 9 | 0 / 10 | 0 / 9 | 0 / 10 | 0 / 9 |
| serious Total, serious adverse events | 0 / 9 | 0 / 9 | 0 / 10 | 0 / 9 | 0 / 10 | 0 / 9 |
Outcome results
Primary
Speech Sound Accuracy
Percent consonants correct for target sounds: The primary outcome measure was percent correct for each participant's target sound-positions in untreated phrase. Participants imitated 20 pre-recorded phrases, each containing the target sound pattern 2 times per stimulus (e.g., for /l/ onset, leave the location), resulting in 40 attempts at each sound-position. For each session, 3 transcribers independently transcribed in Phon software (Hedlund & Rose, 2022) and accuracy was averaged across transcribers. Instances where IPA symbols for the Actual transcription differed from the Target transcription was scored as incorrect.
Time frame: 10 weeks from the start of treatment
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Biofeedback, Massed->Distributed, Face-to-Face | Speech Sound Accuracy | 54.1 Percent correct | Standard Error 13.5 |
| No Biofeedback, Distributed, Face-to-Face | Speech Sound Accuracy | 32.5 Percent correct | Standard Error 14.1 |
| Biofeedback, Distributed, Face-to-Face | Speech Sound Accuracy | 26.8 Percent correct | Standard Error 10.2 |
| No Biofeedback, Massed-> Distributed, Face-to-Face | Speech Sound Accuracy | 42.1 Percent correct | Standard Error 11.1 |
| No Biofeeedback, Distributed, Telepractice | Speech Sound Accuracy | 32.50 Percent correct | Standard Error 7 |
| No Biofeedback, Massed->Distributed, Telepractice | Speech Sound Accuracy | 43.0 Percent correct | Standard Error 7.7 |
Comparison: Main effect of biofeedback at 10 weeksp-value: 0.13295% CI: [-23.2, 3.1]Mixed Models Analysis
Comparison: Main effect of Practice Distribution at 10 weeksp-value: 0.02895% CI: [-25.9, -1.5]Mixed Models Analysis
Comparison: Main effect comparing telepractice vs face-to-face treatmentp-value: 0.18795% CI: [-24.9, 5]Mixed Models Analysis
Comparison: Interaction of biofeedback and practice distribution at 10 weeksp-value: 0.12595% CI: [-5.8, 46.3]Mixed Models Analysis