23 / 44
ACQ
uiring knowledge
in
speech
,
language and hearing
, Volume 10, Number 2 2008
57
Work– l i f e balance : preserv i ng your soul
Clinical case
Subject
The client involved in this project, DB, was a 4-year-old boy.
He was initially seen by a non-PROMPT trained therapist,
and was found to be suitable for a trial of PROMPT therapy.
He was then referred to a PROMPT trained therapist for on
going therapy. The potential of GAS as an outcome measure
for PROMPT therapy arose and was seen as an opportunity to
explore the tool for this purpose. Much of the background
information reported here was gleaned retrospectively from
reports and progress notes from the initial therapist; therefore
sufficient baseline measures desirable for the present study
are lacking.
An initial assessment of DB’s speech and language skills
prior to commencement of PROMPT therapy indicated severe
speech and language difficulties. His hearing was found to be
within the normal range and gross and fine motor skills
(except oromotor skills) were developing normally. He used
mainly vowel sounds when speaking and was unable to be
fully understood in social situations. DB had an expressive
vocabulary of 4–10 recognisable words (e.g., Mum and “gar”
for “car”), using four different consonant sounds (/g/, /h/,
/m/, and /j/). He was unable to complete a formal articulation
assessment due to his limited vocabulary. DB did not use any
grammatical markers, and his mean length of utterance was
1–2 words. He also showed significant delays in receptive
language including difficulty understanding questions and
grammatical markers, and difficulty following multi-step
directions. Social communication difficulties were also noted,
including difficulty initiating communication (including non-
commencement of work towards the goals, and a specific
timeframe must be set for independent evaluation of progress,
or rating, using the pre-set scale (Cardillo & Choate, 1994).
The follow-up rater collates a “summary score” which is used
to find a corresponding T-score (Cardillo, 1994). GAS has been
found to have high inter-rater reliability (Stolee, Rockwood,
Fox & Streiner, 1992) while reliability over time and between
scales has not yet been shown (Caslyn & Davidson, 1978;
Cardillo & Smith, 1994). The validity of GAS as an outcome
measure has been criticised (e.g., Caslyn & Davidson, 1978);
however, a number studies (e.g., Goodyear & Bitter; 1974;
Williams & Stieg, 1987) found GAS to be a valid and reliable
outcome measure in the rehabilitation setting (Hurn, Knee
bone & Cropley, 2006). A review of research in this area has
shown that while GAS has flaws in terms of a “general
measure of outcome”, it is a strong measure of “treatment-
induced change” (Smith & Cardillo, 1994, p. 272). It is recom
mended that it be used in conjunction with other standardised
forms of assessment for pre and post testing (Smith &
Cardillo, 1994; Malec, 1999). GAS has been found to measure
“clinically important change” (Rockwood, Stolee & Fox; 1993,
Rockwood, Joyce & Stolee; 1997), including small changes
which may not be detected in standardised assessments
(Rockwood, Stolee & Fox; 1993; Malec 1999).
PROMPT uses a structured format for a “System Analysis
Observation” based on the stages of the Motor Speech Hierarchy
(TPI, 2007) which forms the basis of assessment, and leads
into the development of three main goals for therapy (TPI,
2007). GAS can be seen as a natural extension of this, allowing
for systematic evaluation of the outcomes of these goals.
Table 1: Goal attainment follow-up guide; DB’s therapy goals
Goal 1
Goal 2
Goal 3
Goal 4
Level of
attainment
Much less
than expected
–2
Somewhat
less than
expected
–1
Expected
level of
outcome
0
Somewhat
more than
expected
+1
Much more
than expected
+2
DB will count from 1-5
without PROMPT
DB will need surface
PROMPTs to produce an
approximation
1
of each
number.
DB will produce one
number without surface
PROMPTs, yet will need
PROMPT for other
numbers.
DB will produce two
numbers without
surface PROMPTs. (may
need visual/auditory
cues)
DB will produce three
numbers without
surface PROMPTs. (may
need visual/auditory
cues)
DB will produce all
numbers without
surface PROMPTs.
DB will pronounce
words with changing
jaw gradient (4-1)
DB will need parameter
PROMPT for jaw
gradient on all target
words
DB will need parameter
PROMPT for 80-90% of
target words.
DB will need parameter
PROMPT for 60-70% of
target words. (may need
visual/auditory cues)
DB will need parameter
PROMPT for 50% of
target words. (may need
visual/auditory cues)
DB will need parameter
PROMPT for less than
50% of target words.
DB will produce CV
words with rounded
vowels
DB will produce 50% of
target words with
surface PROMPT.
DB will produce 70% of
target words with
surface PROMPT.
DB will produce 70% of
target words with verbal
cue to round lips.
DB will produce 90% of
target words with verbal
cue to round lips.
DB will produce 70% of
target words without
verbal cue or PROMPT.
DB will produce
alveolar consonants in
the initial position of
CV words
DB will elevate his
tongue to the alveolar
ridge for /n/ with
PROMPT <50%.
DB will elevate his
tongue to the alveolar
ridge for /n/ /t/ and
/d/ with PROMPT
<50%.
DB will achieve correct
tongue placement for
alveolar sounds
2
on 50%
of attempts with
PROMPT.
DB will produce
alveolar sounds in
isolation with PROMPT
DB will produce
alveolar sounds at the
beginning of CV words
with PROMPT.
1 Expected pronunciation of numbers: one- ‘ooah’, two- ‘oo’, three- ‘ee’, four- ‘for’, five- ‘fah’ or ‘fy’ (as in ‘my’).
2 Tongue placement without production of sound (ie. no airflow) acceptable.