Intensive Stuttering Group Therapy for School-age Children and the Effect of Emotional Reactivity on Therapy Outcomes

The investigators hypothesized that (1) implementation of the usage of individualized desensitization exercises in an intensive stuttering group therapy created more effective treatment results than standard implementation of the desensitization exercises in an intensive stuttering group therapy in school-age children who stutter, (2) recovery rates will differ according to exhibited affects before stuttered utterances, (3) heart rate and skin conductivity findings before, during and after the stuttered utterances will differ from the general findings.

Children (n=12) were recruited and randomized in a 1:1 ratio either Control group (n=6) or Study group (n=6) by the study assistant. Randomization was stratified by sex, using a block size of two. For the first part of the study 6 study assessors (speech and language therapy interns who were experienced in the assessment procedure) who were not otherwise involved in the study, were participated in baseline and follow-up assessments. Children were blinded to treatment arm. Video recordings of assessments of the Turkish version of Stuttering Severity Scale-4 were made. Offline assessments of the stuttering severity were done by two research assistants. One of the research assistants was blinded to group assignment and was not otherwise involved in the studies. Video recordings of assessments were labeled by decimal numbers randomly. The masking of the data was done by the study assistant. The second part of the investigation conducted with children in the study group.

For the first part of the study, intervention and family training was applied by a professional therapist who has 7 years of experience in stuttering therapy. When children applied for the baseline assessments, parent training was applied. Training included; (a) general information on stuttering, (b) underlying facts, and (c) how to communicate with their children. At the training session, parents were given an information brochure for teachers of their child. Family training was conducted a month before intensive stuttering group therapy.

The intervention used a hybrid approach which was combined with the evidence-based techniques: speech reconstruction includes fluency shaping and stuttering modification; cognitive restructuring; emotional restructuring; desensitization; video self-modeling.

Standard therapy application included 7 components:

1. Recognition: Speech mechanism, stuttering, fluency or stuttering modification technic, emotions, thoughts, bodily actions, behaviors etc. What is it? Where/When/How to use? Transferring the rationale of the exercise to the participants.
2. Desensitization: It is the process of gradually exposing the participants to the stuttering induced situations (in which there is a negative and very positive mood) by first determining the stuttering induced situations (anxiety ladder exercise used, to begin with). Problem-solving exercises and role-model exercises applied.
3. Technic usage:
4. Video-model had prepared for fluency shaping which includes easy start, light articulatory contacts, connection, and pausing techniques. At first video model had been watched for once; than clinician asked the children to identify the speech characteristics and they watched the video-model again, chorus readings were done altogether with the clinician and the group, clinician's help retracted gradually and children read the passages alone.
5. To generalize the fluency technique into speech gradual increase method was used; one-word production games, two-word production games, sentence production games, conversation activities, reading aloud activities, role-model activities were used.
6. For stuttering modification pull-out technique was introduced during daily activities.
7. Diversified Technic Usage: The skills learned have been used in complex tasks. For example; Making presentations in front of the group and answering questions.
8. Problem-Solving Skills: Identifying problem situations, tagging the emotions, brainstorming, decision making, role-modeling.
9. Generalization: In order to generalize the learned skills to the daily activities, outside of the therapy room, orienteering activity was organized with the adults they hadn't met before. Each child completed a set of tasks outside of the therapy room. Presentation in front of a group of adults (their parents) was organized.
10. Maintenance: In order to maintain the skills gained in therapy after therapy, video self-modeling technique was used. On the last day of therapy, each subject was visually fluent during free speech and reading tasks and was asked to watch these videos every day after therapy.

For both groups of children, desensitization exercises were conducted. Standard intensive stuttering group therapy was applied to the control group while in the study group, clinician's observation was included in the intervention process. Predetermined stuttering inducing situation's table was prepared for each participant. The scenario table was included 5 situations. Clinician observed the children and marked the situations which increased their stuttering. Mostly marked situations were identified for each child. At the end of the first-week children were given instructions for the second week of the therapy. For example; if the child's stuttering severity increased during reading aloud in front of the group, the intensity of the reading-aloud in front of the group exercises was increased for him/her. They were informed that all of them would have special tasks.

Stuttering severity and cognitive, affective, linguistic and social score measures were done 5 times; 1) one month before the therapy, 2) on the first day of the therapy, 3) on the last day of the therapy, 4) one month after the therapy, and 5) three months after the therapy.

For the second part of the study, the images required for the Emotion Facial Action Coding System analysis were recorded during the therapy through CH_TECH SONY CHIPSET 2 MEGAPIXEL IP cameras. A total of six cameras were placed in the therapy room. The analysis of the facial action coding system was carried out by a certified researcher. Increased stuttering moments of subjects in the study group were detected by monitoring and transcription of the camera and microphone recordings at the end of the day. During the time before the fluency technique was thought to the subjects, 10 stuttering moments were identified and the 5-second window before the moment of stuttering was coded.

The evaluations were classified in 2 groups, positive or negative affect. Facial expressions which are expressions specified as non-encoded in Emotion Facial Action Coding System are identified as neutral facial expressions. Positive affect codings were done via observation of facial expressions; AU 12 (extraction of the lip corner) and AU 6 +12 (cheek up with the extraction of the lip corner). Positive affect may be accompanied by AU1 + 2 (eyebrow elevation), AU25 (mouth opening) or AU 26 (loosening of the jaw) (33). If AU12 occurs without AU6, AU12 should occur at least C severity in the intensity classification from 5 to Likert type A to E If AU12 is present with AU6, the intensity of AU12 is B 109, 139. For the coding of negative affects; AU 9 (nose creasing); AU 10 (upper lip elevation); AU 14 (pitting); AU 15 (depression of the lip corner); It is necessary that at least one of the codes AU 20 (lips stretching) and AU 1 + 4 (lifting the eyebrows to the midline and approaching) occur at intensity B.

Skin conductivity and heart rate recordings of participants in the study group were obtained via E4 wristbands. The wristbands were tested on both wrists during the therapy of the control group. As the non-dominant hand wrist's recordings resulted with lower noise levels, the wristbands were worn on the non-dominant side. For the analysis of skin conductivity; 1) online version of EDA-Explorer, 2) Ledalab and 3) Matlab was used. For the heart rate analysis, Matlab was used.