|
Article Excerpt
This systematic review of the literature addresses interventions for dysarthria that focus on the global aspects of speech. The review is part of the development of practice guidelines for the Academy of Neurologic Communication Disorders and Sciences (ANCDS). A search of electronic databases (PsychINFO, MEDLINE, and CINAHL) and hand searches of relevant edited books yielded 51 articles focusing on loudness, rate, prosody, and general instructions. These articles were rated for the strength of evidence they provide for the effectiveness of intervention. Articles were categorized into phases of research and evaluated in terms of the level of participant description, outcome measures, evidence of research control, and findings. The strongest evidence regarding treatment effectiveness is in the area of modification of loudness in individuals with Parkinson's disease who have hypokinetic dysarthria. Directions for future research are provided in the areas of rigor of evidence and its reporting, outcomes, candidacy criteria, and application of principles of motor learning to intervention.
**********
One of the primary goals of speech intervention for speakers with dysarthria is to improve communicative function. At times, this is accomplished by focusing on specific speech subsystems. For example, the management of the velopharyngeal systems is addressed in a previous review (Yorkston et al., 2001b). At other times, the focus of intervention is on the entire process of speech production. Global aspects of speech production are those that span the different levels of speech production: respiration, phonation, resonance, and articulation (Dromey & Ramig, 1998). Global aspects also span the time domain and are applied throughout an utterance rather than occurring during a specific gesture (e.g., enhancing bilabial closure), syllable or sound (e.g., production of precise sibilants). Changes in loudness, speaking rate, and prosody are the targets of such intervention programs. The purpose of this systematic review is to identify the types and strength of evidence documenting the effectiveness of interventions targeting the global aspects of speech production in speakers with dysarthria.
METHODS
Background
This systematic review is part of the development of practice guidelines for management of dysarthria sponsored by the Academy of Neurologic Communication Disorders and Sciences (ANCDS) and supported in part by ASHA (Office of the VP of Clinical Practices in Speech-Language Pathology, and Steering Committee of Division 2) and by the Department of Veterans Affairs (DVA). Systematic reviews can be viewed as a process of evaluation of evidence from both research literature and expert opinion with the goal of assisting clinical decisions. The procedures for developing these reviews are described elsewhere (Yorkston et al., 2001a).
The Searches
We searched the following electronic databases: PsycINFO covering 1,300 journals (1967 to Nov. 2004), MEDLINE covering 3,900 journals (1966 to Nov. 2004), and CINAHL covering 600 journals (1982 to Nov. 2004). The initial searches were keywords paired with the term, dysarthria, for example, dysarthria and rate, dysarthria and loudness, dysarthria and prosody. In addition to these electronic searches, hand searches of relevant edited books in the area of dysarthria and ancestral searches of extant references (e.g., studies cited within an article or chapter) were conducted. From this large search, those citations related to intervention were described, rated, and compiled in a series of Tables of Evidence. Intervention studies were defined as those focusing on treatment of the speech production for at least one person with a primary communication diagnosis of dysarthria. Thus, articles were excluded that:
1. described the global aspects of dysarthric speech but did not treat it,
2. focused on a single speech subsystem or component, for example, those that focused only on production of sustained phonation in speakers with dysarthria,
3. applied treatment approaches to individuals without impairment, for example, examination of rate and loudness changes in speakers without neurologic impairment,
4. studied techniques for management of global aspects of speech associated with disorders other than dysarthria, for example, treatment of rate and loudness in apraxia of speech, or
5. reported the affects of digital manipulation of recorded sample of dysarthric speech.
Intervention studies that focus on respiratory-phonatory or velopharyngeal aspects of speech production are reviewed in other modules (Spencer et al., 2003; Yorkston et al., 2001b).
Rating the Strength of Evidence
The strength of evidence for behavioral intervention studies can be rated by asking a series of questions.
What Type of Research Is Represented by the Study?
We answered this question by identifying the phase of research adapted from the descriptions of Robey and Schultz (1998). During phase I, hypotheses about treatment efficacy are developed for later testing. Often this involves experimental manipulations to test the potential benefits or activity of a particular treatment; for example, asking a speaker with Parkinson's disease (PD) to speak slowly or loudly and then measuring the acoustic, physiologic, or perceptual consequences of that manipulation. In this review, studies in which global aspects of speech were manipulated experimentally were designated as phase I studies. During phase II, the goals are to formulate and standardize protocols, validate measurement instruments, optimize dosage of treatment, and so on. For this review, articles were placed in this category if a treatment protocol was carried out for a speaker with dysarthria. Articles in this category were case reports or small group studies with no control groups or treatment comparisons. During phase III, treatment efficacy of a specified protocol is formally tested either with single participant design research or group studies with controls such as control groups or treatment comparisons.
How Well Are the Participants Described?
We answered this question by noting the presence or absence of 18 participant descriptors similar to those described elsewhere (Strand & Yorkston, 1994). The level of description was categorized as brief if 1-5 characteristics were included in the article, detailed if 6-10 were included, and comprehensive if more than 10 were included. In addition, information about the following participant-related characteristics is provided: number of participants, type of dysarthria, severity of dysarthria, medical diagnosis, age, and gender.
Are the Consequences of the Intervention Well Described?
We answered this question by noting evidence for control, that is, evidence that changes were the result of intervention and not some other variables. The following are examples of factors that suggest control: presence of stable baseline, outcome measures obtained with and without a device, improved speech performance with intervention in the face of a progressive disorder, and presence of a comparison or control group. The types of outcomes measures were also noted in the following categories: acoustic, physiological, perceptual, or psychosocial. See Technical Report 6 for specific outcomes measures in each of these categories. Finally, a summary of the study conclusions is provided.
RESULTS AND DISCUSSION
A total of 51 intervention studies were identified, obtained, and rated by at least two members of the Writing Committee. Articles describing and reporting the effectiveness of treatment of the global aspects of speech in dysarthria have appeared for over 30 years: nine were published before 1985, 10 between 1985 and 1994; and 32 between the years 1995 and 2004. Intervention studies were categorized into four groups, those focusing on loudness (N = 21), speaking rate (N = 19), prosody (N = 10), and general instructions (N = 6). Note that a number of articles were placed in more than one category. For example, if a study had a condition involving rate and a condition involving loudness, the study was placed in both categories. The following sections describe each of the treatment categories: loudness, rate, prosody, and general instructions. Characteristics of these studies are summarized in a series of Tables of Evidence (Tables 1, 3-5). Studies are listed in chronological order of publication.
Loudness
Table of Evidence: Loudness (Table 1) summarizes 21 articles reporting outcomes of treatment focusing on increasing speech loudness. Note that most studies (N = 11: 52%) are phase II studies (preliminary investigation of intervention protocols), and some (N = 7: 33%) are phase III (a specific protocol is formally tested).
Treatment
While increasing loudness may be a common treatment goal for individuals with various dysarthria types, including flaccid, hypokinetic, and mixed, most evidence-based treatments for increasing loudness stem from research in the area of PD (Adams, 1997). Reduced loudness is one of the primary perceptual features of the hypokinetic dysarthria in PD. This type of dysarthria is common because of the high prevalence of PD in the aging population and the high prevalence of dysarthria in PD. Increasing respiratory/phonatory effort, thus increasing loudness, has been the focus of a systematic program line of treatment research conducted by Ramig and colleagues (Ramig, Fox, & Sapir, 2004). The majority of studies in this category focused on Lee Silverman Voice Treatment (LSVT), an intensive, high-effort speech treatment designed to rescale the magnitude of motor output of speakers with PD (Ramig, Pawlas, & Countryman, 1995). Goals are to increase phonatory effort, vocal fold adduction, and respiratory support. LSVT was designed to improve the perceptual characteristics of voice by targeting loudness. This treatment approach is well described and thus can be replicated from clinic to clinic.
Speakers' Characteristics
A total of 308 participants participated in this group of studies. This total may overestimate the number of individuals because some are represented in multiple studies. For example, studies reporting the impact of LSVT on aerodynamics (Ramig, Countryman, Thompson, & Horii, 1996) are reported separately from those reporting acoustic findings (Ramig et al., 1995). Generally, the participations were well described, with 45% of studies providing comprehensive descriptions of participants and 45% detailed descriptions. The majority of participants (76%) described in this series of studies had a diagnosis of PD and exhibited symptoms consistent with hypokinetic dysarthria, including phonatory hypoadduction and decreased vocal loudness.
A profile of candidacy requirements emerges from a review of the literature. This candidacy profile is based on a constellation of symptoms (reduced loudness, poor respiratory support/effort), good stimulability such that performance improved with appropriate cues, and a high level of motivation to participate actively in an intensive program of practice. Ramig and colleagues (2004) suggest that individuals with mild-moderate PD have the most positive outcomes and thus are the best candidates for LSVT. Preliminary data are reported for speakers with...
|
