INT. J. LANG. COMM. DIS.

, MARCHAPRIL 2007,
VOL. 42, NO. 2, 131153

Research Report
Behavioural intervention effects in dysarthria
following stroke: communication effectiveness,
intelligibility and dysarthria impact

Catherine Mackenzie and Anja Lowit

Department of Educational and Professional Studies, University of Strathclyde,
Glasgow, UK
(Received 28 September 2005; accepted 13 June 2006)

Abstract
Background: Dysarthria is a common post-stroke presentation. Its management
falls within the remit of the speech and language therapy profession. Little
controlled evaluation of the effects of intervention for dysarthria in stroke has
been reported.
Aims: The study aimed to determine the effects of a period of behavioural
communication intervention on communication effectiveness and intelligibility,
and of the speech disorders impact on the person with dysarthria.
Methods & Procedures: Eight people with dysarthria following stroke received
individually tailored intervention programmes of 16 sessions during an 8-week
period. Conversation, and reading aloud of connected speech and of single
words were sampled at four points, at intervals of around 2 months: two before
intervention and two after intervention. The data were perceptually evaluated by
ten listeners, blind to assessment point, for the overall effectiveness of
communication (conversation) and for single-word and reading intelligibility.
The impact of dysarthria was assessed at the beginning and end of the
intervention period.
Outcome & Results: The level of reliability of listener ratings was high. There was
no evidence of group change between assessment points for measures of
communication effectiveness during conversation, or for reading and word
intelligibility. Individual participant analysis indicated that five speakers
improved in at least one of these three measures. In all cases this followed
either stability or deterioration during the 2-month pre-intervention period.
There was very little evidence of deterioration during the 2 months after
withdrawal of treatment. For three participants no intervention-related speech

Address correspondence to: Catherine Mackenzie, Department of Educational and Professional

Studies, University of Strathclyde, Southbrae Drive, Glasgow G13 1PP, UK; e-mail: c.mackenzie
@strath.ac.uk

International Journal of Language & Communication Disorders

ISSN 1368-2822 print/ISSN 1460-6984 online # 2007 Royal College of Speech & Language Therapists
http://www.informahealthcare.com
DOI: 10.1080/13682820600861776
132 Catherine Mackenzie and Anja Lowit

benefit was demonstrated. Age, dysarthria severity or lesion information did

not appear to differentiate those who improved from those who did not.
Improvement occurred in all who began the intervention phase between 5 and 8
months following stroke onset. Group data indicated no change in the
Dysarthria Impact Profile in respect of three sections (The effect of dysarthria on me
as a person; How I feel others react to my speech; and How dysarthria affects my
communication with others). There was a significant difference between the start and
end of intervention for Accepting my dysarthria, suggesting a reduction in the
negative impact of dysarthria.
Conclusions: The results demonstrate that some individuals with dysarthria have a
capacity to respond positively to intervention, some months after stroke, and to
maintain this improvement following 2 months of no intervention.
Consideration is given to how the results of the present study may inform
subsequent phases of dysarthria stroke research.

Keywords: dysarthria, stroke intervention, outcome.

What this paper adds

Dysarthria is common after stroke. Whilst the literature suggests a wide range
of treatment approaches for this disorder, there has been little controlled
evaluation of treatment effects in the stroke population. This preliminary
study shows that significant improvements in communication effectiveness
and intelligibility can be achieved with a modest amount of input, and
maintained in some people with dysarthria subsequent to stroke. Examination
of the participant variables provides some suggestions as to future research
design with this population.

Introduction
The characteristics of dysarthria reflect abnormalities in the strength, speed, range,
timing or accuracy of speech movements, resulting from neuro-pathophysiological
conditions such as weakness and spasticity. Effects are evident in the respiratory,
laryngeal, velopharyngeal and oral articulatory subsystems, singly or in combination
(Yorkston et al., 2001). The efficiency, naturalness and intelligibility of communica-
tion may be affected. Reduced communication skill may be associated with loss,
disadvantage and stress. Society values communicative competence and may
demonstrate negative attitudes towards those who lack this. The speaker with
dysarthria may experience social discrimination or devaluation (Hustad et al. 1998).
Where the individual perceives that communication standard or effectiveness is
reduced, or where the attitudes and reactions of others lead to fear of failure, there
may be reduction in communication contact not only for the speaker with dysarthria
but also for communication partners (Yorkston et al. 1999). If participation in
communication situations is compromised, quality of life may be affected (Yorkston
1996). Thus dysarthria impacts psychologically, socially and emotionally (Lubinski
1991).
Intervention effects in dysarthria following stroke 133

Many differing communication intervention and supportive approaches are

used in dysarthria, such as communication partner education, the introduction of
augmentative or alternative communication, establishing motor control, and
prosthetic management. Behavioural management which aims to maximize
communication efficiency, naturalness and intelligibility through direct symptom
treatments and compensatory strategies (Yorkston et al. 1999) is considered to be the
most common approach in dysarthria (Duffy, 2005).
Dysarthria is reported to be the most frequently acquired speech and language
disorder (Enderby and Emerson 1995). According to the most conservative figures
20% of stroke patients present with dysarthria (Warlow et al. 2000). Stroke is
reported to account for over 29% of spastic and 13% of ataxic dysarthrias, and
smaller proportions of some other diagnostic groups (Duffy 2005). Although there
are strong arguments for practice in the management of communication disorders
to be grounded in evidence (Reilly 2004a), demonstration of a positive outcome for
intervention in dysarthria resulting from stroke is limited. Randomized control trials,
often regarded as the gold standard in healthcare for answering specific questions
about therapy evaluation (Reilly 2004b), have not been conducted in this field
(Sellars et al. 2001). The individual variability of communication presentation,
influenced by features such as stroke location and severity, and the consequent
effects on intervention targets, may indicate that a single case approach is an
appropriate method for evaluating the impact of therapy in dysarthria (Yorkston et al.
1999).
A number of case studies and case series studies have documented a positive
effect of intervention in stroke related dysarthria, in respect of the use of
alternative communication systems (Beukelman et al. 1985, Culp and Ladtkow
1992), speech supplementation using an alphabet board (Beukelman and Yorkston
1977) and palatal lift appliances for the treatment of velopharyngeal incompetence
(Yorkston et al. 2001). Speech and language therapists have an identified role in
dysarthria management (Royal College of Speech and Language Therapists 1996)
though many frequently used intervention approaches, including the commonly
used behavioural methods, lack demonstrated positive results with the stroke
population. As with other communication disorders, the facilitation of adjustment
to altered communication status, including acceptance and increased confidence in
communication, may be a goal of dysarthria management programmes. Here too
there is a lack of data which substantiate intervention-related change in dysarthria
impact.
According to the recommended standards of clinical outcome testing, a series
of well controlled single case treatment studies is a necessary initial exploratory
stage in the route to addressing efficacy, effectiveness and efficiency (Robey and
Schultz 1998). Such Phase 1 studies are brief, with small samples sizes and
do not include control patients (Robey and Schultz 1998). Yorkston et al. (1999)
recommend that intervention targets in dysarthria are individualized rather than
uni-dimensional. The current Phase 1 investigation was concerned with adults with
dysarthria following stroke. The main aim was to determine whether change in
communication effectiveness and intelligibility followed a defined period of
individually tailored behavioural communication intervention which addressed
the functional limitation of dysarthria (Yorkston et al. 1999). A secondary aim was
to determine whether intervention affected the impact of dysarthria on the
individual.
134 Catherine Mackenzie and Anja Lowit

Methods and procedures

Participants
Inclusion criteria were a diagnosis of dysarthria, resulting from stroke, occurring not
less than three months previously and no prior or concurrent additional neurological
disease. Participants were required to have no additional diagnoses of dementia or
aphasia, and to have hearing and vision adequate for participation in assessment and
therapy, as judged by the referring therapist. Nine adults who fulfilled the criteria
were recruited to the study, via three National Health Service (NHS) units. One (P4)
was later excluded because of a further neurological event. Before recruitment,
participants had received individual speech and language therapy intervention and
one participant had also been involved in group therapy. All were premorbidly right
handed and native speakers of English. Time since initial stroke ranged from 3 to
80 months and age from 47 to 75 years. Participant details are given in table 1. At
admission to the study (Assessment 1, see below; and table 3) effectiveness of
communication during conversation, rated by ten listeners on a 1 (low) to 7 (high)
scale, ranged from 1 to 6.

Table 1. Participant details

Months
Age post-stroke at Speech and language
Participant (years) Gender Assessment 1 Stroke details therapy history
1 61 male 80 right cerebellar 4 months individual
haemorrhage plus several group
courses
2 63 male 6 bilateral cerebellar infarcts 5 months individual
3 66 male 6 left partial anterior 4 months individual
circulation infarct
5 69 male 5 bilateral infarcts, larger on several individual
right, affecting frontal, sessions
temporal and parietal
lobes and putamen
and left cerebellar
hemisphere
6 47 male 29 left cerebellar infarct and several months
multiple scattered small individual
infarctions in both
cerebral hemispheres,
particularly right
occipital lobe
7 75 male 36 left hemisphere several months
individual
8 75 female 3 intracerebral bleed in left 3 months individual
basal ganglion; partial
anterior circulation
syndrome
9 64 female 12 bilateral posterior 3 months individual
circulation infarcts
Intervention effects in dysarthria following stroke 135

Assessment
Assessment was conducted at four points:
N Assessment 1 (A1):
2-month non-intervention period.
N Assessment 2 (A2):
2-month intervention period.
N Assessment 3 (A3):
2-month non-intervention period.
N Assessment 4 (A4).
Assessments took place in a quiet room at the participants clinic. Data were
recorded with a Sony Digital Audiotape Recorder, Model TCD-D8 and a Sony
electret condenser microphone, model ECM-MS907, positioned approximately
30 cm from the speakers mouth.
At each assessment session the following data were recorded:
N Ten minutes of conversation about home and family, health, stroke history
and effects, previous employment, interests and the participants typical day.
N Word intelligibility test (Kent et al. 1989). According to the authors, this test
examines acousticphonetic contrasts that are likely to be sensitive to
dysarthric impairments and contribute significantly to speech intelligibility.
The 70 stimulus words were presented individually for reading aloud.
N Reading connected speech. A passage of 150 words (Lowit-Leuschel and
Docherty 2001) was presented for reading aloud.
All assessments were administered by one research speech and language therapist
who had no previous contact with the participants and no involvement in the
intervention programme. Therapy was conducted by a graduating student (two
participants), a research speech and language therapist (four participants) and by the
referring therapists (two participants). Assessment and therapy sessions took place
in local hospital out-patient clinics.
In addition to the above measures, at the first and last treatment session of the
2-month intervention period, the treating therapist administered the Dysarthria
Impact Profile (DIP) (Walshe 2002). This comprises four sections, each containing
between ten and 14 statements to which the respondent indicates strong agree-
ment, agreement, neutrality, disagreement or strong disagreement. The four sec-
tions are: The effect of dysarthria on me as a person (e.g. I am as confident now as I was
before I had a speech problem); Accepting my dysarthria (e.g. I do not try to hide my
speech problem); How I feel others react to my speech (e.g. People are more
condescending to me now because of my speech); and How dysarthria affects my
communication with others (e.g. I rely on others to talk for me whenever possible).

Intervention
From the audiorecordings of the first assessments (A1), the authors drew up
behavioural intervention aims and approaches for each participant (table 2). The
treating therapists were responsible for session by session implementation of the
programmes, being guided in their selection of practice material by their emerging

Table 2. Intervention aims
Participant Intervention aims
1 A. To improve clarity in multisyllable words
B. To improve articulatory precision at sentence level, with particular
attention to voiced/voiceless plosive distinctions (t/d, p/b, k/g),
alveolar oral/nasal distinction (d/n) and l/r distinction
C. To participate intelligibly in dialogue
2 A. To improve articulatory precision in single words, progressing to
short connected utterances, with particular attention to /s/,
consonant clusters (e.g. /bl/ /st/), affricates, voicing contrasts,
vowel length contrasts (e.g. pit/peat)
B. To adopt optimum phrase length and rate for breath control and
maximum clarity in connected speech
C. To improve clarity in multisyllable words
D. To improve stress pattern in short connected utterances
E. To participate intelligibly in dialogue
136
Intervention approaches
A. Syllable by syllable practice of words with three (progressing to four
if appropriate) syllables, aiming for all syllables present and complete,
with correct stress pattern, encouraging exaggerated articulation. Practice
of stimulus words in isolation, then in short sentences
B. Initial minimal pair practice (e.g. pin/bin), encouraging exaggerated
articulation, progressing quickly to practising these words in sentence
context (e.g. She pointed to the bin/She pointed to the pin, Theres
a black coat/Theres a black goat). Encouragement of a slow rate in
sentence practice, with pausing between words
C. Initial practice using prepared written dialogue with therapist and
participant taking turns, progressing to spontaneous discussion topics
where therapist and patient take turns in producing a single sentence/
point each (with later extension to two sentences each). As before, with
attention to slowed rate, stress pattern and exaggerated articulation
A. Single word practice. Progress to practising stimulus words in short
Catherine Mackenzie and Anja Lowit
phrases, with slowed rate
B. During connected speech practice division of utterance into segments
of appropriate length, to avoid exhausting breath supply, encouraging
consistent pattern of attention to breath before each segment of speech
and a deliberate slow rate
C. As for P1, A
D. Contrastive stress drills (incorporating stimulus material from A and B)
E. As for P1, C.

Table 2. Continued
Participant Intervention aims
3 A. To improve clarity in multisyllable words, with particular attention
to adequate length of vowels and closure of syllables
B. To regularize rate of speech, reducing the number of short rushes
C. To participate in dialogue with maximum comprehensibility and
naturalness
5 A. To improve articulatory precision in single words, progressing to
short connected utterances, with particular attention to oral stops
(including m/b, n/d, g/g minimal pair practice for initial and final
targets, e.g. man/ban, come/cub), clusters with stop consonants
and affricates
B. To improve clarity in multisyllable words
C. To slow rate of speech, reducing the number of short rushes
D. To participate maximally in dialogue, sharing responsibility for the
communication exchange
Intervention effects in dysarthria following stroke
Intervention approaches
A. Syllable by syllable practice of multisyllable words, aiming for all syllables
present and closed, with adequate length of vowels and correct stress
pattern (e.g. everything, paramedic, television yesterday). Practice
of stimulus words in isolation, then in short sentences
B. Impose slowed rate, with pauses between words initially tapping on each
syllable in sentence stimuli comprising single syllable words, progressing
to tapping at stressed syllables and incorporating some longer words.
Reminders to keep a slow steady rate and pronounce all sounds in each
word, pausing between words
C. As for P1, C, with attention to slowed rate, stress pattern, vowel length
and syllable closure
A. As for P2, A
B. As for P1, A
C. As for P3, B
D. As for P1, C
137

Table 2.
Participant Intervention aims
6 A. To improve articulatory precision in connected speech, with
particular attention to /h/, / /, /t /, /st/, and /w/
B. To increase clarity in multisyllable words
C. To improve stress pattern in short connected utterances
D. To maximize naturalness in dialogue
7 A. To reduce and regularize rate of speech
B. To adopt optimum phrase length (and rate) for breath control
and maximum clarity in connected speech
C. To improve articulatory precision in connected speech, with
particular attention to /f/, /n/, /d/, /t /, /s/
(especially in clusters e.g. /sk/ and /str/)
D. To improve stress pattern in short connected utterances
E. To maximize comprehensibility and naturalness in dialogue.
138
Continued
Intervention approaches
A. Word level practice, but progressing quickly to practising these words in
longer context. Inclusion of minimal pair practice for s/st (e.g. sun/stun),
and w/ /* (e.g. wine/whine) initially with words in isolation then in
short sentence context
B. As for P1, A
Ci. Contrastive stress drills (incorporating some stimulus materials from A
and B)
Cii. Production of sentences that include mainly full vowels contrasted with
those including many unstressed/reduced vowels (Low et al. 2000), e.g.
John came back through France last Sunday. Don seemed quite cross
with John last week (full vowels). John was sick of Fred and Sandy. Don
was across at Jonathans (unstressed vowels)
D. As for P1, C, with attention to stress pattern, vowel length and
articulatory precision
A. As for P3, B
Catherine Mackenzie and Anja Lowit
B. As for P2, B
C. Word level practice, but progressing quickly to practising these words in
sentence context. Inclusion of minimal pair practice for f/p (e.g. fat/pat),
n/d (e.g. nip/dip), d/z (e.g. do/zoo) and minimal set n/d/z
(e.g. nip/dip/zip)
D. Contractive stress drills (incorporating some stimulus materials from A,
B and C
E. As for P1, C, with attention to slowed rate, stress pattern, breath control
and articulatory precision

Table 2.
Participant Intervention aims
8 A. To improve articulatory precision in connected speech, with
particular attention to / / /t /, /r/ clusters
(e.g. br, gr, tr, fr), /s/ clusters (e.g. /st/)
B. To adopt optimum phrase length for breath control and
maximum clarity in connected speech
C. To improve clarity in multisyllable words
D. To improve stress pattern in short connected utterances
E. To maximize articulatory precision and naturalness in dialogue
9 A. To improve articulatory precision in connected speech
B. To achieve a more natural stress pattern in connected utterances
C. To obtain efficient use of available breath supply in speech
D. To maximize naturalness and speaker confidence in dialogue
*w/ / is a phonemic distinction in Scottish English
Intervention effects in dysarthria following stroke
Continued
Intervention approaches
A. Word level practice, but progressing quickly to practising these words
in sentence context. Inclusion of minimal pair practice for b/br
(e.g. ban/bran), t/tr (e.g. tie/try), s/st (sore/store), s/
(e.g. sip/ship), s/h/ (e.g. sin/thin), /t ) (e.g. shop/chop)
B. As for P2, B
C. As for P1, A, with attention to correct stress pattern, vowel length and
encouraging exaggerated articulation
D. Contractive stress drills (incorporating some stimulus materials from A,
B and C
E. As for P1, C, with attention to stress pattern, breath control and
articulatory precision.
Ai. Minimal pair practice, especially word finally, for p/b (e.g. rope/robe),
k/g (e.g. knack/nag), t/d (e.g. shoot/should), m/b (e.g. sum/sub), with
emphasis mainly on connected utterances
Aii. Practise of phrases containing both /w/ and /l/ (e.g. a wee lamp,
a long way), and both /l/ and /r/ (e.g. a long rope) and use of these
phrases in sentence context (e.g. It was a long way home)
Aiii. Practise of nasal consonants, mainly word finally and intervocalically, in
words and across word boundaries (e.g. Im alright)
B. Practice of sentences with runs of monosyllabic words (e.g. I think John
might come)
C. As for P2, B, with attention to maintaining steadiness across the breathy
group
D. As for P2, E
139
140 Catherine Mackenzie and Anja Lowit

knowledge of the interests and general background of the individuals. Where

appropriate, words and phrases which occurred frequently in the participants
spontaneous utterances were also incorporated. For all participants the overall goal
was the maximization of comprehensibility and effectiveness in conversational
speech. Most treatment was directed beyond the single word level, though where
improvement of articulatory accuracy and multisyllabic word clarity were
intervention aims, some preliminary single word practice was undertaken before
incorporating the practice stimuli in connected speech. Motor learning principles
(Schmidt 1991, Yorkston et al. 1999) (see appendix 1) were applied. Therapists were
further guided to encourage consistent use of communication maximizing strategies,
such as exaggeration of articulatory gestures, giving full attention to communication
partners response, repeating where necessary, and reducing speaking rate (Yorkston
et al. 1999). Frequent use was made of recording and playback, followed by therapist
and participant evaluation. Therapists were directed to endeavour to maximize and
maintain motivation.
Participants received 16 therapy sessions of around 45 min during an 8-week
period. They were encouraged to carry out around 15 min of home practice every
day between sessions. No other communication intervention took place between A1
and A4.

Perceptual analysis of communication data

Ratings of the conversation, reading and single word data were carried out by ten
final-year speech and language therapy students who were enrolled in a dysarthria
advanced study module. From each 10-min conversational sample a section of 2
3 min was selected from around the middle, within which the participant was the
main contributor. For each of the three data sets, recordings were randomized and
coded as to assessment point and participant, to control for bias from these two
factors. The content of the recordings provided no clues as to assessment point.
Each student rated 32 conversation and word intelligibility samples and 31 reading
samples (there was one set of missing reading data):
N Conversation: ratings were made of the overall effectiveness of communication,
considering both intelligibility and naturalness, using a seven-point scale,
based on the work of Ball et al. (2004), where 15not at all effective and
75very effective.
N Word intelligibility test: each stimulus word was presented in a horizontal line
along with three distracters (e.g. stimulus bad: bad, bed, bet, pad), as
suggested by Kent et al. (1989). In addition, a fifth response option (?) was
provided for use where none of the four words provided was considered to
be the word attempted. The position of the stimulus in the display was
randomized. Participant responses were played at 5-s intervals. Listeners
circled the word they heard. For analysis of group performance, the number
of words not correctly identified was totalled across the ten listeners
(maximum possible score5700). Calculations were based on percentage of
correct identifications.
N Reading aloud: ratings of intelligibility were made using the procedure of direct
magnitude estimation (Weismer and Laures 2002). One of the participants
was chosen as the standard for comparison with the other speakers. Given
Intervention effects in dysarthria following stroke 141

that the current participant group presented with a range of dysarthria

severities it was decided to use a speaker with a moderate level of dysarthria
to provide maximum scope for ratings above and below the standard. The
speaker was selected by an experienced clinician by judging his speech
performance against the other participants in the study, as well as the wider
clinical population. A rating of 100 was notionally allocated to the speaker,
and using this standard as a reference, listeners then assigned a numerical
rating of intelligibility to each of the other samples. Thus a speaker
considered to be twice as intelligible as the standard would be given a rating
of 200, and a speaker half as intelligible given a rating of 50. The standard was
played repeatedly after groups of four samples. For the group analysis
geometric mean scores were calculated across the ten listeners.

Dysarthria Impact Profile (DIP)

Responses to each statement were assigned scores between 5 and 1 (55strongly
agree, 45agree, 35not sure, 25disagree, 15strongly disagree). The majority of
statements in the DIP are negatively expressed, e.g. I am not happy with my speech
as it is now, so high scores are associated with negative impact of dysarthria. For
those items which indicate a positive attitude, e.g. It does not bother me to admit
that I have a speech problem, scoring was reversed, so that strong agreement
attracted a score of 1. Scores were calculated for each of the four profile sections.

Statistical analysis
Inter-rater reliability for ratings of communication effectiveness (conversation data),
reading intelligibility and single word intelligibility was assessed by calculating the
intraclass correlation coefficient (ICC) (Shrout and Fleiss 1979) for the ten raters
scores on each of the three measures.
Friedman tests were used to assess variance across the four assessment points,
using the mean rater scores for communication effectiveness, reading intelligibility and
single word intelligibility. Friedman tests were also used to examine the performance of
each participant, using the individual scores of the listeners. Where a significant result
was present (p,0.05, two-tailed), Wilcoxon signed rank tests were used to compare
pairs of assessments (A1 and A2, A2 and A3, A3 and A4, A2 and A4). Wilcoxon tests
were used to compare the DIP beginning and end of intervention scores. Spearman
tests were used to examine correlations between performance measures.

Outcomes and results

Inter-rater reliability
Intraclass correlation coefficients indicated good agreement between listeners
(communication effectiveness, 0.93; reading intelligibility, 0.88; and word intellig-
ibility, 0.86).

Communication data: group results

There was no evidence of group change between assessment points for the
communication effectiveness (conversation), reading and word intelligibility
142 Catherine Mackenzie and Anja Lowit

measures (Friedman tests: conversation, chi-square54.650, p50.199; reading, chi-

square51.087, p50.780; words, chi-square50.750, p50.861). Mean scores are given
in tables 35.

Communication data: individual participant results

Variance across the four assessment points (A1, A2, A3, A4) was present for four
participants in communication effectiveness, for all eight participants in reading
intelligibility and for four participants in word intelligibility (table 6).
P2, P5, P7 and P8 showed significant variance in communication effectiveness
across the assessment points. Between the two pre-intervention assessments (A1/

Table 3. Effectiveness of communication: mean ratings across listeners (standard deviation)

Participant Assessment 1 Assessment 2 Assessment 3 Assessment 4

1 1.40 (0.52) 1.60 (0.52) 1.10 (0.32) 1.40 (0.52)
2 1.00 (0.00) 1.10 (0.32) 1.60 (0.52) 1.50 (0.53)
3 3.80 (0.63) 4.10 (0.74) 4.00 (1.15) 3.50 (0.97)
5 2.00 (0.67) 1.10 (0.32) 2.60 (0.97) 3.00 (0.82)
6 6.00 (0.67) 5.80 (0.92) 6.20 (0.92) 5.50 (0.97)
7 2.40 (0.52) 1.40 (0.52) 2.60 (0.52) 3.10 (0.57)
8 3.10 (1.10) 3.30 (0.48) 4.90 (0.57) 3.80 (0.79)
9 4.60 (0.97) 4.90 (0.57) 4.90 (0.57) 5.40 (0.84)

Scale 17, where 15low and 75high.

Table 4. Reading: direct magnitude estimations: geometric mean ratings across listeners

Participant Assessment 1 Assessment 2 Assessment 3 Assessment 4

1 70.42 94.42 70.62 66.68
2 * 15.78 35.15 25.58
3 87.21 98.58 167.06 165.47
5 28.55 15.00 43.79 47.91
6 177.27 170.50 135.08 183.14
7 80.59 110.02 91.64 110.01
8 164.85 147.17 186.43 127.76
9 181.47 159.33 122.11 208.03

*Missing data.
Higher scores are associated with increased intelligibility.

Table 5. Word intelligibility: mean percentage correct across listeners (standard deviation)

Participant Assessment 1 Assessment 2 Assessment 3 Assessment 4

1 85.57 (5.45) 85.71 (5.26) 85.29 (4.42) 84.00 (3.86)
2 69.23 (5.76) 61.00 (6.67) 78.43 (4.59) 68.43 (3.59)
3 87.29 (1.84) 89.00 (4.90) 88.29 (4.08) 89.71 (3.86)
5 69.86 (8.18) 69.00 (10.69) 73.00 (6.04) 79.29 (5.96)
6 85.57 (5.45) 85.71 (5.26) 85.29 (4.42) 84.00 (3.86)
7 89.43 (3.10) 89.00 (4.26) 87.43 (5.17) 88.14 (1.36)
8 91.86 (3.99) 94.00 (2.31) 94.86 (3.03) 90.71 (2.96)
9 90.43 (3.44) 92.57 (3.22) 89.86 (4.01) 91.14 (3.42)
 Intervention effects in dysarthria following stroke
Table 6. Across assessment point analysis: individual participants

CE A1 versus A2 A3 versus A2 RI A1 versus A2 A3 versus A2 WI A1 versus A2 A3 versus A2

Participant variance A2 versus A3 A4 versus A4 variance A2 versus A3 A4 versus A4 variance A2 versus A3 A4 versus A4
P1 n.s. * ++ 2 22 n.s.
P2 ** + + ** ++ + *** 2 ++ 22 ++
P3 n.s. *** ++ ++ n.s.
P5 *** 22 ++ ++ *** ++ ++ ** ++ ++
P6 n.s. * 2 + *** ++ ++
P7 *** 22 ++ + ++ ** + 22 + n.s.
P8 *** ++ 22 + *** ++ 22 22 ** 22 22
P9 n.s. *** 2 22 ++ + n.s.

CE, communication effectiveness (conversation); RI, reading intelligibility; WI, word intelligibility.
n.s., Variance across A1, A2, A3, A4 non-significant at 0.05, *p(0.05, **p(0.01, and ***p(0.001.
+, Improvement on earlier score, p(0.05; ++, improvement on earlier score, p(0.01; 2, deterioration on earlier score, p(0.05; 22, deterioration on earlier score,
p(0.01.

143
144 Catherine Mackenzie and Anja Lowit

A2) none of these four showed spontaneous improvement and for two (P5, P7) the
scores reduced. All four showed improvement following intervention (A2/A3), and
2 months after treatment had ended improvement relative to pre-intervention status
was still evident (A2/A4). In addition, one of the four participants (P7) continued to
make gains during the post-intervention period (A3/A4).
Across assessment variance was evident in reading intelligibility for all
participants. For four participants (P2, P3, P5, P8) there was intervention-related
improvement (A2/A3). Three of these showed stability prior to intervention (A1/
A2). An A1 score was not available for P2. For three participants (P2, P3, P5) the
gains relative to pre-treatment status were still evident at A4. For the four
participants who did not show an A2/A3 gain, deterioration during intervention was
evident and in all of these there were indications of increased intelligibility during
non-intervention periods.
Four participants (P2, P5, P6, P8) showed significant variance in single word
intelligibility, but there was little indication of an intervention link. Only P2 showed
an A2/A3 increase, and improvement relative to before intervention was maintained
at A4, despite a significant loss in the 2-month post-intervention period (A3/A4).
This participant also showed a decrease in intelligibility between A1 and A2. P5 and
P6, for whom no change was apparent at the end of treatment, showed A3/A4 gains
and P6 showed a similar gain in the pre-treatment period. P5 had a significantly
better intelligibility score at A4 compared with the pre-intervention level. For P8 the
variance was attributable to A3/A4 deterioration.
To investigate association between performance in communication effectiveness
and reading intelligibility, A2 and A3 scores were examined. These comparisons
were selected as reflecting immediately before and after intervention status in the
two measures for which there was some evidence of an intervention-related change.
Correlations were significant (A2, rs520.85, p,0.01; A3, rs50.73, p,0.05).

Dysarthria Impact Profile (DIP)

Mean group scores for the four sections of the DIP are given in table 7. For three
sections, no changes were present between the start and completion of the
intervention period: The effect of dysarthria on me as a person (Z521.193, p50.233), How
I feel others react to my speech (Z520.491, p50.623), and How dysarthria affects my
communication with others (Z521.103, p50.270). For the remaining section, Accepting
my dysarthria, a significant difference was present, reflecting a reduction in the
negative impact of dysarthria (Z521.970, p50.049). Figure 1 shows the total scores

Table 7. Dysarthria Impact Profile: mean group scores (standard deviation) at the beginning
and the end of intervention

Beginning End
Section A (maximum 60) 43.50 (4.21) 39.88 (8.39)
Section B (maximum 50) 32.25 (6.27) 27.75 (6.11)*
Section C (maximum 70) 38.63 (7.15) 39.00 (9.90)
Section D (maximum 60) 32.38 (6.16) 30.50 (6.32)

*p,0.05.
Lower scores are associated with decreased impact.
Intervention effects in dysarthria following stroke 145

Figure 1. Dysarthria impact profile: section B accepting my dysarthria: beginning (BA) and end of
intervention (BB) scores (maximum550).

of each participant for this section of the DIP, at the start of intervention (BA) and
at the end of intervention (BB). All participants, with the exception of P1, had a
reduced raw score at the second assessment, indicating that speakers were generally
less affected by their dysarthria.
To investigate association between speech measures and DIP, communication
effectiveness and reading intelligibility at A2 and A3 were examined in relation to
BA and BB scores, respectively. These comparisons were selected as reflecting
immediately before and after intervention status, in measures for which there was
some evidence of intervention-related change, as above. Correlations were non-
significant before intervention (BA/A2 communication effectiveness, rs520.20,
BA/A2 reading intelligibility, rs520.33) and also post-intervention for reading
intelligibility (BB/A3, rs520.59). However, the correlation with communication
effectiveness showed a significant result post-intervention (BB/A3, rs520.76,
p,0.05).

Discussion
Performance trends
Given the small number of cases in this investigation, it is not surprising that
significant communication gains were not demonstrated across the group. However
within the group, individuals did show the capacity to respond positively to
intervention, some months or years after stroke, and to maintain this improvement
following 2 months of no intervention. The eight participants can be broadly
separated into two subgroups: five who showed improvement in at least one of the
three speech measures (P2, P3, P5, P7, P8) and three for whom no intervention-
related benefit was demonstrated (P1, P6, P9).
Following stroke, the natural course is for some spontaneous recovery of
function, of variable extent and timescale. The rate of recovery is fastest in the first
few weeks (Wade 1992) and although change may continue for many months or
146 Catherine Mackenzie and Anja Lowit

even years (Dombovy and Bach-Y-Rita 1988), most improvement is observed to

occur within the first three months (Royal College of Physicians 1989). Controlling
for spontaneous change is an important feature of intervention research design.
Where, as in this investigation, participants act as their own controls (Howard 1986),
it is desirable that stability is present before the initiation of treatment. If
improvement then coincides with the intervention period this may be interpreted as
a probable treatment effect. If no improvement occurs with treatment, the given
treatment may be judged as ineffective for that individual. Status following cessa-
tion of intervention should also be considered. Where gains are made but not
maintained, this suggests that the treatment has achieved only short-term benefits
and continuing intervention of some form is required. The assessment protocol was
that non-treatment periods, of equivalent length to the intervention period,
preceded and followed the intervention phase, as used in a previous investigation of
aphasia response to intervention (Mackenzie 1991).
In the 2-month period prior to intervention, participants generally demons-
trated stability or deterioration. In this interval there were very few indications of
improvement: two participants in reading intelligibility and one in word intelligibility,
and in none of these did an intervention linked (A2/A3) improvement in the same
measure occur. Interestingly these were the three participants (P1, P6, P7) who were
longest from stroke onset. In all cases where intervention-related improvement was
present, this followed either stability or deterioration during the 2-month pre-
intervention period. This may reasonably be taken as an indication that the
improvement was related to treatment, as opposed to being a naturally occurring
event. Interventionists hope that where treatment related gains do occur the
standard achieved can be maintained when the treatment has been discontinued.
This was the case for all four participants with A2/A3 improvement in
communication effectiveness, and for three of the four with improvement in
reading intelligibility and for the one participant with A2/A3 word intelligibility
improvement. Further significant gains did not occur in the 2 months following the
end of treatment, with one exception (P7, communication effectiveness).
Improvement following cessation of intervention may occur where a participant
continues to apply the advice given during treatment and has become his own
speech therapist (Wertz 1978: 87). There were also a few examples of A3/A4
improvement where the A2/A3 comparison had indicated no intervention-related
progress (P6, P7, P9, reading intelligibility). In such cases it may be that there is a
delayed effect of intervention or this may be an indication of a general instability, as
may be the case with A1/A2 gains.
There was very little evidence of gains in single word intelligibility, despite the
improvements which were present in communication effectiveness and reading
intelligibility measures. The only participant for whom an A2/A3 gain was present had
the lowest score of the group. Before intervention, six of the eight participants had
high scores (above 85%) in this test, so there may have been little scope for
improvement. Whereas connected speech data provides context, such single word
audiorecorded intelligibility testing requires both speaker and listener to deal with
decontextualized words. However for the speaker with dysarthria, being able to focus
on a single word may be facilitative of maximum accuracy and, for the listener absence
of context is compensated for by the multiple choice format of this assessment. A
word may thus be identified correctly despite being characterized by distortions or
substitutions. Both factors may contribute to the high scores most participants had in
Intervention effects in dysarthria following stroke 147

this test and may have limited the scope for increases in intelligibility. Improvements in
accuracy of phonetic production are thus not necessarily reflected in the single word
intelligibility measure. Additional analysis of these data, utilizing narrow phonetic
transcription, is in progress to more closely examine the characteristics of the
articulatory productions and to ascertain whether intervention was associated with
improvement in production accuracy in the word stimuli of this test. Within the
current study, connected speech intelligibility and communicative effectiveness
emerge as the informative outcome measures for therapy.
Multidimensional assessments are often used in standard clinical practice. Such
impairment focused assessments (e.g. Frenchay Dysarthria Assessment; Enderby
1983), are not necessarily measures of communication change. Furthermore they
include many non-speech tasks, such as tongue protrusion, which while providing a
view of the integrity of the motor system, may be poorly correlated with
communication (Duffy and Kent 2001). Also without videorecords, or having
several judges present at the data collection point, inter-rater agreement on such
tasks cannot be demonstrated. The importance of capturing naturalistic data is well
recognized (Enderby and Emerson 1995, Lowit-Leuschel and Docherty 2001) and
the measures selected for this investigation were of the effectiveness of
communication and intelligibility at single word and connected speech levels. The
intraclass correlation coefficients for these three measures were in the range 0.86
0.93, demonstrating high reliability on the analysis sets. Significant correlations
between performance on the communication effectiveness and reading intelligibility
measures both before and at the end of the treatment period indicates that there is
an association between performance in conversation and connected reading.
For three participants (P1, P6, P9) the treatment provided appeared not to be
successful, in terms of achieving improvement in communication effectiveness or
intelligibility. For these three cases either no across assessment variance was present
or the variance present was associated with deterioration over the intervention
period and /or gains or deterioration during non-intervention periods. It is however
to be noted that both P6 and P9 followed the group trend as regards reduction at
the end of the intervention period in the DIP Accepting my dysarthria section. This
may indicate that despite the absence of improved communication some progress
was made in terms of adjustment.

Prognostic factors

Common sense dictates that not all individuals are equally good candidates for
treatment.
(Holland et al. 1996, p. 39)
In acquired dysarthria there is an absence of reliable prognostic data. Research
which looks not only at group outcome but also examines the response of
individuals, may provide some insight into candidate suitability for particular
treatment regimes. It is therefore appropriate to explore tentatively some of the
variables which may be relevant. All participants had the same duration and intensity
of intervention course. All had individually tailored intervention aims. Several
therapists were involved and there was no indication that outcome was determined
by the individual who administered the treatment. Attendance was consistent, which
may be an indication of sustained motivation. No attempt was made to monitor the
148 Catherine Mackenzie and Anja Lowit

extent to which participants practised or attempted to apply advice given in sessions,

outwith the clinic, nor the social support and encouragement available to them.
These related variables have been identified as being of potential importance in
prognosis in adult neurological disorders (Tompkins et al. 1990) and might have
affected the participant group. All participants had a history of speech and language
therapy intervention. P5, who showed a good response in the research study, had the
smallest amount of previous therapy. Age may affect response to treatment, for a
variety of reasons. However, the three non-responsive cases were not the oldest
participants in the group, and P6, at age 47, was the youngest. The group comprised
participants with variable lesion sites and in both responsive and non-responsive
subsets there were individuals with unilateral and bilateral lesions.
Potential for natural recovery is linked to time since stroke (Royal College of
Physicians 1989). P1, P6 and P9 were all at least 1 year post-initial stroke record at
admission to the study. With only one exception (P7), those who made significant
progress entered between 3 and 6 months post-onset and so began the intervention
phase at between 5 and 8 months. No one who entered within this time scale failed
to progress. It would be imprudent to draw any conclusions from this small data set,
especially given the response of P7, but restricting the participant group to a more
limited time after onset should be considered in future research.
Dysarthria severity should also be examined in relation to treatment response. It
has been suggested (Wertz 1978) that prognosis is directly related to intelligibility, in
that the more compromised is intelligibility, the less favourable is prognosis. For the
communication effectiveness measure, which encompasses not only intelligibility but
also speech naturalness, in each of the three non-responsive cases variance across the
assessment points was nonsignificant. At the start of the intervention period P6 and
P9 were the highest rated of the group, with scores of 4.90 and 5.80 on a seven-point
scale, and may therefore have had less capacity for improvement than some other
participants. These participants also had the highest scores for reading intelligibility.
P1, while not the lowest rated participant as regards communication effectiveness,
with a score of 1.40 had a much more severe involvement. By contrast, P2 who had
the lowest pre-intervention rating, showed improvement not only in this measure, but
in reading and single word intelligibility also. It is unlikely that prognosis in dysarthria
will be dictated by any one variable and while severity may be relevant, it may be
negated as a prognostic indicator by other variables, including stage of recovery (Wertz
1978). Individuals will have a range of contributing factors which interact negatively
and positively, and many more participants than in this study are required to shed
further light on this issue.

Adjustment
The maximization of communication effectiveness is a main target of speech and
language therapy management in acquired dysarthria. For speakers with persisting
communication difficulty, as was the case for the participants in this study, there
may also be a need to make psychological and social adjustment to an altered
communication status. While according to Duffy (2005), behavioural management
includes supportive and counselling roles, this area has not received detailed con-
sideration in the dysarthria management literature. In this investigation, psycho-
social effects were not specifically targeted, through the incorporation of a
Intervention effects in dysarthria following stroke 149

programme such as formal counselling or personal construct. The participants did

talk about their speech and the perceived consequences, and therapists responded to
this in a supportive way, aiming to maximize confidence in speaking situations.
Duffy (2005) considers that overt sensitivity to the dysarthric persons situation may
be facilitative of more direct intervention, although this approach is not subject to
formulaic prescription. This aspect of management is not readily measurable but
we believe that the therapists who participated in this study were encouraging,
empathetic and supportive in attitude: intrinsic traits of many clinicians (Duffy
2005, p. 444).
Therapist sensitivity and support may have contributed to the group positive
change in acceptance of dysarthria, when comparing the start and end of the
intervention period. This section of the DIP covered issues such as emotional
reaction to not being understood, general effects of speech difficulty on life,
sensitivity, and worrying about speech. The one participant whose raw score was not
increased at the end of intervention also showed no improvement in the three
communication assessments. Performance on the conversation and connected
reading measures did not correlate with this section of DIP before intervention, but
did at the end of intervention, at which point higher scores on these two speech
measures were associated with better acceptance of the presence of dysarthria. Some
interaction between severity of dysarthria and psycho-social response might be
anticipated, though it is interesting that this was not present prior to intervention. It
may be speculated that the post-intervention correlations reflect an increased
adjustment in those with stronger communication skills. It must also be
acknowledged that an ongoing natural process of adjustment during the first year
after stroke might be relevant to some of the group. Three sections of the DIP did
not show change for the group and examination of the scores of individual
participants did not indicate any pattern in relation to communication improvement.
As the programme was concerned with the speaker with dysarthria and not with
communication partners, change could not be anticipated in the section concerned
with How I feel others react to my speech. However, the absence of change in The effect of
dysarthria on me as a person and How dysarthria affects my communication with others
suggests that future programmes should more specifically target the psycho-social
effects of dysarthria.

Conclusions and clinical implications

There is very little published research on the outcome of intervention in dysarthria
caused by stroke. As a series of well-controlled single-case treatment studies, this
investigation fulfils Robey and Schultzs (1998) recommendation of a necessary
initial exploratory (Phase 1) stage in the route to addressing efficacy, effectiveness
and efficiency. The results provide some guidance for future research methodology
in this field.
The findings of this investigation show that targeted intervention of two
sessions a week for 8 weeks results in a significant improvement in intelligibility
and/or communication effectiveness for some, but not all, adults with established
dysarthria caused by stroke. This progress is likely to be sustained after the cessation
of treatment. Positive change may also occur in the person with dysarthrias
acceptance of the condition. That such results are achievable with a modest amount
150 Catherine Mackenzie and Anja Lowit

of professional input provides support for the continuing use of a behavioural

approach which addresses the patients predominant symptoms in the management
of dysarthria in stroke. Many unanswered questions remain that might be addressed
in future investigations. Would a longer or more intensive treatment course have led
to speech gains in any of the three apparently non-responsive participants? Were
gains maintained beyond the 2-month post-intervention assessment point? Might
gains simply reflect a placebo effect, in the sense of a positive response to thera-
peutic attention? Would group therapy have been equally or more effective? Would
the DIP gain have occurred without direct speech input? Would more attention to
issues of adjustment have affected the results of other sections of the DIP?
The participants in this study comprised a heterogeneous group in many
respects, including age, time after stroke, intervention targets, dysarthria severity and
lesion location. On the basis of available imaging data, varied, and commonly non-
single, lesion sites were implicated in this group. Future research might aim for a
group with a single lesion site, though our research and clinical experience suggests
this is impractical and the literature indicates such classification may not be
informative. Discrete lesions which might be associated with one diagnostic
category are not typical of the dysarthric stroke population. Also within a given
dysarthria category much individual variation of symptoms is noted and there is now
wide recognition that subtypes exist within the established dysarthria categories,
such as ataxic and flaccid (Duffy and Kent 2001). The features of dysarthria
categories overlap to the extent that even experienced clinicians are minimally
successful in diagnosis (Zyski and Weisiger 1987). Thompson et al. (1997), by
providing evidence of considerable variability amongst stroke subjects diagnosed
with the same type of dysarthria (upper motor neurone), stressed the need for
treatment regimes to be specifically and individually tailored. It is thus highly
unlikely that a single set of treatment goals and intervention strategies would be
applicable to any group of people with dysarthria following stroke and intervention
should continue to be individually planned and tailored.
All participants who began treatment in the first year after initial stroke onset
made progress with intervention, while not improving in the 2-month pre-
intervention period. In view of this finding, a subsequent research phase might
specify less than 1 year from stroke onset as an inclusion criterion, while maintaining
3 months as a minimum recruitment point, consistent with current knowledge about
spontaneous recovery (Royal College of Physicians 1989).
The results suggest that even where communication effectiveness is severely
compromised, progress may occur. However, for the two participants with the
highest pre-intervention scores (P6 and P9) no significant change was apparent. No
definitive explanation for this can be provided. Length of time post-stroke may be a
factor, in that both were at least one year following initial stroke. The measures used
may lack sensitivity, but it may be argued that change which is not evident in
evaluation of communication effectiveness during conversation is not a functionally
useful change. Perhaps these participants who entered the study with relatively high
ratings had achieved their potential. In further preliminary exploration of
intervention effects in post-stroke dysarthria, consideration should be given to
identification of a participant group with a more consistent level of severity.
A number of therapists were involved in this study. Although intervention aims
and principles were specified and approaches were suggested, inevitably there would
be variation in the implementation of the programmes. Results did not appear to be
Intervention effects in dysarthria following stroke 151

linked with individual therapists. However to optimize consistency of approach, in

future investigations, a single therapist might conduct all sessions.
Hochstenbach et al. (1998, p 510) found that stroke has a massive effect on
many cognitive processes, with the most prominent deficits related to slowness of
information processing and attention. In the study of post-stroke aphasia, attention
is being drawn to the potential influence of cognitive status on treatment response,
and Helm-Brooks (2002) hypothesises that this may contribute to variability in
response to intervention. This consideration should be applied to other post-stroke
communication disorders. The inclusion of a cognitive assessment such as the
Cognitive Linguistic Quick Test (Helm-Estabrooks 2001) will facilitate the evalu-
ation of the association between dysarthria intervention results and cognitive status.
In dysarthria stroke research, before intervention efficacy, effectiveness and
efficiency may be assessed, further exploratory investigations are required in which
response to intervention is evaluated at individual as well as at group level. The
results of this (Phase 1) investigation indicate that a positive response may be
observed following a modest number of speech and language therapy sessions. In
keeping with the usual characteristics of Phase 1 outcome research there was no
control group, so the research does not inform on the response of a matched group
of dysarthric stroke patients who received no specialist intervention (but ideally an
equivalent period of attention). The recommendations for Phase 2 research include
the establishment of minimum and optimum treatment dosage protocols with a
more defined population, before the introduction of large samples and external
controls which characterize Phase 3 research (Robey and Schultz 1998). The
reported investigation offers some guidance for progression of research in stroke-
related dysarthria.

Acknowledgements
Research was supported by a grant from Chest, Heart and Stroke, Scotland. The
authors also thank the participants, Thia Begg, who conducted the assessments, the
intervention therapists and referring therapists, who are not named for reason of
participant confidentiality, the student raters (Joanne Avery, Elisabeth Bay, Rachel
Courtney, Ali Falconer, Lesley Garrett, Claire Higgins, Jill Kennedy, Galatia Paxinou,
Alison Scott and Katie Walsh), and John Norrie, Centre for Healthcare Randomised
Trials (CHaRT), Health Services Research Unit, Aberdeen University.

References
BALL, L. J., BEUKELMAN, D. R. and PATTEE, G. L., 2004, Communication effectiveness of individuals with
amyotrophic lateral sclerosis. Journal of Communication Disorders, 37, 197215.
BEUKELMAN, D. R. and YORKSTON, K. M., 1977, A communication system for the severely dysarthric
speaker with an intact language system. Journal of Speech and Hearing Disorders, 42, 265270.
BEUKELMAN, D., YORKSTON, K. M. and DOWDEN, P., 1985, Communication Augmentation: A Casebook of Clinical
Management (Austin, TX: Pro-Ed).
CULP, D. and LADTKOW, M. C., 1992, Locked-in syndrome and augmentative communication. In
K. M. Yorkston (ed.), Augmentative Communication in the Medical Setting (Tucson, AZ:
Communication Skill Builders), pp. 59138.
DOMBOVY, M. L. and BACH-Y-RITA, R., 1988, Clinical observations on recovery from stroke. In
S. G. Waxman (ed.), Advances in Neurology. Vol. 47: Functional Recovery in Neurological Disease (New
York, NY: Raven), pp. 265276.
152 Catherine Mackenzie and Anja Lowit

DUFFY, J. R., 2005, Motor Speech Disorders: Substrates, Differential Diagnosis and Management. 2nd edn (St
Louis, MO: Elsevier Mosby).
DUFFY, J. R. and KENT, R. D., 2001, Darleys contributions to the understanding, differential diagnosis,
and scientific study of the dysarthrias. Aphasiology, 15, 275289.
ENDERBY, P. M., 1983, Frenchay Dysarthria Assessment (San Diego, CA: College-Hill Press).
ENDERBY, P. and EMERSON, J., 1995, Does Speech and Language Therapy Work?: A Review of the Literature
(London: Whurr).
HELM-ESTABROOKS, N., 2001, Cognitive Linguistic Quick Test (San Antonio, TX: Psychological Corporation).
HELM-ESTABROOKS, N., 2002, Cognition and aphasia: a discussion and study. Journal of Communication
Disorders, 35, 171186.
HOCHSTENBACH, J., MULDER, T., VAN LIMBEEK, J., DONDERS, R. and SCHOONDERWALDT, H., 1998, Cognitive
decline following stroke: a comprehensive study of cognitive decline following stroke. Journal of
Clinical and Experimental Neuropsychology, 20, 503517.
HOLLAND, A. L., FROMM, D. S., DERUYTER, F. and STEIN, M., 1996, Treatment efficacy: aphasia. Journal of
Speech and Hearing Research, 39, S27S36.
HOWARD, D., 1986, Beyond randomised controlled trials: the case for effective case studies of the effects
of treatment in aphasia. British Journal of Disorders of Communication, 21, 89102.
HUSTAD, K. C., BEUKELMAN, D. R. and YORKSTON, K. M., 1998, Functional outcome assessment in
dysarthria. Seminars in Speech and Language, 19, 291302.
KENT, R. D., WEISMER, G., KENT, J. F. and ROSENBEK, J. C., 1989, Toward phonetic intelligibility testing in
dysarthria. Journal of Speech and Hearing Disorders, 54, 482499.
LOW, E. L., GRABE, E. and NOLAN, F., 2000, Quantitative characterisations of speech rhythm: syllable-
timing in Singapore English. Language and Speech, 43, 377401.
LOWIT-LEUSCHEL, A. and DOCHERTY, G. J., 2001, Prosodic variation across sampling tasks in normal and
dysarthric speakers. Logopaedics, Phonology, Vocology, 26, 151164.
LUBINSKI, R., 1991, Dysarthria: a breakdown in interpersonal communication. In D. Vogel and
M. R. Cannito (eds), Treating Disordered Speech Control (Austin, TX: Pro-Ed), pp. 153181.
MACkENZIE, C., 1991, An aphasia group intensive efficacy study. British Journal of Disorders of
Communication, 26, 275291.
REILLY, S., 2004a, The move to evidence-based practice within speech pathology. In S. Reilly, J. Douglas
and J. Oates (eds), Evidence Based Practice in Speech Pathology (London: Whurr), pp. 317.
REILLY, S., 2004b, What constitutes evidence? In S. Reilly, J. Douglas and J. Oates (eds), Evidence Based
Practice in Speech Pathology (London: Whurr), pp. 1834.
ROBEY, R. R. and SCHULTZ, M. C., 1998, A model for conducting clinical-outcome research: an adaptation
of the standard protocol for use in aphasiology. Aphasiology, 12, 787810.
ROYAL COLLEGE OF PHYSICIANS (RCP), 1989, Stroke: Towards Better Management (London: RCP).
ROYAL COLLEGE OF SPEECH AND LANGUAGE THERAPISTS (RCSLT), 1996, Communicating Quality 2: Professional
Standards for Speech and Language Therapists (London: RCSLT).
SCHMIDT, R. A., 1991, Motor Learning and Performance: From Principles to Practice (Champaign, IL: Human
Kinetics).
SELLARS, C., HUGHES, T. and LANGHORNE, P., 2001, Speech and Language Therapy for Dysarthria due to Non-
progressive Brain Damage (Cochrane Review). Cochrane Library, Issue 2 (Oxford: Update Software).
SHROUT, P. E. and FLEISS, J. L., 1979, Interclass correlations used in assessing rater reliability. Psychological
Bulletin, 86, 420428.
THOMPSON, E. C., MURDOCH, B. E. and THEODOROS, D. G., 1997, Variability in upper motor neurone-type
dysarthria: an examination of five cases with dysarthria following cerebrovascular accident.
European Journal of Disorders of Communication, 32, 397428.
TOMPKINS, C. A., JACKSON, S. T. and SCHULTZ, R., 1990, On prognostic research in adult neurological
disorders. Journal of Speech and Hearing Research, 33, 398401.
WADE, D. T., 1992, Stroke: rehabilitation and long-term care. Lancet, 339, 791793.
WALSHE, M., 2002, You have no idea what it is like not to be able to talk. Exploring the impact and
experience of acquired neurological dysarthria from the speakers perspective. PhD thesis,
Trinity College, Dublin.
WARLOW, C. P., DENNIS, M. S., VAN GIJN, J., HANKEY, G. J., SANDERCOCK, P. A. G., BAMFORD, J. G. and WARDLAW,
J. (eds), 2000, Stroke: A Practical Guide to Management (Oxford: Blackwell Scientific).
WEISMER, G. and LAURES, J. S., 2002, Direct magnitude estimates of speech intelligibility in dysarthria:
effects of a chosen standard. Journal of Speech, Language and Hearing Research, 45, 421433.
Intervention effects in dysarthria following stroke 153

WERTZ, R. T., 1978, Neuropathologies of speech and language: an introduction to patient management.
In D. F. Johns (ed.), Clinical Management of Neurogenic Communicative Disorders (Boston, MA: Little,
Brown), pp. 1101.
YORKSTON, K. M., 1996, Treatment efficacy: dysarthria. Journal of Speech and Hearing Research, 39, S46S57.
YORKSTON, K. M., BEUKELMAN, D., STRAND, E. and BELL, K. R., 1999, Management of Motor Speech Disorders in
Children and Adults (Austin, TX: Pro-Ed).
YORKSTON, K. M., SPENCER, K., DUFFY, J., BEUKELMAN, D., GOLPER, L. A., MILLER, R., STRAND, E. and SULLIVAN,
M., 2001, Evidence-based practice guidelines for dysarthria: management of velopharyngeal
function. Journal of Medical SpeechLanguage Pathology, 9, 257274.
ZYSKI, B. J. and WEISIGER, B. E., 1987, Identification of dysarthria types based on perceptual analysis.
Journal of Communication Disorders, 20, 367378.

Appendix 1: Motor learning principles

N Instructions before the patients attempt: clear, simple instructions, including
illustration where relevant, and with repeated demonstration.
N High amount of practice.
N questionofandstimuli
Variety elicited in a range of situations (e.g. imitation, reading,
answer).
N Blocked practice in early stages, followed by random practice (mixing the
tasks and stimuli within the session).
N Feedback relating to result (whether correct or not) and also performance
(what specific aspect was correct or incorrect).
N Feedback while there is memory of the attempt (a few seconds after perfor-
mance) with time for processing before progressing to the next stimulus/
attempt.
N Frequent feedback in the early phase of a treatment, with summary feedback
(i.e. after a number of responses) more appropriate later on.
N Therapist initially has main responsibility for monitoring and feedback
progressing to participant appraising performance.
N Goal setting: set a specific level of expected performance (70% accuracy on
two successive attempts at a stimulus set, before progression).
N Periodic maintenance practice following achievement of any goal.