Use of the extended therapy outcome measure for children with dysarthria

Abstract

Increasing demand on healthcare resources has led to a greater emphasis on the examination of the impact of service delivery on outcomes. Clinical assessments frequently do not cover all aspects of change associated with interventions for those with complex conditions. This paper reviews the need for more comprehensive outcome measurement suitable for clinical practice and benchmarking. It describes an extension of the Therapy Outcome Measure for specific use in reflecting the impact of the broad range of interventions commonly required when managing children with dysarthria. Three case histories are used to illustrate the approach, and data from four speech-language pathology services are used to illustrate the value of benchmarking.

Keywords::

• Outcome measurement
• assessment
• dysarthria
• benchmarking
• Therapy Outcome Measure

Introduction

The paper considers how outcome measurement can contribute to the improvement of clinical services for children with dysarthria. It reviews the importance of collecting data on every client, so that services can be compared and best practice identified. The paper details the adaptation of the Therapy Outcome Measure (Enderby, John, & Petherham, 2006) for this purpose and provides details of a pilot benchmarking exercise.

Improving clinical services is always challenging. It is particularly difficult for services attending to the needs of children with neurological deficits which compromise communication. The underlying neurological impairment can range in severity, and in the number of other functions it affects in addition to that of speech. Correspondingly, services provided within care pathways for these children include a broad range of interventions in response to the complexity of this condition. The role of the speech-language pathologist (SLP) identified by the Royal College of Speech and Language Therapists in the UK for children with stable dysarthria is outlined in Supplementary Appendix 1 available online at http://informahealthcare.com/doi/abs/10.3109/17549507.2014.902994 (Royal College of Speech and Language Therapists, 2010). It is likely that individual services will emphasize some of these approaches rather than others (for example, a SLP may or may not have the opportunity of promoting access to the curriculum).

Given the range of potential interventions for a child with dysarthria, it is challenging to identify outcome measures that can be used to capture meaningful changes in their communication. Measures need to compare the same child over time, different children, and different services. Capturing this information would be of value for both clinical and research purposes. The Cochrane review of speech and language therapy for young children with dysarthria due to non-progressive brain damage (Pennington, Miller, & Robson, 2009a) aimed to determine the efficacy of speech and language interventions. The review concluded that there were no studies of sufficient quality for inclusion. This was disappointing given the amount of research that has been published in this area.

It is difficult to arrive at a straightforward conclusion about the evidence base for speech and language therapy for children and young persons with dysarthria. Most studies have included a small number of participants, children with different levels of severity of speech difficulty, different associated comorbidities, and different times since onset. Additionally, most studies have used particular aspects of speech as the primary outcome, without considering the impact of therapy on function, autonomy, and acceptability. However, the general impression is that consistent, intensive, and targeted speech and language therapy can have a positive impact upon the speech disorder and associated psychosocial impact (Pennington, Miller, Robson, & Steen, 2010).

Large randomized controlled trials, whilst important, may add little to the evidence base, given the heterogeneity of the client group, the complexity of the underlying conditions, and the individual nature of the communication impairment on the person and their family. Research commonly uses measures of intelligibility (either from word lists or more recently via conversation) as the primary outcome measure for identifying any benefit associated with the intervention of SLPs (Yorkston, Beukelman, & Bell, 2010). This is appropriate if this is the only target of the speech and language intervention. However, having this as a single objective of an intervention is unlikely in clinical practice. The goals of speech and language intervention for children with dysarthria are bound to be broad. Such therapy can be viewed as a complex intervention (Petticrew, 2011) and individual therapy goals can produce different effects. For example, altering an environment can affect the ability to communicate more effectively, or developing confidence may have a direct impact on the vocal output and improve the speech impairment.

Another challenge to research and clinical practice in the area of dysarthria is the complexity of the condition underlying the speech impairment. Most research uses clients with particular speech, language, or cognitive impairments to evaluate the efficacy of certain interventions. It is usual and appropriate to have strong inclusion and explicit exclusion criteria. Whilst this is necessary to identify the particular effects of an intervention on a specific condition, it does lead to most studies recruiting small samples. Reduced statistical power then leads to problems of generalizability to the broader clinical population. Most researchers screen many individuals before recruiting participants to their studies, and exclude many (for a variety of reasons, including concomitant physical, psychological, and social challenges). Thus, whilst research is essential to improving evidence-based practice, we also need more robust information from our general caseloads to inform what we are achieving with the majority of children who are not recruited into research.

Role of outcome measures in health service delivery for children with dysarthria

Globally there have been substantial changes in health service delivery over the last decade. These relate to changes in demography, increased expectations, and new technologies. This has placed considerable pressure on healthcare resources and has led to an increasing interest by service commissioners and providers on cost effectiveness (Department of Health, 2010; NHS, 2012). In the UK, new commissioning arrangements are placing greater emphasis on service providers improving the quality of service delivery, and on the use of explicit outcome measurement tools to examine the effectiveness and efficiency of service delivery.

Outcome measurement is important as it can assist those who purchase services, and also providers of those services, to identify whether the stated objectives of the service are being met. Whilst evidence-based practice has underpinned service improvement, it is generally recognized that health and social care is frequently complex, and requires inspection of more subtle aspects of impact (Pawson, 2003). Historically, much research related to surgical, medical, and allied health interventions considers cost-effectiveness as related to the reduction in disease, mortality, and re-infection (Greenhalgh, Wong, Westhorp, & Pawson, 2011). The effect of a service on reducing disability and improving the quality-of-life (which is the overarching goal of therapy for children with dysarthria) is less often considered. However, it is recognized that many services have social, functional, and psychological interventions as explicit aims of their delivery. These holistic aspects, which include improving coping skills, well-being, and integration, are all harder to define and measure.

Assessments vs outcome measurement

Outcome measurement is different to assessment, in that the latter aims to assess the deficits and abilities in specified areas, whereas an outcome measure aims to identify whether an individual has benefited (or otherwise) from an intervention. Assessments can be used as outcome measures but are infrequently used for this purpose (Enderby et al., 2006).

Whilst clinical assessments may be used (such as assessments of dysarthria, cognitive awareness, or language competence) as outcome measures, they are not necessarily well designed for this purpose. Psychometrically robust assessments often aim to identify deficits and abilities to inform the initiation of treatment but are not often used on completion of treatment. Standardized assessments are very rarely used in the manner required to ensure reliability, as detailed by the test authors. Rather, they are often adapted by clinicians for practical and logistical reasons associated with time and resource constraints.

Informal assessments, case reports, observations, and history taking are all used by clinicians to provide additional information to that of the assessment, to inform clinical judgements relating to the nature of the difficulty and its change over time. On this basis clinicians will form an impression as to whether the service recipient has improved or benefited from the intervention or not. However, these clinical impressions may lack reliability and validity. For example, service users may become more relaxed with the therapist over time and have improved function because of the improved relationship. Alternatively, service providers may accommodate to the particular behaviours of a client. In this case it is possible that the service provider will judge the speech as improving when it is their ability to interpret the speech of the service user more reliably that has changed.

Benchmarking

It is generally recognized that the provision of services tends to be highly variable and inequitable, even in countries with government-funded national healthcare systems such as the National Health Service in the UK (NHS, 2010). In the context of services for children with dysarthria and alternative and augmentative communication (AAC) services, it is clear that some services provide detailed assessment, loans of equipment, therapy to improve articulatory accuracy, improve broader communication, foster confidence and self image, stimulate the use of AAC, and encourage integration. In contrast, other services may only be able to support certain aspects of provision, such as assessment, advice, and a limited range of communication systems.

Services are rarely developed upon child need alone, and frequently reflect historical patterns and interventions of an enthusiast at some time in the past (Gorsky, 2008). If we are to reduce variability and inequity in service provision, as well as striving for year-on-year improvement in our service delivery, it is important that we identify the impact that we are having, and also differences between services in the influence they are having on the recipients of the interventions.

Swage (2004) identified five areas that need to be addressed to improve the quality of health services: the culture of the organization; inequity and variability in the provision of services; involvement of users; sharing good practice; and detecting and dealing with poor performance. Benchmarking can assist with four of these areas by: making comparisons between services to identify best practice; providing a baseline to assess change over time; facilitating the capacity to learn from variation; and contributing to the knowledge base. Benchmarking (which is the process of comparing the outcomes, services processes, and performance metrics with similar services to identify areas of strength and weakness) has utility in the absence of other information. For example, benchmarking may assist in the identification of what works best for whom, and when relationships between processes and outcomes of care are not well understood (Bullivant, 1996; Bullivant & Roberts, 1997; Ellis, 2000; Franklin, 1995; Higgins, 1997; Swage, 2004). This is often the case in speech-language pathology services, especially for children with complex conditions that are relatively low in prevalence, but relatively high in service needs.

Benchmarking can only take place: if the same outcome measure is used by similar services; if this measure appropriately facilitates the collection of data on issues associated with the objectives of the service; if the measure is used consistently and reliably; and if the results are compared with other services with the same objectives. For an outcome measure to reflect the impact of an intervention it needs to be used in a systematic way on more than one occasion (Collis & Bloch, 2012; Ellis, 2000; Franklin, 1995).

In speech-language pathology, the few benchmarking studies reported in the literature have focused on internal benchmarking of services. In this work, performance of a service is compared over time, or alternatively before or after a change in service delivery (Yaruss, LaSalle, & Conture, 1998), rather than external benchmarking, which compares one service against other similar services.

The use of outcome measures and benchmarking in the world of speech and language therapy has changed radically over the last decade. The majority of reports have assessed factors associated with the process of service delivery such as number of sessions available, lengths of treatment, and waiting times. This has been at the expense of assessing the broader impact of therapy on issues such as modifying the speech and language impairment, improving communication and social participation, as well as the impact on well-being (Fuhrer, Jutai, Scherer, & Deruyter, 2003). An outcome measure that could be used to inform the broad impact of speech and language therapy on speech impairment, communication ability, participation, and well-being in children with dysarthria in benchmarking initiatives has the potential to significantly improve services and outcomes for these children. The following section describes a candidate measure for this purpose.

The Therapy Outcome Measure

The Therapy Outcome Measure (TOM; Enderby et al., 2006) is based on the International Classification of Functioning, Disability and Health (ICF) developed by the World Health Organization (WHO, 2001), which provides a structure to reflect issues beyond that of the impairment alone. The TOM was designed to be a simple, reliable, cross-disciplinary, and across-client group method of gathering information on a broad spectrum of issues related to the objectives of services and relevant to all clients receiving speech and language therapy. The TOM allows the clinician to describe the relative abilities of an individual across four dimensions. These follow the dimensions used by the ICF (WHO, 2001) of impairment, activity restriction, and social participation. The dimension of the individual's well-being was added to the TOM, since many goals in therapy aim to improve the emotional status of the child and, where appropriate, the carer.

“Impairment” is concerned with the integrity of body systems, and includes psychological and physiological structures and functioning, and concerns the degree of abnormality observed, in terms of its variance from the norm for a human being (of same age, gender, etc.). For the purposes of this paper we view “impairment” as the degree of abnormality in neuromuscular control of speech. Results gathered from standardized assessment such as the Frenchay Dysarthria Assessment (Enderby & Palmer, 2008) can assist in determining the nature and severity of the impairment.

The category of “activity restriction” is concerned with the limitations on actions or functions for an individual, which would correspond to the degree of communication difficulty experienced by the person with dysarthria. Determining this may be facilitated by conversational analysis and/or all tests of intelligibility. The domain of “participation” concerns consideration of any social disadvantage experienced by the individual, including social participation, interaction, and autonomy. Whereas “well-being” is concerned with emotions, feelings, burden of upset, concern, and anxiety.

The four TOM dimensions (Impairment, Activity, Participation, and Well-being) are rated on an 11-point ordinal rating scale, with 0 representing the severe end of the scale and 5.0 representing normal for age, sex, and culture. The integers are defined with semantic operational codes that identify the severity of the difficulty experienced on each dimension. This allows the assessor to indicate whether the person is slightly better or worse than the descriptor. The procedures for using the TOM require the therapist to assess the individual referred for treatment using their usual assessment procedures such as standardized tests, observation, report, and consideration of medical and social history. The information collected from these sources allows the therapist to judge the appropriate rating to assign.

The reasons for adopting the TOM for children with dysarthria (TOM-DYS) are 3-fold. First, it is based upon the strong conceptual framework of the ICF (WHO, 2001). This framework has been adopted internationally and covers the objectives of services aimed at supporting complex conditions and incorporating a broad range of interventions such as that required for those with dysarthria (including augmentative and alternative communication). The second reason is associated with the relative strength of the reliability and validity of the TOM, which has been used in a broad range of clinical research and settings (Enderby & John, 1999; Hamilton, Granger, & Sherwin, 1987; John & Enderby, 2000; John, Enderby, & Hughes, 2005; John, Hughes, & Enderby, 2002; Roulstone, John, Hughes, & Enderby, 2004; Stineman, Jette, Fiedler, & Granger 1997) and found to have strong psychometric properties.

Third, the TOM allows individuals to undertake any number of assessments they wish, and to convert their findings to the rating scale in a quick and simple fashion, allowing comparison and benchmarking of all aspects of the service. An outcome measure must capture the possible impact of the full range of interventions for the condition. Table I summarizes approaches used in speech-language pathology to manage dysarthria in children, and maps them on to the domains of the ICF. With children the focus of an intervention could be to maximize communicative potential, encourage communication, and support the creation of the most beneficial environment to enable communication. For a child with severe dysarthria, the goal might be to have some degree of functional verbal communication (Enderby & Emerson, 1995), or to find non-speech based alternatives.

Table I. Techniques for intervention in dysarthria classified by the ICF.

ICF dimension	Techniques
Impairment	Normalization of muscle tone and/or increasing strength of movement precision and co-ordination (Pinto, Ozsancak, Tripoliti, Thobois, Limousin-Dowsey, & Auzou, 2004; Schmidt & Fitzpatrick, 1996; Joffe & Reilly 2004). Prosthetic methods for controlling some of the symptoms associated with dysarthria, e.g., palatal lifts (Enderby, Hawthorn, & Servant, 1984). A speech systems approach to control breath support, phonation, and speech rate (Pennington & Noble, 2010; Pennington, Roelant, Thompson, Robson, Steen, & Miller, 2013). Behavioural techniques (Deane, Whurr, Clarke, Playford, & Ben-Shlomo, 2001). Examples include: The Lee Silverman Voice Technique (Fox & Boliek, 2012); Dysarthria Treatment Programmes (Drummond, Worley, & Watson, 2003).
Activity	Facilitating a communicative environment (Pennington & Noble, 2010; Pennington, Thomson, James, Martin, & McNally, 2009b). The use of devices and biofeedback giving feedback to enable individuals to monitor and modify speech characteristics (Pinto et al., 2004; Schmidt & Fitzpatrick, 1996). Assistive devices, ranging from the low-tech such as an alphabet board, to a high-tech computerized Augmentative and Alternation Communication system. Voice amplifiers used to increase effectiveness of communication (Enderby, Judge, Creer, & John, 2013). Altering the environment can improve function (Sellars, Hughes, & Langhorne, 2005), e.g., ensuring that the listener has full view of the dysarthric speakers face (Palmer & Enderby, 2007).
Participation	Providing education about dysarthria to child, family, and/or school, and advice on promoting self-esteem, to increase social interaction and participation in society, facilitate interaction in school, and communication on social settings. This will lead to increased autonomy, which is associated with general well-being and quality-of-life (Pennington & Noble, 2010; Pennington et al., 2009b).

Adapted TOM for dysarthria

The adapted TOM-DYS for individuals of any age with dysarthria allows for a rating of five impairments: physical, cognitive, sensory, speech, and language. This differs from the original as it is clear from the literature that these service recipients most frequently have a complex combination of different impairments (Supplementary Appendix 2 available online at http://informahealthcare.com/doi/abs/10.3109/17549507.2014.902994 provides a detailed description of the TOM-DYS). If services are to be compared then it is important that the complexity of the caseload is appropriately reflected to ensure comparison of alike services.

Furthermore, research indicates that those with a greater number of impairments (the more complex cases) require greater service involvement over a longer period of time than those with individual impairments (Hamilton et al., 1987). Specialist services may attract more complex cases and require a broader skill mix than more local services, and it is likely that outcomes would be different.

Illustrative case histories (names have been altered to maintain anonymity)

The following section uses three case examples to illustrate how the TOM-DYS can be used by service providers to document treatment outcomes for children with dysarthria.

Case history 1

Brendan was a little boy aged 6 years, with cerebral palsy characterized by severe spastic dysarthria and limited control over all limbs. The Frenchay Dysarthria Assessment indicated a pattern that, when converted onto the Therapy Outcome Measure, suggested severe-to-moderate speech impairment, with most consonants being attempted but poorly represented acoustically (a rating of 2). Audiological and visual assessment suggested no impairment. Assessment of language indicated comprehension, vocabulary, and syntactic abilities that were below normal for his age. Brendan was unable to signal his preferences with any degree of consistency, and had few friends outside the family. Whilst Brendan appeared to be happy and only expressed frustration and distress occasionally, his parents showed marked concern and unhappiness over his condition and their ability to cope. A synopsis of these findings is provided in Table II.

Table II. Brendan pre-treatment using the TOMS-DYS.

Impairment (4 scores)		Activity/communication (2 scores)		Participation (1 score)	Wellbeing (1 score)	Carer wellbeing (1 score optional)
Physical	1	Without augmentation	1	2	4	2
Cognitive	3	With augmentation	NA
Sensory	5
Speech	2
Language	2.5

Brendan had 9 months of regular (once or twice weekly) speech and language intervention. Therapy for his speech impairment focused on respiratory support and improving accuracy of the articulation, by concentrating on improving auditory self-regulation. Makaton (http://www.signsimply.com/what-is-makaton.html) was introduced to Brendan, his family, and school. This had an impact on improving his language and cognitive skills. This also played a role in reducing the frustration and concern of his family. The well-being of his family was further assisted by the SLP supporting them in expressing their concerns and guiding them to meet other parents with disabled children. The outcome of his therapy is summarized in Table III.

Table III. Brendan post-treatment.

Impairment (4 scores)		Activity/communication (2 scores)		Participation (1 score)	Wellbeing (1 score)	Carer wellbeing (1 score optional)
Physical	1	Without augmentation	1	3.5	4	4
Cognitive	3	With augmentation	3
Sensory	5
Speech	3
Language	4

Case history 2

Amelia, aged 10 years, became dysarthric following a head injury as a result of a road traffic accident. She was referred to speech-language pathology 6 months after the incident. While able to walk and undertake most personal care she had increased tone in all limbs, rendering her movements to be slow and clumsy. Her speech reflected this spasticity, and while she could make herself understood in a quiet situation she had difficulty communicating in the classroom and with those who were unfamiliar with her. She had lost confidence, and was embarrassed and withdrawn in many social situations. She had become angry and depressed, but her parents were managing well. The situation is summarized in Table IV.

Table IV. Amelia pre-treatment.

Impairment (4 scores)		Activity/communication (2 scores)		Participation (1 score)	Wellbeing (1 score)	Carer wellbeing (1 score optional)
Physical	3.5	Without augmentation	3	3	2	4
Cognitive	4	With augmentation	NA
Sensory	5
Speech	3
Language	5

Amelia received therapy from a SLP 5-times a week for a 2-month period. Intervention focused on improving respiratory support, vocal power, and intonation. She was also introduced to an iPad and voice-activated applications (apps). Psychological support was provided to assist Amelia, with guidance given on how to broaden her ability to express her personality, and how to exploit her general communication skills (see Table V).

Table V. Amelia post-treatment.

Impairment (4 scores)		Activity/communication (2 scores)		Participation (1 score)	Wellbeing (1 score)	Carer wellbeing (1 score optional)
Physical	3.5	Without augmentation	4	4	3.5	4
Cognitive	4	With augmentation	4.5
Sensory	5
Speech	4
Language	5

Case history 3

Robin, aged 3 years, was referred to speech-language pathology after two of failure to thrive which was associated with a genetic abnormality. The condition was thought to be non-progressive. Robin was unable to sit unsupported and had no control over his limbs. He appeared to have severe cognitive impairment and was assessed as being severely visually impaired. He used one vocal tone which was not modified for expression in any way. Whilst impossible to assess with any degree of accuracy it appeared that he had some very limited comprehension of situations, some concrete words, and vocal tone (see Table VI).

Table VI. Robin pre-treatment.

Impairment (4 scores)		Activity/communication (2 scores)		Participation (1 score)	Wellbeing (1 score)	Carer wellbeing (1 score optional)
Physical	0	Without augmentation	0	1	4	3
Cognitive	1	With augmentation	0
Sensory	2
Speech	0
Language	.5

During speech and language intervention it became clear that Robin's cognitive deficits were not as profound as first thought. He was able to indicate choices using large picture icons triggering sound. He also indicated that he could sequence and remember events. His vision remained the same, but he was soon able to indicate positive and negative by vocal tone and some rudimentary articulation. His ability to understand basic instructions and conversation became evident with the use of his communication aid, as did his willingness to initiate interactions. Thus, without his communication aid his ability to communicate had improved, and with the use of his aid he could indicate a little more consistently and reliably. This allowed him to engage more in the special nursery, as reflected in an increased participation score (see Table VII).

Table VII. Robin post-treatment.

Impairment (4 scores)		Activity/communication (2 scores)		Participation (1 score)	Wellbeing (1 score)	Carer wellbeing (1 score optional)
Physical	0	Without augmentation	2	2	4	3.5
Cognitive	3	With augmentation	2.5
Sensory	2
Speech	2
Language	2

Therapy outcome measure for benchmarking

An example of using the adapted TOM-DYS for benchmarking with clinical data provided by four different speech-language pathology services is used here for illustrative purposes only. Figure 1 provides information on the outcomes associated with speech and language intervention for young people with dysarthria.

Figure 1. Comparison of the Therapy Outcome Measure for children with dysarthria (TOM-DYS) scores for Impairment (a), Activity/Communication (b), and Participation (c) across four trusts providing speech and language intervention for children with dysarthria.

Each service recruited 10 consecutive young people (aged under 15 years) and identified their score on TOM-DYS at the beginning and end of speech and language intervention. It is clear in Figure 1 that services 3 and 4 received referrals of persons with more severe dysarthria impairment. Despite this they achieved greater gains. Without knowing the age and diagnosis of these young people, it is inappropriate to speculate whether the gains are wholly attributed to the speech-language pathology intervention. Interestingly, despite the initial impairment score being more severe for services 3 and 4, the communication level (activity) was better than for services 1 and 2 at outset. All of the services showed significant gains in the dimension of participation, with service 2 having the most challenging status on this dimension at outset.

It is clear from Figure 1 that there were differences in the nature of children being referred to these services, and that the outcomes were different. This raises many different questions, but perhaps most importantly it raises the issue of the complexity of the conditions associated with dysarthria. For example, a child with dysarthria and sensory and/or cognitive impairment is likely to require more speech and language therapy and possibly will improve to a lesser extent. The adaptation to the TOM for dysarthria permits appropriate comparisons to be made.

Conclusion

Speech and language therapy for children with dysarthria needs to be multi-faceted, and to embrace the broad impact of this impairment. It is unlikely that the majority of children with dysarthria will be recruited into research aimed at improving the knowledge base; however, our knowledge base can be informed in more ways than by research alone. We therefore need to record the impact of therapy in a way that reflects the complexity of the intervention and the breadth of the goals; however, we need to do this in ways that are reliable and valid, and provide useful data that can be aggregated.

Outcome measures are required if we are to: first, provide appropriate information to commissioners; second, monitor change over time; third, improve the quality of our services; and, finally, to reduce variability in practice by identifying strengths and weaknesses. The development of outcome measures which are reliable, while also being sensitive to small but important changes in the areas of importance to service users, relatives, and practitioners, is difficult. Children, their families, practitioners, professionals-in-training, service providers, and funders; however, will all benefit from collection and thoughtful analysis of such data. Those who provide services to children with dysarthria are strongly encouraged to choose a clinically feasible outcome measure, which has demonstrated reliability and validity, and to use it without fail over time, so that these potential benefits can be realized.

Acknowledgements

Dr Sue Baxter has helped with the preparation of this manuscript. Her assistance is much appreciated. Thanks are also due to the therapists who have participated in the developments reported here.

Declaration of interest: The author reports no conflicts of interest. The author alone is responsible for the content and writing of the paper.