Tjaden (2000)

EBP THERAPY ANALYSIS for

Single Subject Designs

 

SOURCE:  Tjaden, K. (2000). Exploration of a treatment technique for prosodic disturbance following a stroke. Clinical Linguistics and Phonetics, 14,  619-614.

 

REVIEWER(S):  pmh

 

DATE:  4.18.13

ASSIGNED OVERALL GRADE:  D-

 

TAKE AWAY:  The results of this case study suggest this computer-based intervention was not effective in reducing scanning speech by increasing the temporal variability (duration) of adjacent syllables. Nevertheless, the investigator presented thoughtful procedures for perceptually and acoustically measuring  scanning speech and for clinical decision making.

 

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

                                                                                                           

2.  What type of evidence was identified?                              

a.  What  type of single subject design was used?  Case Study: Description with Pre and Post Test Results

b.  What was the level of support associated with the type of evidence?  Level =  D+     

                                                                                                           

3.  Was phase of treatment concealed?

a.  from participants? No

b.  from clinicians? No

c.  from data analyzers? No

 

4.  Were the participants adequately described?  

a.  How many participants were involved in the study?  List here:  2 (1 with aphasia, 1 neurologically intact peer for comparison)

b.  The following characteristics were described:

•  List characteristics controlled:

•  age:  42 years

•  gender:  m                                    

•  educational level of participant: master’s degree prior to stoke, enrolled in another master’s program at the time of investigation

•  etiology:  left hemisphere cerebral infarct, 2 seizures following stroke

•  other medical issues:  migraines, hypertension

•  medications:  several

•  motor involvement:  right hemiparesis

•  previous therapy:  6 weeks at residential rehab facility following stroke; outpatient –about 7 weeks; weekly group therapy at university clinic from a year after stroke to time of study (2 hours, language based)

•  previous intervention plan:  word retrieval, reading, prosody

c.  Were the communication problems adequately described?  Yes

•  List the disorder type(s):  expressive aphasia

•  List other aspects of communication that were described:

prosodic problems remained—strong scanning speech; perceived monotonous rhythm, reduced naturalness

formal assessment prior to investigation:

    •  oral mechanism—right side of face—mild weakness

    •  intelligibility—estimate = 95%

    •  Apraxia Battery for Adults – mild to moderate apraxia of speech, no oral or limb apraxia (increased difficulty with words as they increased in length, groping, awareness and self-correcting of errors, prosodic problems)

characteristics of pre-treatment/baseline sample:

     •  attempted to document

         1.  P’s production  of selected acoustic measures believed to be correlates of scanning speech

         2.  relationship between selected acoustic measures and perceptions of monotonous rhythm and  speech naturalness

–  the acoustic measures were

     •  lengthened syllable durations (slow rate)

     •  reduced temporal variation of adjacent syllables (syllable isochrony)

     •  temporal segregation of adjacent syllables

     •  reduced or flattened fo across adjacent syllables

–  elicited read speech samples in 10 pretreatment sessions for P and peer

    •  the investigator determined treatment targets by comparing P to peer using the Outcomes in #7.  The following decisions were made:

     •  P produced only 6 intraturn pauses within 200 words.   Therefore, intraturn pauses were not targets

     •  P’s fo in target words was actually significantly larger than the comparison peer’s.  Therefore the investigator measured fo in whole sentences; there was not a significant difference.  The investigator did not target fo in treatment.

     •  Total word durations were also significantly longer than peer BUT were not targeted for therapy because conversation with C revealed P’s articulation deteriorated in more rapid speech. 

–  Investigator also explored the relationship between syllable isochrony and judges’ perceptions of speech rhythm (typical flow of speech in neurologically intact speakers) and naturalness (the extent to which the speech sounded ‘normal’).

     •  words with lower isochrony scores (0.4-1.2) were perceived to be less natural and less rhythmic than higher isochrony scores (2.0-3.7)

                                                                                                                       

5.  Was membership in treatment maintained throughout the study?  Not applicable; this was a single case study.

a.  If there was more than one participant, did at least 80% of the participants remain in the study?  Not  applicable

b.  Were any data removed from the study? No.

 

6.  Did the design include appropriate controls?  No, there were limited controls as this was a case study BUT the investigator provided comparison data from a neurologically intact  peer.

a.  Were baseline/preintervention data collected on all behaviors?  Yes

b.  Did probes/intervention data include untrained data?  Yes

c.  Did probes/intervention data include trained data?  No

d.  Was the data collection continuous?  Yes

e.  Were different treatment counterbalanced or randomized? Not Applicable 

 

7.  Were the outcomes measure appropriate and meaningful?  Yes

a.  There were 2 types of outcomes of interest in this investigation:

1.  Initial Pretreatment/baseline measures—the investigator explored these measures to determine which Outcome(s) to target in the intervention.

  OUTCOME A:  Total word duration of a 3 syllable word (with primary stress on 1st or 2nd syllable) imbedded in a read sentence.  (A global index of rate)

OUTCOME B:  The result of dividing the duration of the stressed syllable by the duration of the unstressed syllable on a 3 syllable word (with stress on the 1st or 2nd syllable) imbedded in a read sentence.  (A measure of syllable isochrony which may be a measure of scanning speech; 1.0 = equal durations of each syllable; target ≥ 2.0.)

OUTCOME C:  Frequency of intraword pauses (silences within a word not part of closure for consonants) on a 3 syllable word (with stress on the 1st or 2nd syllable) imbedded in a read sentence)

OUTCOME D:  Duration of intraword pauses on a 3 syllable word (with stress on the 1st or 2nd syllable) imbedded in a read sentence

OUTCOME E:  fo

2. Target Outcome—the investigator focused on this outcome.  See item #4c for rationale.

OUTCOME #1:  The result of dividing the duration of the stressed syllable by the duration of the  unstressed syllable on a 3 syllable word (with stress on the 1st or 2nd syllable) imbedded in a read sentence.  (A measure of syllable isochrony which may be a measure of scanning speech; 1.0 = equal durations of each syllable)

b.  Where any of the outcomes subjective:  no

c.  Which numbers of the outcomes were objective:  all

d.  The outcome measure for which reliability data were provided is Outcome #1, however, the reliability is for calculation of the duration of syllable, not for the calculation of the syllable isochrony measure.

e.  The data supporting reliability of Outcome #1 is 0.98 for duration of syllables, not the calculation of the  syllable isochrony measure.

 

8.  Results:

a.  Did the target behavior improve when it was treated?  No

b.  The overall quality of improvement for  OUTCOME #1 (syllable isochrony) is   ineffective

9.  Description of baseline:

9.  Were baseline data provided?  Yes. The number of data points for OUTCOME #1 (syllable isochrony) is 10

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

c.  What was the percentage of nonoverlapping data (PND)? 30%

d.  Does inspection of data suggest that the treatment was effective? OUTCOME #1 (syllable isochrony):  ineffective

 

10.  What was the magnitude of the treatment effect?  NA

 

11.  Was information about treatment fidelity adequate?  No

 

12.  Were maintenance data reported?  Yes.  Data for 5 post treatment sessions were reported and displayed no systematic improvement.

 

13.  Were generalization data reported? Yes. The naturalness and rhythmicity measures were not compared pre and post test. Review of the figures suggest they did not improve. Also, the probes were sentences and intervention involved reiterated  nonsense syllable and words; therefore they can be considered generalization.  Review of figures indicate not generalization to sentences although the P did meet criteria for moving through the program.

 

OVERALL RATING OF THE QUALITY OF SUPPORT FOR THE INTERVENTION:  D-

 

NOTE: 

1.  A peer who did not have history of neurological impairment was administered probes on the same schedule as the P to provide comparison data.  He did not receive therapy.

2.  The investigator provided a thoughtful discussion of the rationale for selecting treatment targets.

 

SUMMARY OF INTERVENTION

 

PURPOSE: To explore the effectiveness of a computer based intervention using stress patterning to increase the temporal variation of adjacent syllable in the speech of a patient with left hemisphere damage.

POPULATION:  expressive aphasia—left hemisphere stroke

 

MODALITY TARGETED:  production

 

ELEMENTS/FUNCTIONS OF PROSODY TARGETED:  stress, duration

 

OTHER ASPECTS OF LANGUAGE/COMMUNICATION TARGETED (Dependent variable):  naturalness, rhythmicality (These measures were collected pre and post intervention but the investigator did not compare them but rather used them as correlates.)

DOSAGE:  24, 30-minute sessions; schedule was variable — some weeks one session, other weeks four sessions

 

ADMINISTRATOR:  SLP

 

STIMULI: visual, auditory, motor

,

GOAL ATTACK STRATEGY: vertical

 

MAJOR COMPONENTS:

 

•  This is a hierarchical intervention that moved from nonsense syllable to speech.

•  Auditory, visual, and computational feedback were provided.

•  Used the Computerized Speech Lab 4300 (CSL)

•  Steps:

LEVEL 1 (criterion: for 2 consecutive sessions, 8/10 probe sentences with syllable isochrony measure ≥ 2.0)

Step 1

–  target:

a.  each trial within a session involved production of 10 reiterant sequences of dadada

b.  5 had stress on syllable 1 (DAdada) and 5 on syllable 2 (daDAda)

–  Prior to therapy:  C provided P with directions regarding how to interpret visual and computational feedback.

–  At the beginning of each session, C recorded and played for P 10 reiterant sequences (5 with stress on syllable 1, 5 with stress on syllable 2.)

–  C presented targets to P on index cards that had been shuffled to randomize.

–  P produced an attempt that was recorded and analyzed by CSL.

–  3 types of feedback to P:

a.  Visual waveform on CSL

b.  C calculated the syllable isochrony measure  (target 2.0 or better)

c.  C replayed P’s attempt on a loudspeaker.

–  after producing the 10 reiterant sequences, P produced the probe sentences.

Step 2

–  Identical to Step one although C played the recorded models of reiterated syllable every 3rd trial.

LEVEL 2 (criterion: for 2 consecutive sessions, 8/10 probe sentences with syllable isochrony measure ≥ 2.0)

Step 1

–  target:  production of reiterant nonsense syllables and real words with identical  stress patterns.

a.  each trial within a session involved production of 10 reiterant sequences of dadada paired with a 3 syllable word with identical stress patterns

b.  5 pairings had stress on syllable 1 (DAdada and honesty) and 5 on syllable 2 (daDAda and rehearsal)

–  every 3 trials P listened to C’s prerecorded production of one set of targets (DAdada, bicycle; daDAda and banana)

–  P produced an attempt that was recorded and analyzed by CSL.

–  3 types of feedback to P:

a.  Visual waveform on CSL of both words

b.  C calculated the syllable isochrony measure for the real word only  (target 2.0 or better)

c.  C replayed P’s attempt on a loudspeaker.

–  P did not reach criterion over 7 sessions; therefore, Step 1a was added.

Step 1a

–  the same procedures as Step 1 and

–  C taught P to tap out the rhythm of the targets with his unaffected hand.

–  P reached criterion in 5 sessions and then returned to Step 1.

Step 2

–  target:  production of 3 syllable real words with stress on the first or second syllable

–  The procedure was similar to previous steps except

–  at the beginning of each session, C played 10  three syllable words (5 with stress on syllable 1; 5 with stress on syllable 2)

–  C changed criterion for acceptable syllable isochrony  for moving to the next step to 2.0 to 3.5 because P had displayed a tendency to produce exaggerated syllable isochrony thus disrupting speech naturalness.

–  After 5 sessions of Step 2 (and a total of 24 treatment sessions), the intervention was terminated due to lack of progress.

NOTE: 

 

1.  The investigator clearly described treatment procedures in the appendix.

2.  Perceptions of rhythmicity and natural were measured before and after intervention by a panel of judges.  The pre and post measurements of rhythmicity and naturalness were not compared. Rather, the investigator explored their relationship with the P’s isochrony da

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: