McCabe et al. (2014)

February 24, 2021

EBP THERAPY ANALYSIS for 

Single Case Designs

NOTES:  

•  The summary of the intervention procedure(s) can be viewed by scrolling about two-thirds of the way down on this page.

Key:

 C =  Clinician

 CAS =  Childhood Apraxia of Speech

CTPP =  Comprehensive Test of Phonological Processing 

 EBP =  evidence-based practice

 KP =  used knowledge of performance 

 NA = not applicable 

 P =  Patient or Participant

 pmh =  Patricia Hargrove, blog developer

 PCC =  Percent Consonants Correct 

 PND =  Percentage of Nonoverlapping Data 

 PPVT = Peabody Picture Vocabulary Test 

 PVC =  Percent Vowels Correct

 ReST =  Rapid Syllable Transition Training

 SLP =  speech–language pathologist

 SS =  Standard Score

TAP =  Test of Auditory Processing

 WNL =  within normal limits 

SOURCE:  McCabe, P., Macdonald-D’Silva, A. G., van Rees, L., Ballard, K. J., & Arciuli, J. (2014).  Orthographically sensitive treatment for dysprosody in children with childhood apraxia of speech using ReST intervention. Developmental Neurorehabilitation, 17 (2), 137-146. DOI: 10.3109/17518423.2014.906002 

REVIEWER:  pmh

DATE: February 24, 2021

ASSIGNED OVERALL GRADE:  B+  The highest grade possible based on the design of this investigation is A-   (Single-case experimental design). The Assigned Overall Grade is not a judgment about the quality of the intervention; it is rating of the evidence presented in the investigation.

TAKE AWAY:  This investigation explored the effectiveness of Rapid Syllable Transition Training (ReST) on the production of lexical stress in the speech of 4 children diagnosed with Childhood Apraxia of Speech (CAS). Using an AB design, the investigators identified changes from pre- to post- treatment and retention, 4 weeks after the termination of the intervention. The findings suggest that lexical stress improved as the following the intervention and that the changes are maintained.

1.  What was the focus of the research?  Clinical Research  

2.  What type of evidence was identified?                              

–  What  type of single subject design was used? Single Subject Experimental Design with Specific Clients: AB 

  What was the level of support associated with the type of evidence?  Level = A-

3.  Was phase of treatment concealed?                                              

•  from participants?  No 

•  from clinicians?  No 

•  from data analyzers?  No 

4.  Were the participants (Ps) adequately described? Yes 

–  How many Ps were involved in the study?  4

–  CONTROLLED CHARCTERISTICS:    

•  expressive language: sample of at least 50 utterances

•  receptive language: within normal limits (WNL)

•  diagnosis: CAS

•  hearing level: WNL

•  comorbid developmental or genetic problems: excluded

•  oral motor skills: WNL for structure, strength, muscle tone, and reflexes

–  DESCRIBED CHARACTERISTICS:

•  age:  55 to 8-6

•  gender:  4m                                   

•  expressive language:

     – Clinical Evaluation of Language Fundamentals (4th ed; CELF) Expressive Language Index: Standard Score (SS) = 53-80

•  receptive language: 

     – Peabody Picture Vocabulary Test (PPVT):  SS = 90-117

     – CELF Receptive Language Index:  SS = 96-105

     – Test of Auditory Processing (3rd ed; TAP) Word Discrimination:  SS = 5-9

     – Test of Auditory Processing (3rd ed; TAP) Word Memory:  SS = 6-12

•  literacy:

     – Woodcock Reading Mastery Test-Revised Basic Skills Cluster:  SS = 88-127

     – Woodcock Reading Mastery Test-Revised Word Identification:  SS = 90-129

     – Woodcock Reading Mastery Test-Revised Word Attack:  SS = 81-121

     – Woodcock Reading Mastery Test-Revised Lower Case Letters Checklist percent: = 48-94

     – Comprehensive Test of Phonological Processing (CTPP) Phonological Awareness Composite Score:  SS = 64-106

     – CTPP Phonological Memory Composite Score:  SS = 70-91

     – CTPP Memory for Digits:  SS = 6-10

     – CTPP Non-word Repetition Score:  SS = 4-9

     – CTPP Rapid Naming Composite Score:  SS = 91-136

     – Neale Analysis of Reading Ability (3rd ed; NARA-3)

          ∞ Accuracy: Reading Age  <6 – 7.7 years

          ∞ Comprehension: Reading Age  6.3 to 7.5 years

          ∞ Rate: Reading Age  6.8 – >13

–  Were the communication problems adequately described? Yes

•  Disorder type: Childhood Apraxia of Speech

5.  Was membership in treatment maintained throughout the study?  Yes 

•  If there was more than one participant, did at least 80% of the participants remain in the study?  Yes 

6.  Did the design include appropriate controls?  Yes 

•  Were baseline/preintervention data collected on all behaviors?  Yes

  Did probes/intervention data include untrained stimuli?  Yes 

•  Did probes/intervention data include trained stimuli?  Yes 

•  Was the data collection continuous?  Yes_x__, for several of the outcomes.  

•  Were different treatment counterbalanced or randomized?  NA  

7.  Were the outcome measures appropriate and meaningful? 

•  OUTCOME #1: Prosodic (lexical stress) accuracy of targeted bisyllable pseudo words during treatment sessions

•  OUTCOME #2: Segmental accuracy of targeted bisyllable pseudo words during treatment sessions

•  OUTCOME #3: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation) of trained targets

•  OUTCOME #4: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation) of untrained targets

•  OUTCOME #5: Percent Vowels Correct (PVC) in connected speech

•  OUTCOME #6: Percent Consonants Correct (PCC) in connected speech

•  OUTCOME #7: Percent correct stress patterns in connected speech

•  OUTCOME #8: PPVT Standard Score (this was a foil outcome; that is, improvement was not expected)

–  All of the outcomes that were subjective.

–  None of the outcomes were objective.

–  RELIABILITY DATA:

•  Inter-rater reliability of accuracy (stress and segmental) of clinician’s (C’s) judgment of participant’s (P’s) productions: 88%

•  Intra-rater reliability of accuracy (stress and segmental) of C’s judgment of P’s productions: 93%

•  Inter-rater reliability of phonemic transcriptions: 83% to 90%

•  Intra-rater reliability of phonemic transcriptions: 95% to 97%

  Treatment fidelity: 75% to 83% (errors tended to be related to delaying feedback

8.  Results:

–  Did the target behavior(s) improve when treated?  Yes, for the most part, but none of the Ps in this investigation achieve Mastery in the 12 sessions.

–  DESCRIPTION OF RESULTS

•  OUTCOME #1: Prosodic (lexical stress) accuracy of targeted bisyllable pseudo words during treatment sessions: Strong Improvement

     ∞ P1 = baseline = below 10%; final session = 66%   

     ∞ P2 = baseline = below 30%; final session = 82%   

     ∞ P3 = baseline = 15%; penultimate session = 75%

     ∞ P4 = baseline = around 30%; final 3 sessions = averaged 79%

•  OUTCOME #2: Segmental accuracy of targeted bisyllable pseudo words during treatment sessions:  Moderate to strong improvement

     ∞ P1 = baseline = below 10%; final session = 66%   

     ∞ P2 = baseline = below 10%; final 3 sessions = 39% to 59%   

     ∞ P3 = baseline = about 50%; penultimate session = 75%

     ∞ P4 = baseline = around 50%; final 3 sessions = averaged 79%   

•  OUTCOME #3: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation)  of trained targets: Moderate improvement

     ∞ P1 = baseline =  0%       ; final probes = 10%; retention probe = 50-60%

     ∞ P2 = baseline =  0%; final probe = about 40%; retention probe = about 30%

     ∞ P3 = baseline = 20% or under; final probe = about 50%; retention probe= about 40%

     ∞ P4 = baseline = about 20%; final probe = about 40%; retention probe = about 40%

•  OUTCOME #4: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation) of untrained targets: Limited improvement

     ∞ P1 = baseline =   0%; final probe = 0%; retention probe = under 20%

     ∞ P2 = baseline = under 10%; final probe = about 30%; retention probe = 0%

     ∞ P3 = baseline = 20% or under; final probe = about 30%; retention probe = about 30%

     ∞ P4 = baseline = 0% to 40%; final probe = about 20%; retention probe = about 40%. 

•  OUTCOME #5: Percent Vowels Correct (PVC) in connected speech: Limited improvement

     ∞ P1 = pretreatment = 74%; post treatment = 81%; retention = 80%      

     ∞ P2 = pretreatment = 78%; post treatment = 87%; retention = 87%      

     ∞ P3 = pretreatment = 90%; post treatment = 88%; retention = 92%

     ∞ P4 = pretreatment = 85%; post treatment = 84%; retention = 91%

•  OUTCOME #6: Percent Consonants Correct (PCC) in connected speech:  Limited improvement to ineffective

     ∞ P1 = pretreatment = 81%; post treatment = 80%; retention = 54%

     ∞ P2 = pretreatment = 95%; post treatment = 95%; retention = 88%

     ∞ P3 = pretreatment = 70%; post treatment = 86%; retention = 74%

     ∞ P4 = pretreatment = 60%; post treatment = 70%; retention = 66%

•  OUTCOME #7: Percent correct stress patterns in connected speech: Limited to moderate improvement

     ∞ P1 = pretreatment = 46%; post treatment = 43%; retention = 70%

     ∞ P2 = pretreatment = 53%; post treatment = 81; retention = 76%      

     ∞ P3 = pretreatment = 79%; post treatment = 77%; retention = 85%     

     ∞ P4 = pretreatment = 64%; post treatment = 68%; retention = 83%     

•  OUTCOME #8: PPVT Standard Score (this was a foil outcome; that is, improvement was not expected) Ineffective

     ∞ P1 = pretreatment = 90; post treatment = 84   

     ∞ P2 = pretreatment = 90; post treatment = 96   

     ∞ P3 = pretreatment = 91; post treatment = 90   

     ∞ P4 = pretreatment = 117; final session = 119  

9.  Description of baseline: 

–  Were baseline data provided?  Variable

  Was baseline low (or high, as appropriate) and stable?

•  OUTCOME #3: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation)  of trained targets: Baseline low and stable for all P

•  OUTCOME #4: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation) of untrained targets: Baseline was low and stable for Ps 1, 2, and 3 but not 4.

  Was the percentage of nonoverlapping data (PND) provided? 

Yes  _____  No  ________  Only for the following Outcome-

•  OUTCOME #3: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation) of trained targets: 

     – PND for P1 = 75% (fairly effective)

     – PND for P2 = 100% (highly effective)

     – PND for P3 = 75% (fairly effective)

     – PND for P4 = 100% (highly effective)

•  OUTCOME #4: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation) of untrained targets: 

     – PND for P1 = 25% (unreliable/ineffective)

     – PND for P2 = 25% (unreliable/ineffective)

     – PND for P3 = 75% (fairly effective)

     – PND for P4 = 0% (unreliable/ineffective)

10.  What is the clinical significance?  NA

11.  Was information about treatment fidelity adequate?  Yes. The investigators calculated treatment fidelity for each of the Ps. It ranged from 75% to 83% with errors tending to be related to delaying feedback.

12.  Were maintenance data reported?  Yes. Four weeks after the post treatment assessments, retention was measured for selected outcomes. The amount of retention varied based on the outcome and individual Ps.

     – OUTCOME #3: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation)  of trained targets

     ∞ Strong retention for P1and P4.

     ∞ Moderate retention for P2 and P3 

     – OUTCOME #4: Percent perceptual accuracy (target perceived as having both correct prosody and correct articulation) of untrained targets:

     ∞ Limited retention P4 

     ∞ Strong retention P4 (perhaps P1

     ∞ Failure to retain P2  

•  OUTCOME #5: Percent Vowels Correct (PVC) in connected speech

     ∞ all Ps strong 

•  OUTCOME #6: Percent Consonants Correct (PCC) in connected speech

     ∞ Failure –retention was markedly lower than pretreatment  for P1 

     ∞ Strong retention (P2)

     ∞ Limited retention (P3, P4)  

•  OUTCOME #7: Percent correct stress patterns in connected speech:

     ∞ Strong retention (all Ps)

13.  Were generalization data reported? Yes

•  There were 3 types of generalization data: 

     – performance on trained verse untrained probes at selected intervals (baseline and probes), 

     – performance in connected speech, 

     – performance on the PPVT before and after treatment. 

•  Performance on trained verse untrained probes at selected intervals (baseline and intervention probes): The percent overall accuracy of trained and untrained targets during baseline was similar for all Ps. However, the percent overall accuracy of trained targets exceeded baseline for all Ps for intervention probes. 

•  All Ps increased their percent of correct stress in connected speech during retention testing, although connected speech was not targeted during treatment.

•  None of the Ps increased their PPVT scores from the beginning to the end of treatment. This was not expected because this was considered to be a foil measure.

14.  Brief description of the design:

• The authors provided supplementary data online in addition to the information in the Results section. Only the data presented in the article are described in this review.

• The investigators administered ReST to 4 children who had been diagnosed with CAS. performance in connected speech.

• The investigators measured selected outcomes at a series of 3 baseline and 3 probe sessions as well at a retention session, 4 weeks following treatment. Other outcomes were measured before and after treatment.

• The analysis of the data included visual analysis as well as the calculation of PND.

ASSIGNED OVERALL GRADE OF THE QUALITY OF SUPPORT FOR THE INTERVENTION:  B+

SUMMARY OF INTERVENTION

PURPOSE: To explore the effectiveness of ReST in improving the production of lexical stress

POPULATION:  Childhood Apraxia of Speech (CAS); children

MODALITY TARGETED:  production

ELEMENTS/FUNCTIONS OF PROSODY TARGETED:  lexical stress

ELEMENTS OF PROSODY USED AS INTERVENTION:  stress, duration, rate

OTHER ASPECTS OF LANGUAGE/COMMUNICATION TARGETED:  articulation 

OTHER TARGETS:  a foil target of receptive vocabulary 

DOSAGE:  60 minute individual sessions, 4 sessions per week for 3 weeks (12 sessions); homework was not assigned.

ADMINISTRATOR:  SLP

STIMULI: pseudowords 

MAJOR COMPONENTS:

• The focus of ReST is the production of bisyllable nonsense words with stress patterns (Strong-Weak, SW, or Weak-Strong, WS). Because previous research has indicated that even children as young as 5 (the youngest P in this investigation) have a tendency to produce certain pseudo words as SW or WS, the tendencies were adhered to in the construction of pseudowords for the targets and probe stimuli. 

• Treatment session consisted of pre-practice (10-20 minutes) and practice (40-50 minutes) phases.

PRE-PRACTICE PHASE 

• In place of the terms Weak and Strong, the clinician (C) used Short and Long. 

• C presented a bisyllable pseudoword target to the participant (P) randomly and asked P to identify the pseudoword as Long-Short or Short-Long.

• C corrected P if the wrong stress pattern was identified.

• C then modeled the target word and requested P to produce it. 

• C provided 100% feedback regarding the accuracy of the stress pattern produced by P. If P’s production of stressing was in error, C used knowledge of performance (KP) feedback and described how P should modify the production (e.g., “Try to make the first part even shorter”, p 140).

• C was also allowed to provide other cues such as hand clapping or shaping to facilitate correct production.

• Regarding the segmental accuracy of the target (not the primary focus of this investigation), C provided knowledge of results (KR) feedback. That is C only indicated whether the speech sound production was correct or incorrect.

• When P produced 5 correct consecutive trials, C proceeded to the Practice Phase

PRACTICE PHASE

• C presented the treatment targets (19 bisyllable pseudo words) in random order within sets to allow for the production of at least 100 pseudowords per session. 

• Depending on the P’s reading level, P either read aloud a pseudoword or imitated C’s production of the C reading aloud the pseudoword.

• Three to 5 seconds after P’s attempt, C provided KR feedback on the combined prosodic and segmental accuracy at the 50% level.

• Mastery was defined as 80% correct performance over 3 consecutive sessions. (Typically developing children can achieve this in 3 to 4 sessions.) None of the Ps in this investigation achieve Mastery in the 12 sessions.

=============================


de Swart et al. (2003)

February 2, 2021

EBP THERAPY ANALYSIS

Treatment Groups

Note: Scroll about 80% of the way down the page to read the summary of the intervention procedures.

Key:

 C =  Clinician

 EBP =  evidence-based practice

 LSVT = Lee Silverman Voice Treatment 

 NA = not applicable 

 P =  Patient or Participant

 PLVT = Pitch Limiting Voice Treatment

 PD =  Parkinson’s disease

 pmh =  Patricia  Hargrove, blog developer

 SLP =  speech–language pathologist

SOURCE: de Swart, B. J. M., Willemse, S. C., Massen, B. A. M., & Horstink, M. W. I. M. (2003). Improvement of voicing in patients with Parkinson’s disease by speech therapy. Neurology, 60, 498-500.

REVIEWER(S): pmh

DATE: January 29, 2021

ASSIGNED GRADE FOR OVERALL QUALITY:  No grade assigned. This was not an intervention investigation, rather it could be considered ‘proof-of-concept’ research in which information from a single 30-minute procedure has application for the development of an intervention procedure.  

TAKE AWAY: This clinically related (not clinical intervention) research compares Pitch Limiting Voice Treatment (PLVT), Lee Silverman Voice Treatment (LSVT), and habitual speaking style to determine if PLVT can increase loudness like LSVT and also modulate pitch to avoid increased pitch level. This single session task served as a proof of concept for PLVT because the investigation indicated that both PLVT and LSVT resulted in increased loudness but only PLVT limited pitch increases.

1.  What type of evidence was identified? 

•  What was the type of evidence? Single Group Experimental Design with 3 tasks and 3 conditions (i.e., a repeated measure design) 

•  What was the level of support associated with the type of evidence? Level =  Not Applicable (NA); this was not an intervention investigation

2.  Group membership determination: 

•  If there was more than one group, were participants (Ps) randomly assigned to groups? Not Applicable (NA)

3.  Was administration of intervention status concealed?

  from participants? No

•  from clinicians? No

•  from analyzers? Unclear  

4.  Was the group adequately described?  Yes 

– How many  Ps were involved in the study? 

•  total # of Ps: 32

•  # of groups: 1

– CONTROLLED CHARACTERISTICS

•  cognitive skills: Ps with dementia were excluded

•  expressive language: Ps who were diagnosed as having “severe, hardly intelligible dysarthria” (p.498) were excluded

•  diagnosis: Parkinson’s disease (PD)

•  social/emotional status: Ps with depression were excluded

•  neurological status: Ps with other comorbid neurological conditions were excluded 

•  hearing status: Ps with hearing loss were excluded

•  previous speech therapy: Ps with a history of speech therapy within a year of the investigation were excluded

•  Other: the Ps were from an outpatient service and they were consecutively identified from the P enrollments; Ps with “on-off” phenomena (p. 498) were excluded

– DESCRIBED CHARACTERISTICS

•  age: 36 years to 75 years

•  gender: 17m; 15f

•  vocal status: diagnosed with mild to severe voice disorders 

•  medications: All Ps were taking PD medications at the time of the investigation 

•  time since diagnosis: 1 to 18 years

–  Were the groups similar before intervention began?  NA 

  Were the communication problems adequately described? Yes 

•  disorder type:  (List) dysarthria associated with PD

•  functional level: mild to severe

5.  Was membership in groups maintained throughout the study?

•  Did the group maintain at least 80% of their original members? Yes

•  Were data from outliers removed from the study?  No 

6.   Were the groups controlled acceptably?  NA

7.  Were the outcomes measure appropriate and meaningful? Yes

•  OUTCOME #1: Acoustic measure of loudness

•  OUTCOME #2: Acoustic measure of pitch

•  OUTCOME #3: Acoustic measure of jitter 

•  OUTCOME #4: Acoustic measure of duratioN

–  NONE of the outcomes were subjective.

–  ALL of the outcome measures were objective.

8.  Were reliability measures provided?  No

  Interobserver for analyzers?  No  

•  Intraobserver for analyzers?  No  

  Treatment fidelity for clinicians?  

9.  What were the results of the statistical (inferential) testing and/or  the description of the results?

—  What level of significance was required to claim significance?  p ≤ 0.01 

•  OUTCOME #1: Acoustic measure of loudness

     – Both PLVT and LSVT increased significantly from habitual level in all tasks

•  OUTCOME #2: Acoustic measure of pitch

     – LSVT increased significantly from habitual level in all tasks

     – PLVT increased significantly from habitual level only in reciting

•  OUTCOME #3: Acoustic measure of jitter

     – Both PLVT and LSVT decreased significantly from habitual level in the vowel task

     – The difference between PLVT and LSVT is not significant

•  OUTCOME #4: Acoustic measure of duration

      – Neither PLVT nor LSVT increased significantly from habitual level

—  What the statistical tests were used to determine significance?  Place xxx after any statistical test that was used to determine significance.  

•  ANOVA: (multivariate, repeated measures) 

•  Other:  There was also a Bonferoni Correction.

  Were confidence interval (CI) provided?  No __x___

10.  What is the clinical significance?  NA

11.  Were maintenance data reported?  No 

12.  Were generalization data reported?  Yes  

13.  Describe briefly the experimental design of the investigation.

• Thirty-two adults with PD were enrolled in this investigation by researchers from the Netherlands. The site was an outpatient clinic and the patients (Ps) were identified as consecutive Ps from the clinic’s patient rolls. The single experimental session for each P lasted approximately 30 minutes.

• The experiment involved 3 tasks and 3 conditions with each task/condition pairing being produced 2 times.

     – Tasks: 

          ∞ sustaining “ah” as long as possible, 

          ∞ reciting the months of the year, 

          ∞ reading a short passage in which the same 2 sentences form the middle of the passage were selected for analysis.

     – Conditions: 

          ∞ spontaneous speaking style (“the way you speak at home,” p. 498)           

          ∞ LSVT style (“think loud, think shout,” p. 499)

          ∞ PLVT style (“speak loud and low,” p. 499)

• Initially the experimenter gave the P a verbal direction but if the P was unsuccessful, the experimenter demonstrated the targeted behavior.

• The performance of the Ps was audiotaped for later acoustic analysis.

• The results were analyzed using multivariate analysis-of-variance with repeated measures.

ASSIGNED OVERALL GRADE FOR QUALITY OF EXTERNAL EVIDENCE:   NA

SUMMARY OF INTERVENTION

PURPOSE: To explore if instructing Ps with PD to “speak loud and low” limits the pitch rise associated with increased loudness.

POPULATION: Parkinson’s disease adults

MODALITY TARGETED: expression

ELEMENTS/FUNCTIONS OF PROSODY TARGETED: pitch, loudness, duration

OTHER ASPECTS OF LANGUAGE/COMMUNICATION TARGETED:  jitter

DOSAGE: one, 30-minute session

MAJOR COMPONENTS: 2 interventions are briefly described: PLVT and LVST.

PLVT: 

• The clinician (C) directs the P to produce the target speaking “loud and low” (p. 499).

LSVT

• The clinician (C) directs the P to “think loud, think shout” (p. 499) when attempting to produce targets.

_______________________________________________________________