Bornhofen & McDonald (2008)

Comparing Treatment of Groups

 

SOURCE:  Bornhofen, C., & McDonald, S. (2008). Comparing strategies for treating emotional perception deficits in traumatic brain injury. Journal of Head Trauma Rehabilitation, 103-115.

 

REVIEWER(S):  pmh

 

DATE: 5.07.13

ASSIGNED GRADE FOR OVERALL QUALITY:  C+

 

TAKE AWAY:  This investigation focused on overall emotional perception; a small number of outcomes are concerned affective prosody as it is only one component of emotional perception. Although this may be interpreted as a disadvantage, the interventions more closely replicate daily living compared to interventions that  focus solely on one aspect of emotional perception (e.g., prosody).  Overall, the findings revealed some success with respect to emotional perception and some generalization outcomes.

 

1.  What type of evidence was identified?

a.  What was the type of evidence? Prospective, Randomized Group Design with Controls

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

                                                                                                           

2.  Group membership determination:

a.  If there were groups, were participants randomly assigned to groups?  Yes

3.  Was administration of intervention status concealed? 

a.  from participants?  No

b.  from clinicians?  No

c.  from analyzers?  Variable. The post  intervention analyzers were blind to group membership.

 

4.  Were the groups adequately described?  Yes

a.         How many participants were involved in the study?

•  total # of participant:  18

•  # of groups: 3

•  # of participants in each group:  6, 6, 6

•  List names of groups:     EL (Errorless Learning), SIT (Self-instruction training), WL (Waitlist)

b.  The following variables were actively controlled:

INCLUSION CRITERIA:

•  severe TBI

•  post traumic amnesia at least 1 day

•  reports of social/interactional problems

•  at least 6 months post trauma

•  at least 2 SD below mean on pretests of social perception

EXCLUSION CRITERIA:

•  history of psychosis or severe depression

•  below borderline cognitive functioning

–  The following characteristics were descriptors:

(no significant differences on measures with asterisk*; remember the N is small!)

•  mean age*:  43.75; 35.4; 31.2 

•  gender:  17m, 1f

•  cognitive skills:  

     –  85.3; 93; 86.6 (mean Weschler Test of Adult Reading scores as a measure of premorbid cognition)

     –  6; 5.4; 5.8 (mean SS Logical Memory I)

     –  5.75; 5.4; 6.6 (mean SS Logical Memory II)

     –  5; 7.4; 6,6 (mean SS Similarities)

     –  6.5; 7; 7.4 (mean SS Matrix Reasoning)

     –  6.25; 6.8; 4.2 (mean SS Letter-Number Sequencing)

     –  6; 5.4; 5.8 (mean SS Symbol Search)

•  face recognition:   (mean adjusted score Benton Face Recognition Test)

     –  6; 5.4; 5.8

•  educational level of clients:  8.75; 11.4; 10.8 (mean years)

•  months post onset:  60; 79.6; 148.2  (mean)

•  days of posttraumatic amnesia:  73.25; 32.5; 80.8 (mean)

c.   Were the groups similar before intervention began?  Unclear  _x__.  Although investigators found no significant differences among groups for pretest characteristics, there was considerable range in the preintervention scores for the 3 groups.   (Remember, there were only 6 members in each group.)

d.  Were the communication problems adequately described?  No

•  disorder type:  social interaction problems associated with TBI

 

5.  Was membership in groups maintained throughout the study?

a.  Did each of the groups maintain at least 80% of their original members?  No. EL maintained 67%  ; SIT maintained   83%; WL maintained 67%

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

 

6.   Were the groups controlled acceptably?

a.  Was there a no intervention group?  Yes

b.  Was there a foil intervention group? No

c.  Was there a comparison group?  Yes

d.  Was the time involved in the foil/comparison and the target groups constant?  Yes

 

7.  Were the outcomes measure appropriate and meaningful?  Yes

a.  Outcomes (dependent variables):

INTERVENTION OUTCOMES      

  OUTCOME #1: Performance on the Facial Expression Same/Different Task

  OUTCOME #2: Performance on the Facial Expression Naming Task (label chosen from a list of 7 emotions)

  OUTCOME #3: Performance of the Facial Expression Matching Task

  OUTCOME #4: Performance on task in which actor enacted an emotion and P selected label from a list of 7 words (cues were only nonverbal)

  OUTCOME #5: Increased ability to interpret social inferences from emotional expression (TASIT, pt 2)

  OUTCOME #6:  Increased ability to differentiate sarcasm and lies (TASIT, pt 3)

GENERALIZATION OUTCOMES

  OUTCOME #7:  Self-report of overall psycho-social functioning (Sydney Psychosocial Reintegration Scale)

  OUTCOME #8:  Self-report of depression (Depression Anxiety Stress Scales)

  OUTCOME #9:  Relative of overall psycho-social functioning (Sydney Psychosocial Reintegration Scale)

  OUTCOME #10: Relative’s ranking of negative and positive behaviors seen in clinical populations   (Katz Adjustment Scale- Relative form)

  OUTCOME #11:  Relative reporting of positive and negative social behaviors (Social Performance Survey Schedule)

b.  All of the outcome measures were subjective.

c.  None of the outcome measures were objective.

 

8.  Were reliability measures provided?

a.  Interobserver for analyzers?  No

b.  Intraobserver for analyzers?  No

c.  Treatment fidelity for clinicians?  Yes.  Two trained raters viewed 8 randomly ordered intervention sessions and assessed whether selected C behaviors occurred. The investigators reported that the raters achieved good interrater reliability.  The raters’ scoring indicated that  EL intervention behaviors were markedly more likely to occur during EL sessions and SIT behaviors were more common in SIT sessions.  Nonspecific intervention techniques (e.g., C displayed warmth and caring) were observed at equal rates in EL and in SIT sessions.    

 

9.  What were the results of the statistical (inferential) testing?

a.   PRE VS POST TREATMENT: INTERVENTION OUTCOMES WITH REPORTED SIGNIFICANT DIFFERENCES:

  OUTCOME #3: Performance of the Facial Expression Matching Task— Form A of Test; EL and SIT; pre vs post (p ≤ 0.05); pre vs 1 month follow up (both < 0.01);

  OUTCOME #5: Increased ability to interpret social inferences from emotional expression (TASIT, pt 2)—SIT Form A; pre vs post (p ≤ 0.05)

  OUTCOME #11:  Relative reporting of positive and negative social behaviors (Social Performance Survey Schedule) SIT—positive behaviors; pre vs post (p ≤ 0.05)

b.  What was the statistical test used to determine significance:  ANOVA and the investigator corrected for the small N by using Ley’s  procedure.  See page 107 and  appendix.

c.  Were confidence interval (CI) provided?  No

                                               

10.  What is the clinical significance? 

a.  Results of EBP testing

•  ETA:   using Cohen’s guidelines the clinical differences (≥ 0.20) were reported

INTERVENTION OUTCOMES      

  OUTCOME #1: Performance on the Facial Expression Same/Different Task—EL (0.42) and SIT ( 0.33) for Form A but not Form B;  pre vs post   

  OUTCOME #2: Performance on the Facial Expression Naming Task (label chosen from a list of 7 emotions  Pre vs Post: SIT form A (0.25)

  OUTCOME #3: Performance of the Facial Expression Matching Task ) Pre vs Post:  EL for form A (0.76), for B (0.21); SIT for form A (0.55) for form B (0.25)

  OUTCOME #4: Performance on task in which actor enacted an emotion and P selected label from a list of 7 words (cues were only nonverbal)-TASIT, pt 1) Pre vs Post:  SIT for form B (0.24)

  OUTCOME #5: Increased ability to interpret social inferences from emotional expression (TASIT, pt 2)   Pre vs Post:  SIT for form A  (0.47)

GENERALIZATION OUTCOMES

  OUTCOME #8:  Self-report of depression (Depression Anxiety Stress Scales) Pre vs Post:  EL (0.30)

  OUTCOME #10: Relative’s ranking of negative and positive behaviors seen in clinical populations   (Katz Adjustment Scale- Relative form) Pre vs Post:  Positive Scale –EL (0.43)

b.  Interpretation of EBP testing.  

 Large: none

•  Moderate:  Outcome #3

•  Small:  Outcomes #1. 2, 3, 4, 5, 8, 11

•  No:  Outcomes # 1, 2, 4, 6, 7, 9, 10

  OUTCOME #1: Performance on the Facial Expression Same/Different Task—EL (small) and SIT ( small ) for Form A but not Form B –no

  OUTCOME #2: Performance on the Facial Expression Naming Task (label chosen from a list of 7 emotions  Pre vs Post: SIT form A (small); form B (no)

  OUTCOME #3: Performance of the Facial Expression Matching Task — Pre vs Post:  EL for form A (moderate), for B (small); SIT for form A (moderate) for form B (0.25)

  OUTCOME #4: Performance on task in which actor enacted an emotion and P selected label from a list of 7 words (cues were only nonverbal)-TASIT, pt 1) Pre vs Post:  SIT for form B (small); form A (no)

  OUTCOME #5: Increased ability to interpret social inferences from emotional expression (TASIT, pt 2)   Pre vs Post:  SIT for form A  (small)

  OUTCOME #6:  Increased ability to differentiate sarcasm and lies (TASIT, pt 3) No

GENERALIZATION OUTCOMES

  OUTCOME #7:  Self-report of overall psycho-social functioning (Sydney Psychosocial Reintegration Scale)  No

  OUTCOME #8:  Self-report of depression (Depression Anxiety Stress Scales) Pre vs Post:  EL (small)

  OUTCOME #9:  Relative of overall psycho-social functioning (Sydney Psychosocial Reintegration Scale)  No

  OUTCOME #10: Relative’s ranking of negative and positive behaviors seen in clinical populations   (Katz Adjustment Scale- Relative form) No

  OUTCOME #11:  Relative reporting of positive and negative social behaviors (Social Performance Survey Schedule)  Pre vs Post:  Positive Scale –EL (small)

 

11.  Were maintenance data reported?  Yes.  Outcomes were measured  1 month and 6 months after post testing. Only the Matching Facial Expressions task form A  (Outcome #3) yielded a significant difference and a clinically significant (large and moderate) difference at 1 month post intervention. This degree of improvement was not maintained at 6 months. In addition, the differences were no longer significantly different.

 

12.  Were generalization data reported? Yes. Outcomes #7 through #11 are generalization outcomes. Outcomes #8 and #11 showed small clinically significant improvement.

 

NOTE:  the N was very small.

 

ASSIGNED GRADE FOR QUALITY OF EXTERNAL EVIDENCE:  C+

 

SUMMARY OF INTERVENTION

 

PURPOSE:  To compare two interventions for improving emotional perception.

POPULATION:  adults with severe traumatic brain injury

 

MODALITY TARGETED:  receptive

 

ELEMENTS/FUNCTIONS OF PROSODY TARGETED:  affective prosody

 

OTHER TARGETS:  overall emotional perception

DOSAGE:  2 ½ hour sessions, 1 time a week, 10 weeks; small group (2 or 3 Ps)

 

ADMINISTRATOR:  therapist (probably not an SLP)

 

STIMULI:  Not clear

GOAL ATTACK STRATEGY:  Not clear

 

MAJOR COMPONENTS:

 

The 2 intervention approaches* were

1.  Effortless Learning (EL)

2.  Self-Instruction Training (SIT)

1.  EL

•  The task involved teaching Ps to  identify an affective state using prosody, facial expression, and body language as cues.

• C

  – discouraged P from guessing during sessions (e.g., C always provided a “not sure” option for P during treatment),

  – designed sessions to progress gradually from easy to more difficult discriminations, and

  – provided extensive practice of each phase of treatment to provide high rate of correct responses.

•  C immediately corrected errors.

•  C initially provided exaggerated cues for the different emotions and gradually faded to more subtle cues.

2.  SIT

•  C presented an acronym (WALTER) to P to facilitate problem solving with respect to emotional perception.

  – the following is quoted from p. 107 of article

  1.  What am I deciding about?

  2.  What do I already know about it?

  3.  What do I need to look/listen for?

  4.  Try out my answer.

  5.  Evaluate how it went.**

  6.  Reward myself for having a go.**

• P practiced using “WALTER” while making increasing difficult discriminations during intervention

*  at the time of publication, the investigators were preparing a manuscript describing more thoroughly the interventions.

** this are optional steps.

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: