[Research]
Music and Stroke Rehabilitation: A Narrative Synthesis of the Music-Based Treatments used to Rehabilitate Disorders of Speech and Language following Left-Hemispheric Stroke
By Kevin Draper
Abstract
Stroke is a leading cause of long-term disability. A stroke can damage areas of the brain associated with communication, resulting in speech and language disorders. Such disorders are frequently acquired impairments from left-hemispheric stroke. Music-based treatments have been implemented, and researched in practice, for the past thirty years; however, the number of published reports reviewing these treatments is limited. This paper uses the four elements of the narrative synthesis framework to investigate, scrutinise and synthesise music-based treatments used in the rehabilitation of patients with speech and language disorders. A systematic review revealed that fifteen studies meet the inclusion criteria set out. It was found that the music-based treatments utilised included: Melodic Intonation Therapy (MIT), Modified Melodic Intonation Therapy (MMIT), adapted forms of MIT, the Singing Intonation, Prosody, breathing (German: Atmung), Rhythm and Improvisation (SIPARI) method and a variety of methods using singing and songs. From a synthesis of the data, three themes emerged which were key elements of the interventions; they were: (a) singing songs and vocal exercises, (b) stimulating the right hemisphere and (c) use of speech prosody. These themes are discussed and implications for newly-qualified practitioners are explored.
Keywords: left-hemispheric stroke; speech and language disorders; music therapy; music and stroke; aphasia.
Introduction
Every year around 10,000 people in Ireland have a stroke, with approximately 2000 resulting in death. It is estimated that 30,000 people in Ireland are living with disabilities caused by a stroke or a series of strokes (Irish Heart Foundation, n.d., para. 1). The most common are varying severities of left- and right-sided hemiparesis and acquired speech and language disorders such as aphasia, apraxia of speech, and dysarthria. Consequently, professional therapeutic services are continuously required in rehabilitation programmes designed for stroke survivors. There is an ever-growing body of literature investigating the use of music-based methods in rehabilitation following stroke in the areas of speech and language disorders, physical rehabilitation, emotional and social support, and memory loss. For many years, research has been undertaken to examine the efficacy and validity of these treatments in the rehabilitation process. However, the methods and interventions adopted need to be assessed in a transparent manner and arranged systematically. This paper seeks to investigate, review and synthesise the music-based treatments used to rehabilitate left-hemispheric stroke survivors with speech and language disorders.
Positioning of the Researcher: A Case Study
My interest in the area of speech and language rehabilitation following stroke began in 2013 during college work-experience placement at a rehabilitation hospital in the east of Ireland. During this time, a patient was referred to me by a speech and language therapist. John [pseudonym] was a fifty-five-year-old man who had recently suffered a left-hemispheric cerebral vascular accident (CVA). He was diagnosed with global aphasia and apraxia of speech. Post stroke he was now a wheelchair user and had mild paresis in his upper limbs, but his prognosis for physical abilities was optimistic. His most concerning post stroke impairment involved his verbal communication skills.
John attended a total of three with me. As a student, with limited experience working with people with speech and language disorders, I had little prior knowledge of the potential benefits which music-based treatments could offer him. However, basic patient information gathered during the referral process informed me that he enjoyed singing along to the radio but could not articulate the words of the songs. While adopting a client-led approach, the music-therapy sessions usually involved song singing where I would play a song on the guitar and sing, and John would vocalise along with me. He did not have the ability to pronounce the lyrics but vocalised the melodies using the vowel sounds “a” and “ou”, and also a “mi” sound. A particularly interesting moment in the second session occurred when John spontaneously began to vocalise the melody of a song without me accompanying him. He sang two verses of an Irish folk song; then he stopped, smiled and nodded to me. I felt that this moment of spontaneity was especially striking because he sang the melody in his own way and in such a manner that I was able to accurately identify the song. Articulating this melody to an extent where it was recognisable required rhythm, intonation, prosody, pitch, and oral muscle control.
Due to his injuries John had lost the ability to form a two-word intelligible phrase. Nonetheless, in music-therapy sessions he demonstrated an ability to vocalise melodies to an extent where the specific songs were clearly recognisable. However brief my time with John, the experience left me feeling very curious with many questions. Most importantly, I wanted to know how he was able to vocalise melodies which required various vocal and oral skills and abilities while simultaneously having a significantly impaired ability to communicate using speech. I wanted to know what was happening neurologically in the relationship between music and language, and between singing and speaking. I wondered if a long-term music-based treatment could have benefitted his rehabilitation. I considered how musical methods could be used when working with John, and also I speculated if I, as a student, would have had the skills and time to implement them.
Left-Hemispheric Stroke
Baker and Tamplin (2006) explained that a cerebral vascular accident or CVA occurs when a part of the brain is deprived of blood flow and subsequently dies. Neurological damage can occur as a result of two types of CVA. Ischaemic stroke, which is the most frequently diagnosed type, occurs when there is a blockage in an artery supplying blood to the brain (Safranek, 2011). Haemorrhagic stroke occurs when sudden high blood pressure causes arteries within the brain to rupture (Wong, 2004). The resulting effects of stroke vary by its type and location in the brain. Following a left- or right-sided CVA, there will often be residual damage to one of the main hemispheres in the brain. Discussing left-sided CVAs specifically, Baker and Tamplin (2006) and Safranek (2011) described how it affects the right side of the body causing hemiplegia or hemiparesis and sensorimotor difficulties such as poor balance, gait and decreased speed of movement. Difficulties can also materialise in memory problems, as well as behavioural and personality changes. In the context of the current study it is important to note that left-sided CVAs can result in speech and language disorders including aphasia, dysarthria and apraxia of speech.
Speech and Language Disorders
Left-sided CVAs commonly cause three types of speech and language disorders; these are dysarthria and apraxia of speech, and aphasia. This section is dedicated to briefly discussing these three disorders.
Aphasia is defined as an acquired language disorder caused by some form of damage to the brain, the most frequent being stroke (Hallowell & Chapey, 2008). Wong (2004) described aphasia as damage to the part of the brain that controls language. It is characterised by an impairment of the language modalities of speaking, listening, reading, and writing. Aphasia is the most frequently diagnosed communication disorder following a left-hemispheric stroke (Baker & Tamplin, 2006). According to the National Stroke Association of America, there are 800,000 diagnosed strokes each year in the USA with 25–35% causing some form of aphasia (as cited in Conklyn, Novak, Boissy, Bethoux, & Chemali, 2012). Schlaug, Norton, Marchina, Zipse, & Wan, (2010) and Conklyn et al. (2012) reported corresponding statistics for the USA regarding the development of aphasia following stroke. Similarly from a European perspective, Jungblut (2005) stated that 30% of stroke survivors in Germany acquire aphasia.
There are several different types of aphasia which cause impairment ranging from mild to severe. These include fluent aphasia, non-fluent aphasia, anomic aphasia, global aphasia and primary progressive aphasia. The literature sourced for the current synthesis revealed that 96% of the left-hemispheric stroke survivors who participated in the studies had non-fluent aphasia (see Appendix A and Appendix B); this is also known as Broca’s or expressive aphasia. It is characterised primarily by a deficiency in language formulation and production (Hallowell & Chapey, 2008). Speech can be slow with long pauses between words; the individual may have difficulty in naming objects and may demonstrate poor sentence construction and disturbed prosody (Baker & Tamplin, 2006). However, auditory and reading comprehension can remain intact.
Dysarthria – defined as a disorder of movement or movement control of oral communication – is a motor speech disorder which can occur following a stroke. It refers to the weakness of muscles responsible for oral, laryngeal and respiratory control (Duffy, 1995). Baker and Tamplin (2006) described dysarthria as poor articulation and vocal control, while Wong (2004) stated that tongue control may be impaired affecting a person’s ability to speak clearly, chew food and clear food from the mouth, as well as affecting voice consistency. Two common types of dysarthria include flaccid dysarthria where the oral muscles become weak and tend to droop, and spastic dysarthria where the oral muscles become tight and vocal sounds have to be forced and tense (Wong, 2004).
Duffy (1995) defined apraxia of speech as a neurological speech disorder which is distinguished by an impaired ability to plan or programme conscious speech commands into oral motor movements which are necessary for speech production. It can also be diagnosed as oral apraxia, but in all cases it presents as a difficulty or inability to coordinate movements of the tongue, lips or vocal cords (Wong, 2004).
Methodology
Narrative Synthesis
Narrative synthesis is a widely-used, systematic reviewing method (McDermott, Crellin, Ridder & Orrell, 2013). Popay, Roberts, Sowden, Petticrew, Arai, Rodgers, Britten, Roen, & Duffy, (2006) devised a narrative synthesis guide in order to make the approach more systematic and to minimise bias. It defined narrative synthesis as an approach to systematic reviews and the collection of findings that primarily relies on the use of words and text in summarising and describing the findings. Findings can also be presented using visual representations such as graphs, statistics, tabulations, conceptual maps, etc. McDermott et al. (2013) demonstrated that the key to successful narrative synthesis is to review what worked but also to investigate how and why an intervention might have worked. It is suggested that this research method is particularly suited to music-therapy literature (ibid.).
Narrative synthesis involves four stages: (a) theory development, (b) preliminary synthesis of findings, (c) exploration of relationships between studies, and (d) assessing the robustness of the synthesis. The current study proposes to utilise these four interactive elements in order to investigate, scrutinise, and synthesise the existing music-based methods used to rehabilitate individuals with disorders of speech and language.
Narrative Synthesis Stage One: Theory Development
The first stage of the narrative synthesis approach involved gathering background information in order to understand the theory behind the specific area of research. After the relevant information had been identified it enabled me to theorise how and why certain interventions work. By working through this process, I was able to determine the direction I was coming from as a researcher and acknowledge my own beliefs; in turn it enabled me to recognise my own biases. Formulating a personal theory allows a researcher to identify questions and themes that are important to their own research, as well as possible implications for practice. My theory development involved considering the link between music and language from birth, and also how this may impact or influence a person with communication impairments following stroke.
Link between music and language.
The link between music and language has been established in previous research and is widely accepted (Hartley, Turry, & Raghaven, 2010; Hurkmans, de Bruijn, Boonstra, Jonkers, Bastiaanse, Arendxen, & Reinders-Messelink, 2012). Infants are born with an innate ability to imitate, respond to, match, and vocalise maternal singing and speech (Briggs, 1991; Hargreaves, 1996). Chen-Hafteck (1997) provided evidence on the link between music and language in early child development. Crystal (1987) stated that at around 8–20 months an infant displays the first signs of gross oral activities required for speech development (as citied in Chen-Hafteck, 1997). As the infant grows older, their vocalisations – described as cooing – are continually becoming more coordinated. This is indicative of their growing vocal control, which is essentially a musical ability. The authors proposed that these learned abilities by infants are prelinguistic or premusical, which reinforces the close relationship between music and language. At this point, it must be considered that musical ability is possibly ingrained in human nature from birth.
There is also a strong link between music and language in people with speech and language impairments. Disorders such as aphasia occur because of damage to the areas of the brain necessary for communication. However, Brust (1980; 2001) outlined that for more than 200 years, it has been reported that patients who have severely-impaired speech abilities retain the ability to produce well-articulated and accurate words when singing (as cited in Wilson, Parsons, & Reutens, 2006).
The infant and the stroke patient.
As discussed above, the literature provides strong evidence for the link between music and language in infancy and a relationship between singing and speaking following brain injury. It has been suggested that infants have musical abilities (Briggs, 1991; Chen-Hafteck, 1997; Hargreaves, 1996), but they do not have the linguistic skills necessary for verbal communication until 1 or 2-years old. They rely on body language and preverbal methods to communicate their needs. Thus, in many ways, people with acquired speech and language disorders communicate on the same level as an infant. They may have lost several communication abilities and may also rely on non-verbal communication. Therefore, with this evidence in mind, the capabilities of a person with a speech or language disorder resulting from a stroke are, somewhat, comparable to the musical and communication abilities of an infant.
Musical ability following stroke.
In relation to musical abilities, it has been observed that some stroke patients with aphasia have intact musical skills, in particular their unimpaired ability to sing. This observation led to the development of melodic intonation therapy or MIT (Albert et al., 1973). Evidence suggests that music-based treatments hold specific benefits for the recovery of speech and language following stroke. In more recent years, other musical methods have been developed, such as modified melodic intonation therapy or MMIT (Baker, 2000) and the SIPARI method: Singing, Intonation, Prosody, Atmung (breathing), Rhythm, Improvisation (Jungblut and Aldridge, 2004). It is very much an expanding body of literature and research because the musical protocols being devised are still being evaluated and their effectiveness, in the rehabilitation of stroke patients, is being continually evaluated. Despite the existing wealth of literature on the topic there are few studies to date which provide in-depth analysis, exploration and synthesis of these musical intervention techniques.
Data Search
Search strategy.
For this synthesis, the literature search was greatly influenced by search methods utilised by Woodward (2012). The current search used a step-wise approach that involved the following elements: searching databases, scrolling through citations, deep journal searching, and researching outside sources. A table was developed in order to organise and guide the search strategy (see Table 1). Electronic searches were conducted primarily on AMED, MEDLINE, PsycINFO, CINAHL, Cochrane Library, SAGE, Science Direct, and Google Scholar. Relevant journals were identified and searched; these included Journal of Music Therapy, Nordic Journal of Music Therapy, and Music and Medicine. Three key subject areas formed the scope of the search: music, speech and language disorders, and stroke. Within these areas, subcategories of keywords were created to identify relevant studies. To ensure that all relevant areas were covered, search terms were used in combination and these included the keywords within the subcategories. When conducting the search, it was also necessary to develop a list of inclusion and exclusion criteria (see below) which could be used to determine which studies were relevant to this particular synthesis.
| Music | Speech and Language Disorders | Stroke |
|---|---|---|
| Music therapy | Aphasia | Left-hemispheric stroke |
| Musical methods | Dysarthria | |
| Music interventions | Apraxia | |
| Speech disorder | ||
| Language disorder |
Inclusion criteria:
- Studies in the English language published in peer-reviewed journals or book chapters between 2003 and 2013;
- Studies exploring the use of music-based treatments as interventions;
- Studies focusing on the rehabilitation of patients with speech and language disorders;
- Studies that included left-hemispheric stroke patients;
- Studies that explored the use of music as a key method in addressing speech and language disorders.
Exclusion criteria:
- Studies that were published pre2003;
- Studies that included right-hemispheric stroke patients.
An issue that arose during the data search was the difficulty in sourcing relevant literature pertaining to the key subject areas of this study. A search for “music” and “stroke” identified an abundance of articles across the databases. Furthermore, specific searches for “music therapy”, “stroke” and “aphasia” revealed over 100 studies across databases, however, many were not relevant to this synthesis. When the combined keywords were searched, and when the inclusion and exclusion criteria were subsequently applied, a significant number of studies had to be excluded revealing a relatively small number of applicable studies (see below). The difficulty in sourcing literature was probably due to the very specific focus of the current synthesis.
Results
Narrative Synthesis Stage Two: Developing a Preliminary Synthesis
Description of search results and data.
The data search revealed over 20 relevant studies that were then scrutinised under the inclusion and exclusion criteria. The participant information sections indicated that several of these studies focused on right-hemispheric stroke patients and, therefore, were excluded. Other studies were conducted prior to the year 2003 and were excluded on this basis. Some of the studies did not focus on the rehabilitation of individuals with speech and language disorders, but were instead concerned with the link between emotional difficulties and communication impairment following stroke; these studies were also excluded. A total of 15 studies met the inclusion criteria. Fourteen of these were sourced from peer-reviewed journals and one study formed a chapter in a book.
The synthesis process.
The synthesis process was broken into two steps. The first step was a preliminary synthesis which involved the creation of a literature matrix (see Appendix A). The purpose of the matrix was to organise and summarise each of the 15 studies, and combine all of them together in one table. The matrix made the, somewhat, overwhelming task of extracting relevant information from each study more manageable. The matrix was made up of several columns of basic information including methodology, number of participants, focus of the study, and musical interventions. It also presented a brief narrative summary of each study that was influenced by a methodology developed by Lucas, Baird, Law, & Roberts, (2007) that provided guidance on textual study commentaries for narrative synthesis reviews. The summaries conveniently supply the principal points of each study making it unnecessary for the reader to read, or scroll through, each article in entirety. The literature matrix was also a starting point which would facilitate further synthesis of the data during the course of this research.
Following the narrative synthesis approach, the second step of the synthesis process involved the creation of an extensive and in-depth table in which all 15 studies were synthesised (see Appendix B). The purpose of this table was to provide more comprehensive details of each study than the literature matrix (Appendix A) had produced, and to begin to identify specific information, characteristics and patterns across the studies which were deemed potentially important to the current study.
Overview of the Literature
By using both the literature matrix (Appendix A) and the detailed synthesis table (Appendix B) a full overview of the studies was obtained. Firstly, several characteristics were identified as particularly important starting points when analysing the data. The following sub-sections outline the preliminary stages of data analysis and also attempt to answer some important research questions that arose during this process.
Study methodologies.
As stated above, in total there were 15 studies included in the synthesis. Of these 15, 12 studies utilised quantitative research methods while the other three utilised qualitative descriptions and evaluations of interventions and outcomes. Further breaking down the methodologies used in each of the 15 studies, we see that five studies used a randomised controlled trial (RCT) approach (Conklyn et al., 2012; Jungblut, 2005; Jungblut, Suchanek, & Gerhard, 2009; Lim, Kim, Lee, Yoo, Hwang, Kim, & Kim, 2012; Straube, Schulz, Geipel, Mentzel, & Miltner, 2008). Three of the RCTs involved between 17 and 30 participants, while of the remaining two RCTs one involved five participants and the other only two. Of the 10 studies that did not use RCTs, four studies used a multiple-participant case study design each involving between six and eight participants (Kim & Tomaino, 2008; Racette, Bard, & Peretz, 2006; Schlaug, Marchina, & Norton, 2009; Vines, Norton, & Schlaug, 2011). The remaining six studies were conducted as single-participant case studies.
Musical interventions.
The initial and most basic question that this research sought to answer involved identifying the musical interventions utilised in the 15 studies. It was found that a variety of musical interventions were utilised and explored. Of the 15 studies:
- five studies investigated the efficacy of MIT;
- five studies incorporated a variety of methods using songs and singing;
- two studies used adapted versions of MIT;
- two studies explored the SIPARI method;
- and one study investigated MMIT.
Speech and language disorders.
During the early stages of this synthesis, it was also important to identify, within each of the 15 studies, the types of speech and language disorders concerned. This would determine the focus of the literature and who could benefit from these music-based treatments. In total there were 108 study-subjects involved across all 15 studies. Ninety-nine of these subjects had suffered various types of left-hemispheric strokes. Five other participants had damage of the right cerebral hemisphere and four more participants were healthy adults recruited as control subjects in one study (Straube, Schulz, Geipel, Mentzel, & Miltner, 2008).
Of the 99 diagnosed with left-hemispheric stroke, 96 of those were diagnosed with non-fluent Broca’s-type aphasia. Three participants were diagnosed with global aphasia. The 99 left-hemispheric stroke participants also displayed signs of other speech and language disorders. Six participants had non-fluent aphasia, dysarthria and apraxia; four participants had aphasia and apraxia of speech; and two participants had non-fluent aphasia and dysarthria
Discussion
Narrative Synthesis Stage Three: Exploring the Relationship Among Studies
Purpose of this stage.
When conducting stage three of the narrative synthesis the relationships within, and among, the 15 studies were explored; the synthesis table (Appendix B) proved a useful aid in this process. This stage also considered the overall purpose of the current study. Based on my position, at that particular time in my career, it was acknowledged that the information collected should have relevant implications for newly-qualified practitioners. Therefore, during detailed examinations of the 15 studies the need to extract clear, practical, and easily understandable information was recognised. When on placement I worked with a stroke patient with global aphasia and during the synthesis process important information was also gleaned; both of these factors informed me of the types of music-based treatments available when working with this client population, as well as providing indications of the types of disorders and clients which may benefit from such treatments.
It was also necessary to explore both the treatments and the studies in more detail in order to determine how and why the treatments did, or did not, produce an effect. In addition, during this process it was necessary to identify themes which revealed information of particular relevance to newly-trained music therapists. This information was obtained by exploring and identifying patterns and common factors among the music-based treatments and the studies.
Characteristics of studies: Focus, analysis, and outcomes
The relationships among the studies were explored in terms of their focus, data collection and analysis, and their measures and outcomes.
Focus of the studies.
Seven studies had a specific focus on investigating the efficacy, benefits and potentials of music-based treatments (Conklyn et al., 2012; Hough, 2010; Jungblut, 2005; Jungblut et al., 2009; Kim & Tomaino, 2008; Wilson, Parsons, & Reutens, 2006; Zipse, Norton, Marchina, & Schlaug, 2012). Of these seven, Jungblut (2005) and Jungblut et al. (2009) investigated the efficacy of the SIPARI method using participants with non-fluent aphasia. Four of the seven studies investigated the effects of MIT and adapted forms of MIT, while one study explored the potential of a unique musical treatment protocol.
A further four studies were similar to one another in that they each explored two interventions or approaches simultaneously. Within these four studies, Racette, Bard, & Peretz, (2006) and Straube et al. (2008) investigated singing and speaking with participants. They aimed to identify the potentials of singing and to determine if the effects of singing are solely responsible for improving the speech of aphasic patients. Using an RCT design, Lim et al. (2012) investigated the effects of neurologic music therapy (NMT) and speech language therapy. The NMT consisted of MIT and therapeutic singing. Similarly, Schlaug et al. (2008) examined two interventions using participants who were randomly assigned to attend MIT sessions and speech-repetition therapy.
Data collection and analysis.
Several methods of data collection were applied throughout the 15 studies. Five studies presented pre and post treatment data collection and analysis (Conklyn et al. 2012; Jungblut, 2005; Lim et al. 2012; Schlaug et al., 2009; Vines et al., 2011). Five other studies were designed by establishing a baseline prior to treatment; this was followed by an analysis of outcomes at specified follow-up assessments (Hough, 2010; Jungblut et al., 2009; Schlaug et al., 2008; Wilson et al., 2006; Zipse, Norton, Marchina, & Schlaug, 2012). For example, Schlaug et al. (2008) established a baseline assessment of speech output and conducted follow-up assessments after 40 and 75 sessions of treatment. Three studies used qualitative descriptive analysis of participant capabilities and assessments at the beginning of treatment and reported the outcomes on an on-going basis (Hartley et al., 2010; Kim and Tomaino, 2008; 2010). The final two studies investigated the relationship between singing and speaking (Racette et al., 2006; Straube et al., 2008); they collected their data when experimental conditions were created and evaluated it post treatment.
Measures and outcomes.
Identifying the relationship between the measures taken in each study and the outcomes of each study was important in order to determine which aspects of the participants’ disorders were targeted and measured during each study. All 15 studies provided outcomes and recommendations for future research. The participants in nine of the quantitative studies showed significant improvements in their outcome measures. For example, an RCT conducted by Conklyn et al. (2012), that used a control group and a treatment group, demonstrated a significant difference in pre and post assessments in the treatment group following a session of MMIT. Significant improvements in assessment tests of expressive linguistic skills, speech output, phrase production and fluency of speech were found in studies conducted by Hough (2005), Jungblut (2005), Jungblut et al. (2009), Schlaug et al. (2009), Vines et al. (2011) and Zipse et al. (2012). The outcomes of the qualitative study by Tomaino (2010) also revealed an improvement in singing ability at 4 weeks into the study. After 4 months of treatment the lone participant was singing more freely, had complete recall and flow of words, and had improved speech production, speech fluency, and word retrieval.
Themes and Implications for Practice
Bearing in mind previous information identified and explored concerning the relationships among, and characteristics of, the 15 studies, this section aims to outline the principal themes that emerged during the synthesis process. These themes specifically relate to why and how the interventions may work, as well as commonalities between approaches and key elements of the interventions. During the exploration and synthesis of the 15 studies, three major themes emerged; these were:
- the use of singing and vocal exercises;
- stimulating the right hemisphere;
- and the use of speech prosody in musical exercises.
Singing songs and vocal exercises.
A total of eight studies utilised various musical methods of singing and vocal exercises (Hartley et al., 2010; Jungblut, 2005; Jungblut et al., 2009; Kim and Tomaino, 2008; Lim et al., 2012; Racette et al., 2006; Straube et al., 2008; Tomaino, 2010). When singing or exploring songs, all eight studies used songs that were familiar to the participants or they used prelearned songs. Singing familiar songs – which forms a key component of the SIPARI method – was central to the treatment interventions used by Jungblut (2005) and Jungblut et al. (2009). The six other studies also explored familiar song-singing as a treatment protocol; in addition, they looked at the efficacy of singing and the relationship between singing and speaking. Racette et al. (2006) and Straube et al. (2008) found that speech improvements and superior performance during singing can not be explained solely by the effect of singing. However, Racette et al. (2006) proposed that singing along to an auditory model has greater potential to improve intelligibility than speaking alone. The single participant in the study undertaken by Straube et al. (2008) was able to sing phrases more accurately than speak the same phrases. It was suggested that this was not due to the effect of singing but possibly due to a memory-based mechanism linking melody to words.
Vocal or oral exercises, in the form of musically-assisted breathing exercises, featured within four of the eight studies (Jungblut, 2005; Jungblut et al., 2009; Lim et al., 2012; Kim and Tomaino, 2008). In each of these four studies participants were encouraged to focus on breathing while being supported and assisted by therapists. For example, Kim and Tomaino (2008) conducted an exercise which encouraged breathing into single-syllable sounds; the participants were instructed to focus on breathing and to gently accompany an exhale with a vocal sound. These sounds were then developed into pitched syllables and, in turn, into vowel sounds. Similar exercises were executed in the two studies investigating the SIPARI method where participants were encouraged to focus on respiration, and to support and regulate articulation and respiration (Jungblut, 2005; Jungblut et al., 2009).
Stimulating the right hemisphere.
It is suggested that MIT works on the basis of engaging language-capable right-hemispheric regions of the brain to compensate for the damaged left hemisphere (Albert et al., 1973). In 11 of the 15 studies, an integral part of various musical interventions with stroke survivors involved an attempt to stimulate and engage their undamaged right hemispheres (Conklyn et al., 2012; Jungblut, 2005; Jungblut et al., 2009; Kim and Tomaino, 2008; Lim et al., 2012; Schlaug et al., 2008; Schlaug et al., 2009; Tomaino, 2010; Vines et al., 2011; Wilson et al., 2006; Zipse et al., 2012). Tentative evidence suggests that improvements resulting from MIT are linked with increased right-hemispheric activity (Hough , 2010; Schlaug et al., 2009; Zipse et al., 2012;). Vines et al. (2011) tested a hypothesis based on this evidence where direct current brain stimulation was used to increase excitability of the right hemisphere during MIT treatment. In turn, current was also applied to reduce excitability of the right hemisphere. Results showed significant improvements in fluency of speech when areas of the right hemisphere were stimulated, thus providing support strongly in favour of the authors’ hypothesis.
To further increase phrase production during MIT, right-hemispheric networks are stimulated by tapping the left hand. Seven studies which utilised MIT, MMIT, and adapted forms of MIT engaged the right hemisphere by tapping the participant’s left hand or by the participant moving their left hand rhythmically or using their left hand to tap on a drum. The two studies exploring the SIPARI method engaged the right hemisphere using rhythmic exercises and improvisations involving instruments and the voice (Jungblut, 2005; Jungblut et al., 2009). Kim and Tomaino (2008) also used rhythmic speech cueing to stimulate the right hemisphere. Participants were guided in clapping or tapping on a drum along with the speech rhythm of the target phrase which was being exercised vocally. In contrast, the study undertaken by Hough (2010) – which investigated an adapted form of MIT – did not use the left-hand tapping method. Hough’s results were in disagreement with the need for left-hand tapping; the participant in his study showed significant improvements in standardised tests without engaging in left-hand tapping. Hough therefore suggested that it is not a critical component of MIT. Despite this conclusion, further research is necessary in order to establish the validity of these findings.
The practice of stimulating the right hemisphere was common throughout the literature. Each musical intervention adopts a slightly different approach but essentially they all share a common aim, i.e. to increase right-hemispheric activity in order to augment speech production.
Use of speech prosody.
In 11 studies vocal and oral musical exercises were used in conjunction with each other as part of treatment interventions (Conklyn et al., 2012; Hough, 2010; Jungblut, 2005; Jungblut et al., 2009; Kim and Tomaino, 2008; Lim et al., 2012; Schlaug et al., 2008; Schlaug et al., 2009; Vines et al., 2011; Wilson et al., 2006; Zipse et al., 2012). Of these 11 studies, eight of them explored various forms of MIT; they focused in particular on the use of phrases and sentences that are intoned. These intoned phrases were examined on the basis of three elements of spoken prosody: variation in pitch, tempo and rhythm of the utterance, and stress points (Baker, 2000). While several types of MIT were revealed in this research, these core exercises appeared consistently throughout the studies. Their use was also evident in three further studies which did not use MIT. When investigating the SIPARI method, intonation, prosody, and rhythmic speech exercises also formed key elements in the approaches taken by Jungblut (2005) and Jungblut et al. (2009). Kim and Tomaino (2008) also incorporated vocal intonation and musical dynamics into treatment protocols.
Implications for newly qualified practitioners.
When exploring the commonalities between the 15 music-based treatment studies, the main themes that emerged included: singing songs and vocal exercises, increasing activity of the right hemisphere, and the use of speech prosody exercises (discussed above). Based on the evidence and knowledge gained from the literature, this paper proposes that it is essential that newly-qualified practitioners are made aware of these three themes when conducting music-based interventions with post stroke patients displaying speech and language disorders. Table 2 provides essential information for any practitioner. This table would have proved beneficial to my own experience as a student when working with post stroke aphasic patients.
| Themes/ Interventions | Essential Elements | Benefits for Client |
|---|---|---|
| Singing and Vocal Exercises |
|
|
| Stimulating the Right Hemisphere |
|
|
| Use of Speech Prosody |
|
|
Limitations
Narrative Synthesis Stage Four: Assessing the Robustness of the Synthesis
As far as I am aware, this is the first narrative synthesis review that looks at music-based treatments used to rehabilitate left-hemispheric stroke patients displaying disorders of speech and language. It was intended from the outset that the review would be transparent; it adopted a defined search strategy and clearly specified inclusion and exclusion criteria. The process has aimed to be systematic in the analysis and exploration of the resulting data. However, it is acknowledged that certain factors may draw the robustness of the synthesis and the strength of its findings into question.
The data analysis found that in 13 studies there were significant improvements reported in the outcomes following the various treatments. Due to a number of factors, the potential for generalisation of these results within this synthesis must be considered. It must be noted that the number of participants involved in each of the 15 studies was small. The RCT by Conklyn et al. (2012), with 30 participants overall in both the control and treatment groups, involved the largest number of participants across all of the studies. Of the multiple-participant case studies, Racette et al. (2006) – with eight participants – had the most participants. All other studies were single-participant case studies.
It must also be noted that the number of sessions, as well as the duration of treatment, varied greatly across the 15 studies. For example, Conklyn et al. (2012) conducted just a single session investigating the effects of MMIT. In comparison, Schlaug et al. (2008) carried out 75 MIT sessions. Few studies were carried out over similar time frames. Hartley et al. (2010) reported on 9 years of music-therapy treatment with a single participant; their research provided qualitative analysis of improved linguistic skills. Jungblut et al. (2009), who also reported significant improvements, conducted 3 years of treatment that included 360 SIPARI sessions with a single participant. On the other hand, several studies, while also reporting improved linguistic skills, conducted their various treatments over a relatively short period of time comprising a number of weeks (Hough, 2010; Jungblut, 2005; Kim and Tomaino, 2008; Lim et al., 2012).
Three studies investigated the potentials of unique treatment protocols. Hough (2010) and Zipse et al. (2012) proposed adapted forms of MIT that revealed improvements in phrase production for each of their single participants. Kim and Tomaino (2008) also investigated the efficacy of an innovative treatment protocol and provided qualitative descriptions of guidelines for practitioners. Although the three studies reported improved outcomes, each study represents early research into newly-developed musical interventions. These interventions, therefore, require further research before their efficacy can be established and included in future reviews of this nature.
Bearing these points in mind, the robustness of this review is, somewhat, weakened. Thus, it proved difficult to consolidate the three themes (discussed above) which appeared relevant to newly-qualified practitioners because these themes had emerged from evidence which was, at times, tentative. While I proposed that these themes are essential elements in music-based work with patients suffering from speech and language disorders, due to a lack of consistency in approaches, I cannot suggest a duration or frequency of treatment. It is anticipated that this uncertainty may be damaging to the compilation of music-therapy programmes by newly-qualified practitioners because the amount of treatment required to achieve positive results cannot be estimated.
This synthesis was also subjected to my own biases because I had previous experience working with a patient with aphasia. Prior to conducting this research, my preconceived aspirations anticipated confirmation from the literature that these musical treatments produced positive effects. Therefore, it was difficult to maintain a neutral perspective when analysing the outcomes of the studies. For example, when considering the conclusions made by Racette et al. (2006) and Straube et al. (2008) which argued that music-based interventions were not responsible for improvements in speech, these findings were, somewhat, rejected in favour of the evidence provided in the majority of studies which directly linked improvements in speech with music-based interventions.
Conclusions
The current paper provides a narrative synthesis review of music-based treatments used to rehabilitate individuals with disorders of speech and language following left-hemispheric stroke. Although there is a wealth of literature pertaining to the relationship between music and stroke, a systematic search for recent literature specifically focusing on speech and language disorders revealed only 15 studies; this indicates that up-to-date literature in this area is limited. This review is in agreement with a similar review undertaken by Hurkmans et al. (2012) that found that all 15 studies reported positive treatment outcomes.
After conducting an in-depth synthesis, five types of music-based treatments were revealed; these were: (a) MIT, (b) songs and singing, (c) adapted forms of MIT, (d) SIPARI method, and (e) MMIT. Through further exploration and analysis of the interventions and studies, three major themes emerged; these were: (a) singing songs and vocal exercises, (b) stimulating the right hemisphere, and (c) use of speech prosody. By being familiar with these three themes, it is proposed that a newly-qualified practitioner will have an adequate starting point when working with left-hemispheric stroke clients displaying speech and language disorders. However, it does not necessarily follow that these themes should form a pioneering treatment protocol. It is also hoped that practitioners working in the field of neurological rehabilitation can use this synthesis, and these themes (Table 2), as a source of preliminary information to explore the musical interventions that exist, as well as their core elements. It is also recommended that newly-qualified practitioners conduct comprehensive research of all music-based treatments which they intend to implement and that they do not rely, solely, on this review. If this is done, consistency and a systematic approach in the implementation of interventions may be achieved.
Appendices
Appendix A: Literature Matrix (pdf)
Appendix B: Preliminary Synthesis of Studies (pdf)
References
Albert, M. L., Sparks, R. W., & Helm, N. A. (1973). Melodic intonation therapy for aphasia. Archives of Neurology, 29, 130–131. 10.1001/archneur.1973.00490260074018
Baker, F. A. (2000). Modifying the melodic intonation therapy program for adults with severe non-fluent aphasia. Music Therapy Perspectives, 18, 110–114. 10.1093/mtp/18.2.110
Brust, J. C. M. (1980). Music and language: Musical alexia and agraphia. Brain, 103, 367-392. 10.1093/brain/103.2.367
Brust, J. C. M. (2001). Music and the neurologist: A historical perspective. Annuals of the new york academy of sciences, 930, 143-152. 10.1111/j.1749-6632.2001.tb05730.x
Baker, F., & Tamplin, J. (2006). Music therapy methods in neurorehabilitation: A clinicians manual. London, UK: Jessica Kingsley Publishers.
Briggs, C. A. (1991). A model for understanding musical development. Music Therapy,10, 1–21. 10.1093/mt/10.1.1
Chen‐Hafteck, L. (1997). Music and language development in early childhood: Integrating past research in the two domains. Early child development and care, 130(1), 85-97. 10.1080/0300443971300109
Conklyn, D., Novak, E., Boissy, A., Bethoux, F., & Chemali, K. (2012). The effects of modified melodic intonation therapy on nonfluent aphasia: A pilot study. Speech, Language and Hearing Research, 55, 1463–1471. 10.1044/1092-4388(2012/11-0105)
Duffy, J. R. (1995). Motor speech disorders: Substrates, differential diagnosis, and management. (2nd ed). London: Elsevier Mosby.
Hallowell, B., & Chapey, R. (2008). Introduction to language intervention strategies in adult aphasia. In R. Chapey (Ed.), Language intervention strategies in aphasia and related neurogenic communication disorders. (5th ed). (pp. 3–19). Philadelphia, PA: Lippincott and Wilkins. Hargreaves, D. J. (1986). The Developmental Psychology of Music. Melbourne, Australia: Cambridge University Press.
Hartley, M. L., Turry, A., & Raghaven, P. (2010). The role of music and music therapy in aphasia rehabilitation. Music and Medicine, 2(4), 235–242. 10.1177/1943862110382661
Hough, M. S. (2010). Melodic intonation therapy and aphasia: Another variation on a theme. Aphasiology, 24(6-8), 775–786. 10.1080/02687030903501941
Hurkmans, J., de Bruijn, M., Boonstra, A. M., Jonkers, R., Bastiaanse, R., Arendxen, H., & Reinders-Messelink, H. A. (2012). Music in the treatment of neurological language and speech disorders: A systematic review. Aphasiology, 26(1), 1–19. 10.1080/02687038.2011.602514
Irish Heart Foundation (n.d.). Facts on Heart Disease and Stroke. Retrieved from http://www.irishheart.ie/iopen24/facts-heart-disease-stroke-t-7_18.html
Jungblut, M. (2005). Music therapy for people with chronic aphasia: A controlled study. In D. Aldridge (Ed.), Music therapy and neurological rehabilitation:Performing health (pp. 189-209). London, UK: Jessica Kingsley Publishers.
Jungblut, M., & Aldridge, D. (2004). The music therapy intervention SIPARI® with chronic aphasics-research findings. Neurological Rehabilitation, 10(2), 69–78.
Jungblut, M., Suchanek, M., & Gerhard, H. (2009). Long-term recovery for chronic global aphasia: A case report. Music and Medicine, 1(1), 61–69. Retrieved from http://mmd.iammonline.com/index.php/musmed/article/view/MMD-2009-1-1-7
Kim, M., & Tomaino, C. M. (2008). Protocol evaluation for effective music therapy for persons with non fluent aphasia. Topics in Stroke Rehabilitation, 15, 555–569. 10.1310/tsr1506-555
Lim, K-B., Kim, Y-K., Lee, H-J., Yoo, J., Hwang, J. Y., Kim, J-A., & Kim, S-K. (2012). The therapeutic effect of neurologic music therapy and speech language therapy in post stroke aphasic patients. Annals of Rehabilitation Medicine, 37, 556–562. 10.5535/arm.2013.37.4.556
Lucas, P. J., Baird, J., A, L., Law, C., & Roberts, H. M. (2007). Worked examples of alternative methods for the synthesis of qualitative and quantitative research in systematic reviews. BMC Medical Research Methodology, 7(4), 1–7.
McDermott, O., Crellin, N., Ridder, H. M., & Orrell, M. (2013). Music therapy in dementia: A narrative synthesis systematic review. International Journal of Geriatric Psychiatry, 28, 781–794. 10.1002/gps.3895
Popay, J., Roberts, H., Sowden, A., Petticrew, M., Arai, L., Rodgers, M., Britten, N., Roen, K., & Duffy, S., 2006). Guidance on the conduct of narrative synthesis in systematic reviews. A product from the ESRC Methods Programme. Final Report. Lancaster University.
Racette, A., Bard, C., & Peretz, I. (2006). Making non-fluent aphasics speak: Sing along! Brain, 129, 2571–2584. 10.1093/brain/awl250
Safranek, R. (2010). The use of music therapy in stroke rehabilitation. Music Therapy.
Schlaug, G., Norton, A., & Marchina, S. (2008). From singing to speaking: Why singing may lead to recovery of expressive language function in patients with Broca’s aphasia. Music Perception, 25(4), 315–323. 10.1525/mp.2008.25.4.315
Schlaug, G., Marchina, S., & Norton, A. (2009). Evidence for plasticity in white matter tracts of chronic aphasic patients undergoing intense intonation-based speech therapy. Annals of the New York Academy of Science, 1169, 385–394. 10.1111/j.1749-6632.2009.04587.x
Schlaug, G., Norton, A., Marchina, S., Zipse, L., & Wan, C. Y. (2010). From singing to speaking: Facilitating recovery from nonfluent aphasia. Future Neurology, 5(5), 657-665. 10.2217/fnl.10.44
Straube, T., Schulz, A., Geipel, K., Mentzel, H., & Miltner, W. H. R. (2008). Dissociation between singing and speaking in expressive aphasia: The role of song familiarity. Neuropsychologia, 46(5),1505–1512. 10.1016/j.neuropsychologia.2008.01.008
Tomaino, C. M. (2010). Recovery of fluent speech through a musicians use of prelearned song repertoire: A case study. Music and Medicine, 2(2), 85–88. Retrieved from http://mmd.iammonline.com/index.php/musmed/article/view/MMD-2010-2-2-3
Vines, B. W., Norton, A. C., & Schlaug, G. (2011). Non-invasive brain stimulation enhances the effects of melodic intonation therapy. Frontiers in Psychology, 2(230) 1-10. 10.3389/fpsyg.2011.00230
Wilson, S. J., Parsons, K., & Reutens, D. C. (2006). Preserved singing in aphasia: A case study of the efficacy of melodic intonation therapy. Music Perception, 24(1), 23–36. 10.1525/mp.2006.24.1.23
Woodward, A. (2012). Arts-based practices in regions affected by war. Voices: A World Forum for Music Therapy, 12(2). 10.15845/voices.v12i2.633
Wong, E. H. (2004). Clinical guide to music therapy in adult physical rehabilitation settings. Silver Spring, MD: The American Music Therapy Association. Inc.
Zipse, L., Norton, A., Marchina, S., & Schlaug, G. (2012). When right is all that is left: Plasticity of right-hemisphere tracts in a young aphasic patient. Annals of the New York Academy of Science, 1252, 237–245. 10.1111/j.1749-6632.2012.06454.x
