Intervention for dysarthria associated with acquired brain injury in children and adolescents
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To assess the efficacy of intervention delivered by Speech and Language Pathologists/Therapists targeting dysarthric speech in children resulting from acquired brain injury.
Background
The term 'Acquired Brain Injury' (ABI) encompasses a wide range of aetiologies within the paediatric population with neurological injury, including diagnoses of cerebrovascular accident and brain tumour. The largest population however, is commonly children who have sustained a traumatic brain injury such as that incurred in a fall or motor vehicle accident (Parslow 2005). The incidence of paediatric acquired brain injury is significant, with reports of up to two hundred and eighty per 100, 000 children being admitted to hospital with traumatic brain injury in the UK alone each year (Hawley 2003).
The high incidence of acquired brain injury in children is of major public concern, as it is the most frequent cause of acquired disability in children, leaving a large proportion of those who survive with multiple long‐term impairments (Tennant 1995; Michaud 1993). One chronic impairment commonly associated with ABI is a form of speech disorder called dysarthria (Cahill 2002; Cornwell 2003a; Cornwell 2003b).
Dysarthria is a motor speech disorder that may affect the range, rate, strength and co‐ordination of the muscles used for speech. This problem may affect multiple subsystems required for accurate speech production including: articulation, resonance, respiration and phonation (Van Mourik 1997). Dysarthria can significantly affect the intelligibility of speech, resulting in the child with dysarthria often being misunderstood and experiencing communication breakdown. This communication breakdown may have negative affects on a child's education, socialisation, potential for later employment, and hence participation in and contribution to society.
Intelligibility is defined as the degree to which a speaker can be understood by the listener (Yorkston 1996), and ratings of intelligibility therefore tell us how a speech disorder impacts upon the ability of a given person to communicate (DeBodt 2002).
A review of studies reporting the clinical presentation of childhood acquired dysarthria, concluded that there were no definite similarities between adult and paediatric clinical presentations of dysarthria in the literature published during 1980‐1997 (Van Mourik 1997). The authors concluded that paediatric dysarthria requires its own classification system (Van Mourik 1997). These findings are not surprising given that dysarthria in children is typically complicated by a variety of developmental issues. These issues cannot be underestimated, as paediatric dysarthria occurs in a context of brain maturation, rapid physical growth, and cognitive and psychosocial development. In addition, children are still acquiring their sound production system, and the oral musculature and quality of oral movements change during development (Qvarnstrom 1994). Thus the features of childhood dysarthria following brain injury, and the potential benefit of particular speech treatments cannot be extrapolated from the adult literature.
There is currently a push for evidence‐based practice and accountability of practice in regard to all aspects of health treatment. Despite this fact, there is very little evidence base for the treatment of paediatric dysarthria following ABI. Existing techniques are reported for use with the adult population and are largely based on traditional perceptual models of speaker‐based treatment techniques including drills to improve stress or intonation based on improving breath capacity or altering breath pattern (Bellaire 1986; Rosenbek 1991; Yorkston 1988; Yorkston 1999), treatment designed to alter rate control (Beukelman 1978; Crow 1989; Helm 1979) or adjust speech movement patterns to help produce a perceptibly more intelligible speech outcome (Yorkston 1988; Yorkston 1999), and programs aiming to increase vocal intensity such as the Lee Silverman Voice Treatment (Ramig 1995).
Traditional speech therapy based in the community remains largely subjective and non‐quantitative, i.e. a clinician will train a child to improve their speech sounds as rated using auditory‐based, or perceptual outcomes of speech (Dagenais 1995; Secord 1989). The perceptually‐based systems such as those listed above have merit in that they are based on representing the actual speech output, which is what the listener attends to (Wood 1999). However, there are many problems with over‐reliance on subjective perceptual approaches for measuring and treating disordered speech. One example of issues with perceptual approaches is the use of diagnosis using traditional phonetic transcription systems highlighted by Wood and Hardcastle (Wood 1999). There is a tendency for clinicians to favour 'categorical' errors (e.g. sound substitutions like saying 'pin' when the person meant to say 'bin'), over 'non‐categorical' errors (e.g. when sounds are actually distorted so /d/ may be produced in a prolonged and imprecise manner) (Ziegler 1989). Children with dysarthria most commonly experience sound distortions, the errors that are most often poorly transcribed by clinicians, demonstrating the unreliability of traditional auditory‐based techniques for treating this population. If a clinician trained in listening to speech is unable to detect subtle differences such as those heard in dysarthria, then the use of such perceptually based techniques bodes poorly for the client who would also inevitably not be able to correctly hear and define and rectify their own disorder based on auditory feedback alone.
There has been an attempt to make the treatment, and the outcome measures for rating adult treatment for dysarthria, objective and quantifiable by using instrumental techniques such as electropalatography (EPG) (Hardcastle 1985), or the VisiPitch© or Speech Viewer © (Bougle 1995). EPG for example, requires the participant to wear a custom‐made acrylic mould of the individual's hard palate that is embedded with 62 sensor electrodes. The electrodes record tongue contact with the palate during speech. This technique is unique because it provides a dynamic visual presentation of lingual movements that could not previously be seen by clinicians or individuals with speech impairment (Hardcastle 1991; Hardcastle 1985). The additional visual feedback provided by EPG during speech therapy may be particularly beneficial in treating patient's following brain injury because these patients often have difficulties with understanding verbal or auditory instructions, which are traditionally used by clinicians in remediation programs (Chapman 1997; Yorkston 1997). Visual biofeedback techniques provide a more tangible or concrete example of the patient's speech pattern for remediation avoiding the need for complex instructions. Whilst both the traditional perceptually‐based and instrumentally based biofeedback techniques have been advocated for the remediation of speech disorders due to brain damage in adults, there have been few investigations of treatment efficacy for children with speech disorders due to brain injury, and we cannot assume that the same principles and practices can be successfully applied to the paediatric population. It is also not known how age, severity of brain injury, site of brain lesion not severity of dysarthria impact upon treatment success.
Summary
Dysarthria is a common and often persistent sequelae of traumatic brain injury which may lead to deleterious consequences in regard to participation in society and quality of life. To date there has been no investigation of the effectiveness of perceptual and instrumental treatments for dysarthria in the paediatric population with acquired brain injury. Given the variety of developmental stressors such as neurological maturation, and maturation of the subsystems of speech production (e.g. lips, tongue, etc), we cannot extrapolate the findings of adult investigations in this area to the paediatric population. Therefore, it is the aim of the current paper to systematically review evidence on the efficacy treatment for dysarthria in the paediatric population with acquired brain injury through the evaluation of quasi‐randomised and randomised controlled trials.
Objectives
To assess the efficacy of intervention delivered by Speech and Language Pathologists/Therapists targeting dysarthric speech in children resulting from acquired brain injury.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (RCT) studies and quasi‐experimental design studies (e.g., studies in which participants were allocated on alternate days) will be considered for inclusion in this review.
Types of participants
Children aged 3‐16 years with acquired dysarthria. Participants will be grouped together by aetiology, (e.g., posterior fossa tumour, other tumours/gliomas, TBI post‐fall, TBI post‐RTA etc). Paediatric populations with co‐morbid conditions such as cognitive or language impairment will be included in the review.
Types of interventions
1) Perceptually‐based therapy: Intervention using traditional drill exercises without instrumentation in the absence of feedback other than auditory feedback, and typically in the absence of using any instrumentation e.g., exercises of the lips or tongue to increase the rate, strength, range or co‐ordination of the musculature supporting articulation; drill breathing exercises to increase respiratory/breath support for speech; voicing drills to increase the loudness of phonation, etc.
2) Instrumentally‐based biofeedback approaches: Interventions that use some form of instrumentation and that provide visual or other forms of biofeedback in addition to auditory feedback (e.g., electropalatography; kinematics; visual biofeedback acoustic treatment).
Control groups may include no treatment or wait‐list control.
Types of outcome measures
Three levels of outcomes will be considered for analysis used to measure change (Law, Garrett & Nye, 2003) in articulation, phonation, resonance, prosody and respiration or a combination of any of these areas typically affected by dysarthria:
1. At the isolated function level of speech production, e.g. a reduction in single phoneme or sound production duration during articulation.
2. At the single word or sentence level of general speech production, e.g. improvement in rate of single word reading.
3. At the broader level of speech production where outcomes demonstrate functional gains in communication, e.g. improvement in timing of spontaneous speech; reported improvement in speech rate via child or parent report of speech function.
Both standardised and informal outcome measures will be evaluated, including clinician, parent or child questionnaire reports on outcome. Outcome measures for all features of dysarthria (i.e. nasality; articulation; laryngeal function; respiratory function) will be considered.
Search methods for identification of studies
The following databases will be searched from inception:
Cochrane Central Register of Controlled Trials (CENTRAL), published in The Cochrane Library
MEDLINE
CINAHL
Dissertation Abstracts
EMBASE
ERIC
Linguistics and Language Behaviour Abstracts (LLBA)
PsycINFO
The search strategy below will be used for MEDLINE:
1 Dysarthria/
2 Speech Disorders/
3 ataxia.tw.
4 (dysarth$ or dysphon$ or anarth$ or dyspros$ or aphon$ or dyston$).tw.
5 ((speech or articulat$ or voice or vocal or communicat$) adj3 (disorder$ or impair$ or problem$ or difficult$)).tw.
6 ((phonat$ or prosod$ or intonat$ or respirat$) adj3 (disorder$ or impair$ or problem$ or difficult$)).tw.
7 or/1‐6
8 Child/
9 Adolescent/
10 (child$ or girl$ or boy$ or pre school$ or preschool$ or adolescen$ or teen$).tw. (718291)
11 or/8‐10
12 (rehabilitat$ or therap$ or train$ or management or assist$ or measure$ or assess$ or remedia$ or augment$ or recover$).tw.
13 technique$.tw.
14 12 or 13
15 7 and 11 and 14
16 randomized controlled trial.pt.
17 controlled clinical trial.pt.
18 randomized controlled trials.sh.
19 random allocation.sh.
20 double blind method.sh.
21 single‐blind method.sh.
22 or/16‐21 (394278)
23 (animals not human).sh.
24 22 not 23
25 clinical trial.pt.
26 exp Clinical Trials/
27 (clin$ adj25 trial$).ti,ab.
28 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or mask$)).ti,ab.
29 placebos.sh.
30 placebo$.ti,ab.
31 random$.ti,ab.
32 research design.sh.
33 or/25‐32
34 33 not 23
35 34 not 24
36 comparative study.sh.
37 exp Evaluation Studies/
38 follow up studies.sh.
39 prospective studies.sh.
40 (control$ or prospectiv$ or volunteer$).ti,ab.
41 or/36‐40
42 41 not 23
43 42 not (24 or 35)
44 24 or 35 or 43
45 44 and 15
Terms and filters will be modified as appropriate for other databases.
We will request information on unpublished trials from authors of published studies, and experts and information groups in the areas of linguistics and speech therapy.
Data collection and analysis
Titles and Selection of trials
Titles and abstracts will be independently screened for inclusion by both authors (AV) and (AM). In cases of uncertainly over whether an abstract meets the inclusion criterion by either author, the full text article will be obtained. Each paper will then be evaluated independently by the two reviewers (AV) and (AM) for inclusion. In the event of disagreement over inclusion of a particular paper, a consensus will be formed by AV and AM re‐assessing the inclusion criterion together.
Quality Assessment
Included trials will be evaluated against a number of criteria Both reviewers will assess and independently rate the methodological quality of the included studies for the aspects outlined below, using the following three quality codes as described in the Cochrane Collaboration Handbook (Higgins 2005):
(A) Adequate
(B) Component not reported or unclear
(C) Component reported but inadequate
-
Method of allocation: (e.g., Adequate: use of pre‐numbered or coded identical containers administered serially to participants; Unclear: no description of allocation methods; Inadequate: Poor or nil concealment of allocation, e.g., alternate assignment).
-
Blinded to treatment allocation: Were the assessors blind to treatment allocation? (e.g., Adequate: clinician, patient and outcome assessor blinded to allocated treatment group of patient; unclear: blinding of allocation not reported or cannot be verified by contacting investigators/investigators cannot be contacted; Inadequate: clinician, patient or outcome assessor not blinded to allocation group).
-
Proportion of losses to follow up (e.g., Adequate: losses to follow up equally distributed between comparison groups; Unclear: losses to follow up not reported; Inadequate: studies will be excluded where dropout is disproportionate to the remaining study population). Missing data will initially be sought via contact with the corresponding author. In regard to participant drop out, if the rate of attrition reaches a 30% threshold in an included study, the study will be included in the systematic review but not in the meta‐analysis . The maximum allowed difference in the dropout rate between the two groups will be 10% before a study included in the review is excluded from meta‐analysis.
-
Intention‐to‐treat analysis (e.g., Adequate: where all trial participants were included in the final analysis, and analysed in the groups to which they were randomised regardless of : i) treatment type received, and ii) other protocol irregularities; Unclear: Intention‐to‐treat analysis not reported; Inadequate: Not all trial participants included in the final analysis due to protocol irregularities.
Studies rated as (C) will be analysed separately in sensitivity analyses.
Data management
In addition to outcome data, the following information will be documented by both reviewers using a data management form to be developed and piloted: participant details; setting (e.g., community clinic, school, hospital); type of intervention; length and frequency of intervention; professions involved; duration of impairment; level of severity; co‐morbidity; assessment tools employed. Any information that is missing or unclear will be requested from the corresponding author. Outcome data will be independently entered on to RevMan by (AV) and (AM) and a re‐evaluation of the data and entries performed together by AV and AM to reach consensus on points of disagreement.
Missing data
In the event of missing data within published studies, authors AM and AV will contact primary investigators for assistance and information.
Data synthesis
Meta‐analysis will be only performed where studies employ similar interventions and where study populations are clinically homogenous.
Continuous data
To enable the combination of studies measuring the same outcome using different methods, continuous data will be summarised using standardised mean differences.
Binary data
Binary outcomes are likely to be common in early reports within the field (e.g., improved outcome vs no change/worse). Data will be analysed by calculation of the relative risk with a 95% confidence interval.
Subgroup analysis:
If possible, subgroup analysis will be undertaken by age group, site of brain lesion, severity of brain injury (measured via the Glasgow Coma Scale), the severity of dysarthria
Sensitivity a nalysis
Sensitivity analysis will be used to assess the robustness of the overall findings by examining the impact of study quality, e.g. lack of allocation concealment or high rates of loss to follow‐up, the impact of missing data or the impact of imputations, and the rigor of eligibility criteria employed in the study.
We will also evaluate the possibility of one or more large studies dominating the results.
Investigation of heterogeneity
Consistency of results will be assessed by examining I2 (Higgins 2002). I2 is a quantity describing approximately the proportion of variation in point estimates that is due to heterogeneity of a sample rather than error in sampling of the population. A test of homogeneity will be used to determine that the heterogeneity is genuine. In the event of too few studies being available to make this test feasible, a random effects model will be applied.
Assessment of bias
Where appropriate, the possibility that the study selection was affected by bias will be assessed using funnel plots to investigate any relationship between effect size and study precision (closely related to sample size). Such a relationship could be due to publication or related biases or due to systematic differences between small and large studies, or a statistical artefact of the chosen effect measure.
