Research which led to the development of the CAFET-for-Kids program was in part supported by a Department of Health and Human Services Small Business Innovative Research grant (#2 R44 NS 22756-02) to Annandale Fluency Clinic, Inc. The granting agency was the National Institutes of Neuro-Communicative Disorders and Stroke. Under this grant, a clinical study of the efficacy of the CAFET-for-Kids program was performed from Nov. 1987 through May 1989.
The following is an overview of the method used in the testing of the program and the results with that experimental group of children at the Annandale Fluency Clinic. A detailed description of this study is in preparation and will be submitted for peer review.
Subject Selection
The children ranged in age from five years, 0 months to nine years, eleven months at the time of the
evaluation. Subjects were obtained through announcements and interviews in local Washington
metropolitan area newspapers and on local radio talk shows. Since the program was supported by NINCDS,
there was no charge for the treatment. We were fortunate to get a socio-economic cross-section of the area,
with one exception: all children were of two-parent families. This was not by design. It happened
coincidentally. We were concerned that if we stated certain requirements in these announcements, parents might
be tempted to "bend the truth" a little in order to get their child into the program. Therefore,
the only requirement aired was that the child have passed his or her fifth birthday, but not yet
the tenth.
Parents who called were sent an application form which amounted to a fairly complete case history. They were also sent complete information on the nature of the program to be tested. From the resulting twenty-nine applicants, only those were accepted who had, according to the parents, been stuttering at least twelve months. In fact, in the final group of subjects, none had stuttered less than two years.
We further narrowed the group in that no child was accepted who had any other physical handicap or neurological disorder. Hearing acuity was within normal limits for all children. Twelve children were then evaluated, and only those were accepted whose receptive and expressive language scores were at the 50th percentile or above. Several parents whose children had been initially accepted into the program declined to participate because of the requirement that the child be brought to therapy twice weekly. Two children who started the study dropped out, in one case because the mother felt that it was too stressful and in the second case the child was unwilling to miss his after school play time.
Pre-Treatment Evaluation Protocol
The evaluation included a pure-tone audiometric screening at 20dBHL (ANSI, 1969), the Clinical
Evaluation of Language Functions (Wiig and Semel, 1980) Elementary Level Screening, a Peabody Picture
Vocabulary Test-Revised, and a parental interview which included preparation of a Cooper Stuttering
Chronicity Prediction Check-list (Cooper, 1976).
Speech Samples
In-Clinic
At the evaluation, five one-minute samples were taken at five levels of linguistic complexity
while the child wore two respiratory sensors, one around the upper chest, and the second at a level
approximately one inch below the bottom rib. The tie-clip microphone was clipped to a neck band holding it
six inches from the child's mouth. Two examiners were in the room with the child. The child was seated in a child's
chair opposite the examiners. The examiners were able to see the computer monitor but the child was not.
The child was asked to help in putting on the sensors and "setting up" the computer. It was explained that it was difficult for the computers to measure if the child did not sit fairly still, so the child was asked to keep his/her hands on the thighs. While one examiner set up the computer for the measurements, the other chatted with the child.
For each minute, the examiner explained the task to the child, then several practice trials were made. It was explained that the computer would "beep" when it was ready to listen. That meant the child could begin the talking task. All children were able to respond appropriately to the instructions. During the time the computer was on, the sampling rate was 1000Hz. That is, measurements were being taken from each channel at a rate of one every third millisecond. The data was stored on diskette.
When the examiner timed-out (using the space bar) to present a stimulus, the computer produced an audible tone. Another tone was produced when the timing recommenced. These were recorded with the child's and examiner's speech on high quality reel-to-reel audiotape, so that transcriptions could be made and compared to the computer measurements of respiratory and phonatory activity.
Speech tasks for this computerized data acquisition began with a sentence imitation task. Binet sentences were used, starting with the two trial items, and continuing up to the sentences appropriate for the child's chronological age. If time was left in the minute, it was filled with sentences from the CELF sentence imitation subtest.
The second and third tasks were story retelling using the Goldman-Fristoe Test of Articulation stories "A Bad Night for Jerry" and "Jack and Ricky." The computer was not timed-out after the child began retelling the story. The fourth task used the "Cookie Theft" picture from the Boston Aphasia Examination, with a story composed by the examiners. The story was filled with details, such as the children's names and ages, and the children were asked to retell as much as they could remember.
The final task was conversational. The child was asked to begin by stating his name and where he lived. The remainder was conversational, about toys, school, the child's family, vacation, sports, etc. Using the space bar made it possible to time-out when the examiner needed to take a talking turn.
Home Baselines
Using the instructions given in the appendix to this manual, parents were instructed to make baseline tapes.
High quality cassette tapes were supplied to the parents at the time of the evaluation. They were also given
a mailer in which to return the tape to the clinic for review. Five consecutive days of baseline samples
were required for a child to enter the treatment program.
Treatment Logistics
Frequency
It was the intent of this study to demonstrate not only the basic efficacy of the program in establishing
and maintaining fluent speech, but to see whether or not this could be done in a non-intensive, public school
type treatment paradigm. Therefore, children were seen twice weekly, for one half-hour session. The clinic
followed the local county school calendar, including school holidays, teacher workdays, snow-days, etc. Missed
sessions (for any reason) were not made up.
General Treatment Procedure
In order to more closely adhere to public school treatment environment, we limited parental
involvement. Parents were allowed to observe every fourth session at which time home exercises
were demonstrated. We relied primarily on a "speech" notebook" to communicate with parents.
The treatment program began in November (1987) and continued until Ma (1988). By the end of April, most children had attended 19 or 20 individual sessions, at which time group therapy was introduced. The children were seen in two groups (3 and 4 respectively) for a total of 3 sessions.
Results
The treatment program ended for all children in the experimental group in May of 1988.
In April of 1989, the parents were contacted, and a reevaluation was scheduled. Speech samples were
again, collected in-clinic using the computerized data acquisition program. The parents and child
were interviewed separately. All children who completed the program were available for follow-up.
The parents were again given a cassette tape for home baseline tapes.
Summary results are seen in the results table. The speech data given are averages of the five pre- and post-treatment home baseline tapes.
|
||||||||||||||||||||||||||||||
Of these seven children, two were considered by the parents to be still stuttering. S#2, who was the youngest,
demonstrated no appreciable gains. It should be noted that this was not a "relapse," but rather, this child did not
benefit from the program at any time. Several factors were involved, among them a long sickness during the
winter months which caused his attendance to be sporadic, emotional difficulties in adjusting to a younger sibling,
and overall delayed fine motor development. Although he had scored above the 50th percentile on
the standardized language tests, we soon discovered that his expressive language was at approximately the four year level.
This interfered significantly with the treatment. S#5, who was the second youngest of the group,
was a charming, extremely verbal excitable child. In our opinion, one year post treatment he was
demonstrating exclusively normal disfluencies. The parents, however, although they acknowledged his progress, were
not entirely satisfied with the results. The child himself stated that he thought his speech was "fine."
