Radice-Neumann et al. (2009)

 

EBP THERAPY ANALYSIS

Treatment Groups

 

 

Note: Scroll about two-thirds of the way down the page to read the summary of the procedure(s).

 

Key:

C = Clinician

d = Standardized Mean Difference

EBP = evidence-based practice

f = female

FAR = Facial affect recognition training

m = male

N = number

NA = not applicable

P = Patient or Participant

pmh = Patricia Hargrove, blog developer

SEI = Stories of emotional inference training

SLP = speech–language pathologist

 

 

SOURCE: Radice-Neumann, D., Zupan, B., Tomita, M., & Willer, B. (2009). Training emotional processing in persons with brain injury. Journal of Head Trauma Rehabilitation, 24, 313-323.

 

REVIEWER(S):  pmh

 

DATE: February 11, 2016

 

ASSIGNED GRADE FOR OVERALL QUALITY: A-   (The highest possible grade based on the design of the investigation was A. )

 

TAKE AWAY: Participants (Ps) with acquired brain injury (ABI) improved facial recognition of affect and/or the ability to infer affect and/or their social-emotional behavior following the administration of one of two affect interventions that did not focus of prosodic affect. However, the Ps’ recognition of affective prosody did not improve following either of the two interventions. The findings suggest that affective prosody needs to be directly treated.

 

 

  1. What type of evidence was identified?

                                                                                                           

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

                                                                                                           

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

                                                                                                           

  1. Group membership determination:

                                                                                                           

  • If there was more than one group, were participants (Ps) randomly assigned to groups? Yes

 

  1. Was administration of intervention status concealed?

                                                                                                           

  • from participants? No
  • from clinicians? No
  • from analyzers? No

                                                                    

 

  1. Were the groups adequately described? Yes

 

– How many Ps were involved in the study?

 

  • total # of Ps: 21
  • # of groups: 2
  • List names of groups and the # of participants in each group:
  • Facial affect recognition training (FAR) = 11 (data from one P was removed from the investigation, so the final number, N, was 10)
  • Stories of emotional inference training (SEI) = 10 (one P dropped out of the investigation so the final N was 9.)

 

– The following characteristic were CONTROLLED

  • age: between the ages of 18 and 65
  • severity of brain injury: Severe (Glasgow Coma Scale ≤ 8)
  • time post injury: at least 1 year
  • expressive language: no “substantial” problems (p. 315)
  • receptive language: comprehended short paragraphs and the emotion words used in the investigation; no “substantial” problems (p. 315)
  • facial recognition skills: at least one standard deviation below the mean
  • social-emotional status: with the exception of depression no psychiatric comorbidity
  • hearing status: hearing loss was corrected
  • visual status: impairment was correct
  • Other: no current abuse of alcohol or drug

 

– The following P characteristics were DESCRIBED

  • age: (statistical test revealed no significant difference)

     – FAR = 47 years

     – SEI = 38 years

  • gender:

     – FAR = 9m, 2f

     – SEI = 3m, 6f

  • residence

     – FAR = Western New York (US) – 7; Ontario (Canada) – 4

     – SEI = Western New York (US) – 6; Ontario (Canada) – 3

  • race

     – FAR = white-9; African American – 1; Native American – 1;

     – SEI = white-8; Asian – 1

  • cause of brain injury

     – FAR = motor vehicle – 9; fall – 2

     – SEI = motor vehicle – 6; fall – 1; other – 2

  • years post Injury: (statistical test revealed significant difference)

     – FAR = 16 years

     – SEI = 8 years

  • Glasgow Coma Scale Score;

     – FAR = 3.6

     – SEI = 4.6

  • Brock Adaptive Functioning Questionnaire: (test of executive function, statistical testing revealed no significant difference)

     – FAR = 61

     – SEI = 56

  • receptive language: (statistical testing revealed no significant difference)

     – FAR = 6.9

     – SEI = 6.2

  • Beck Depression Inventory: (statistical testing did not reveal significant difference)

     – FAR = 17

     – SEI = 20

  • DANVA2-Faces: (statistical testing did not reveal a significant difference)

     – FAR = 14

     – SEI = 15.9

 

–   Were the groups similar before intervention began? Yes, for the most part

 

Were the communication problems adequately described?

 

  •  disorder type: Normal functioning on the test that measured the ability to recognized affective prosody was a score of 14-21, with a highest possible score of 24. The pretest scores for the two groups were

– FAR: mean = 13.25; standard deviation = 2.37

– SEI: mean = 14.72; standard deviation = 3.40. (Prior to intervention, some of the Ps of both groups apparently were functioning below normal limits and some were functioning within normal limits. The investigators did not provide individual scores but it is unlikely any of the Ps performed at the test ceiling.)

 

 

  1. Was membership in groups maintained throughout the study?

                                                                                                             

  • Did each of the groups maintain at least 80% of their original members? Yes, however, one SEI P dropped out of the investigation.

                                                               

  • Were data from outliers removed from the study? Yes, the investigators removed the data for one FAR P from the investigation.

 

 

  1. Were the groups controlled acceptably?

                                                                                                             

  • Was there a no intervention group? No
  • Was there a foil intervention group? No
  • Was there a comparison group? Yes
  • Was the time involved in the comparison and the target groups constant? Yes

 

 

  1. Were the outcomes measure appropriate and meaningful? Yes

                                                                                                             

– The outcomes were

NOTE: All of the outcomes are listed, although the focus of this analysis will be on the outcome (OUTCOME #1) concerned with prosody.

 

  • OUTCOME #1: Improved performance on test of the recognition of prosodic affect—DANVA2-AP (NOTE: The investigators had NOT expected transference to this outcome.)
  • OUTCOME #2 –Improved to infer how they (the Ps) would feel in the described situations—LEAS-Self
  • OUTCOME #3: Improved to infer how others would feel in the described situations—LEAS-Others
  • OUTCOME #4: Improved performance on a test of the recognition of facial affect—DANVA2-AF
  • OUTCOME #5: Improved ability to recognize affect from short videos which included facial, prosodic, gestural, and postural cues–EET
  • OUTCOME #6: Improved social-emotional functioning—BAFQ

 

–  List the outcome measures (using outcome numbers from item 7a) that are subjective: ALL

 

 List the outcome measures (using outcome numbers form item 7a) that are objective? NONE

 

                                         

 

  1. Were reliability measures provided?

NOTE: The reliability measures are reports of previously reported data.

                                                                                                            

  • Interobserver for analyzers? Yes

– OUTCOME #2 –Improved to infer how they (the Ps) would feel in the described situations—LEAS-Self: a minimum of 0.84

– OUTCOME #3: Improved to infer how others would feel in the described situations—LEAS-Others: a minimum of 0.84

 

  • Intraobserver for analyzers? No

 

  • Treatment fidelity for clinicians?

 

  • Test-retest reliability? Yes

OUTCOME #1: Improved performance on test of the recognition of prosodic affect—DANVA2-AP: 0.73 to 0.93

– OUTCOME #4: Improved performance on a test of the recognition of facial affect—DANVA2-AF: 0.81

 

 

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

 

  • Summary Of Important Results

 

— What level of significance was required to claim significance? p = 0.05 (0.05 was the level listed in Table 3. The prose reports a lower level was used for most comparisons to correct for multiple comparisons.)

 

 

PRE AND POST TREATMENT

 

NOTES:

  1. Although there were 2 groups of Ps, the investigators did not present the results of statistical comparisons of the 2 intervention groups
  2. * = although the p value was ≤ 0.05, correction for multiple comparisons indicated the difference was not significant

 

  • OUTCOME #1: Improved performance on test of the recognition of prosodic affect—DANVA2-AP

– FAR results:

  • pretest compared to posttest–no significant difference
  • pretest to follow-up– no significant difference

 

– SEI results:

  • pretest compared to posttest– no significant difference
  • pretest to follow-up– no significant difference

 

  • OUTCOME #2 –Improved to infer how they (the Ps) would feel in the described situations—LEAS-Self

– FAR results:

  • pretest compared to posttest- post test is significantly higher
  • pretest to follow-up– no significant difference

 

– SEI results:

  • pretest compared to posttest– no significant difference
  • pretest to follow-up – post test is significantly higher

 

  • OUTCOME #3: Improved to infer how others would feel in the described situations—LEAS-Others

– FAR results:

  • pretest compared to posttest– post test is significantly higher
  • pretest to follow-up– no significant difference
  • posttest to follow-up – no significant difference

 

– SEI results:

  • pretest compared to posttest– no significant difference
  • pretest to follow-up– no significant difference

 

  • OUTCOME #4: Improved performance on a test of the recognition of facial affect—DANVA2-AF

– FAR results:

  • pretest #1 compared to posttest– post test is significantly higher
  • pretest #2 compared to posttest– post test is significantly higher
  • pretest #1 and #2 to follow-up– follow up is significantly higher
  • posttest to follow up—significant difference *

 

– SEI results:

  • pretest #1 compared to posttest– post test is significantly higher*
  • pretest #2 compared to posttest– post test is significantly higher*
  • pretest #1 to follow-up– follow up is significantly higher *
  • pretest #2 to follow-up– no significant difference

 

 

  • OUTCOME #5: Improved ability to recognize affect from short videos which included facial, prosodic, gestural, and postural cues—EET

– FAR results:

  • pretest compared to posttest– no significant difference
  • pretest to follow-up– no significant difference

 

– SEI results:

  • pretest compared to posttest- no significant difference
  • pretest to follow-up– no significant difference

 

  • OUTCOME #6: Improved social-emotional functioning—BAFQ

– FAR results:

  • pretest compared to posttest– post test is significantly better (lower scores reflect improvement)

 

– SEI results:

  • pretest compared to posttest– no significant difference

 

 

– What was the statistical test used to determine significance? t-test; ANOVA

 

– Were confidence interval (CI) provided? No

 

 

  1. What is the clinical significance

 

– The investigators provided the following evidence-based EBP measure: Standardized Mean Difference- (d)

 

– Results of EBP testing and the interpretation

 

  • OUTCOME #1: Improved performance on test of the recognition of prosodic affect—DANVA2-AP

– FAR results:

  • pretest compared to posttest— d = 0.02   very small

 

– SEI results:

  • pretest compared to posttest– d = 0.07 very small

 

  • OUTCOME #2 –Improved to infer how they (the Ps) would feel in the described situations—LEAS-Self

– FAR results:

  • pretest compared to posttest    d = 0.81 large

 

– SEI results:

  • pretest compared to posttest d = 0.37 small to medium

 

  • OUTCOME #3: Improved to infer how others would feel in the described situations—LEAS-Others

– FAR results:

  • pretest compared to posttest   d = 1.37   very large

 

– SEI results:

  • pretest compared to posttest   d = 0.14 small

 

  • OUTCOME #4: Improved performance on a test of the recognition of facial affect—DANVA2-AF

– FAR results:

  • pretest #1 compared to posttest: d = 2.07   very large

 

– SEI results:

  • pretest #2 compared to posttest: d = 0.33   small

 

  • OUTCOME #5: Improved ability to recognize affect from short videos which included facial, prosodic, gestural, and postural cues—EET

– FAR results:

  • pretest compared to posttest d = 0.44 medium

 

– SEI results:

  • pretest compared to posttest d = 0.57 medium

 

  • OUTCOME #6: Improved social-emotional functioning—BAFQ

– FAR results:

  • pretest compared to posttest   d = 0.24 small

 

– SEI results:

  • pretest compared to posttest   d = 0.09 very small

 

 

  1. Were maintenance data reported? Yes. With the exception of OUTCOME #6 (Improved social-emotional functioning—BAFQ), all of the outcomes were tested at a follow-up session, administered 2 weeks after the termination of therapy. For the most part, gains made in treatment were maintained or at least improved from the pretest.

 

 

  1. Were generalization data reported? Yes.
  • Affective prosody was not directly treated in either intervention; there was not generalization to affective prosody.
  • FAR intervention did not include training in the interpretation of short video stories but OUTCOME #5 (Improved ability to recognize affect from short videos which included facial, prosodic, gestural, and postural cues) assessed this skill. Neither intervention yielded significant improvement on this outcome.
  • SEI intervention was not concerned with inferring emotions. OUTCOME #2 (Improved to infer how Ps would feel in the described situations—LEAS-Self) and

OUTCOME #3 (Improved to infer how others would feel in the described situations—LEAS-Others) assessed inference. The SEI group did not improve significantly on these outcomes although FAR improved significantly.

 

 

  1. Describe briefly the experimental design of the investigation.
  • The investigators randomly assigned Ps with ABI to one of 2 treatment groups: FAR and SEI.
  • For the most part, the Ps in the 2 groups displayed similar characteristics prior to intervention.
  • The investigators assessed the Ps 3 times: pretests (2 weeks before intervention and immediately before the initiation of therapy), posttest (within 3 days of the end of therapy), and follow-up (2 weeks after the termination of therapy.) One outcome (#6) was only measured at the pre- and post-test assessments.
  • For most part, there was not a significant difference between the first and second pretests, accordingly the pre- and post- test scores of most outcomes were combined for statistical analysis.
  • Therapy sessions were individual, 1 hour long, 3 sessions a week, for 2 to 3 weeks. The average number of sessions per P for both groups was 6.5.
  • The statistical comparisons compared

– pretests vs posttest,

– pretests vs follow up,

– posttest vs follow up, and

– pretest #1 vs pretest #2.

  • The investigators did not report the results of comparisons of the 2 different treatment groups (Far vs SEI.)

 

 

ASSIGNED OVERALL GRADE FOR QUALITY OF EXTERNAL EVIDENCE: A-

 

 

SUMMARY OF INTERVENTION

 

PURPOSE: To compare the effectiveness of interventions to improve affective recognition.

 

POPULATION: Traumatic Brain Injury

 

MODALITY TARGETED: comprehension

 

ELEMENTS/FUNCTIONS OF PROSODY TARGETED: prosodic affect

 

OTHER TARGETS: facial affect, social-emotional function, recognition of affect, inference of affect. (NOTE: These outcomes are not cross referenced because it was not clear that prosody was used to treat these targets.)

 

DOSAGE: individual, 1 hour sessions, 3 times a week for 2-3 weeks.

 

ADMINISTRATOR: “therapist” but the type of therapist is unclear

 

MAJOR COMPONENTS:

 

There were 2 intervention: FAR and SEI

 

FAR

 

  • FAR focused on helping Ps attend to their own affect and inferring the affect of others by attending to facial cues.

 

  • The investigators provided an Appendix detailing the intervention.

 

  • FAR comprised 2 phases: facial recognition and recognition of P’s own affect

 

  • Phase 1

 

– The clinician (C) used a computer program to present facial cues associated with selected affects.

– C and P discussed each of the emotions and applied them to Ps’ experiences.

– C provided feedback to P.

– The difficulty of the task of identifying the emotion increased and the cues for attending to facial representations of affect decreased as P progressed.

– Criterion for each of the exercises in this Phase was 85% correct.

 

Phase 2

 

– P described the physical reactions he/she expected to experience with each of the targeted emotions.

– C selected exemplars/events from the P’s discussions of his/her experiences associated with emotions/affects noted in the previous phase.

– For each of the events, C asked P to

  • provide details about the event
  • describe the context preceding the event
  • describe how he/she felt about the event including physical responses to the emotion
  • suggest why he/she felt the way he/she did
  • describe how he/she responded to the event
  • note the consequences

 

 

SEI

 

  • SEI focused on interpreting social stories and inferring affect from the context.

 

  • C presented 20 social stories focusing on

– attention to cues to emotions of characters

– relating the social story to the P’s life experiences

 

  • The C presented the social stories visually and orally using a computer.

 

  • Key words and cues about emotional status were highlighted in the social story texts on the computer.

 

  • C directed P to read the story and then answer questions about the emotional status of the characters.

 

  • C provided feedback as well as directed P to reread portions of the text, as necessary.

 

  • Following a correct answer, P asked C

– why he/she thought the answer was correct

– how he/she would have felt in a similar context

– if he/she had experienced a similar situation

– if he/she had experience a similar emotion

 

  • Criterion for advancement was 80% correct.

 

________________________

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: