Preliminaries To Treatment Outcomes Research for Adults Who Stutter

About the presenters: J. Scott Yaruss is an assistant professor at the University of Pittsburgh and Co-director of the Stuttering Center of Western Pennsylvania. Yaruss's research focuses on the development of stuttering in children, including analysis of linguistic and motoric factors that affect children's fluency. Yaruss also studies the evaluation of treatment outcomes for adolescents, and adults who stutter.

Robert Quesal is a professor at Western Illinois University, where he serves as program director. He teaches courses in fluency disorders, anatomy, and speech and hearing science. He has presented papers at regional, national, and international meetings. He has served as newsletter editor for ASHA's Special Interest Division 4 - Fluency & Fluency Disorders, since 1997. He has been listowner of the STUT-HLP listserv since June of 1993.


Preliminaries to Treatment Outcomes Research for Adults Who Stutter

by J. Scott Yaruss and Robert Quesal
from Pennsylvania and Illinois, USA

In recent years, there has been increasing emphasis on the importance of documenting the outcomes of treatment, not only within the specialty of fluency disorders, but also across the broader discipline of speech-language pathology. The reasons for this resurging interest are numerous, including the ethical need to demonstrate how well our treatments work, the need to respond to the reasonable and appropriate requests of consumer groups who wish to play a more active role in setting the standards for treatment, and the need to respond to increasing pressure from third-party payers to justify the costs of treatment.

Although there is widespread agreement that our field needs to pay more attention to the documentation of treatment outcomes, there is less agreement about how we should proceed with this task. For example, some authors have pointed out that the outcomes of certain treatment approaches have already been demonstrated, so we simply should select and utilize those strategies that have been proven to achieve their goals (e.g., Cordes, 1998). In general, the treatments that have been evaluated in this way tend to involve training speakers to use prolonged speech to reduce or eliminate stuttering events. Clinicians who use these treatment approaches can have confidence that their treatment will achieve its stated goals. Likewise, clients, consumer groups and insurance companies can have a clear idea of what to expect when individuals are enrolled in these types of treatments.

Of course, prolonged-speech is not the only approach to treating stuttering. For numerous reasons, including different philosophies about the nature of stuttering, different opinions about the clinician’s role in treatment, and different client preferences for the goals of treatment, many clinicians have chosen to select other kinds of treatments. For example, many treatment approaches focus not just on reducing stuttering events, but also on reducing the broad spectrum of negative consequences that stuttering can have for people who stutter (e.g., Manning, 1996). These consequences can include emotional reactions such as low self-esteem, shame, or fear of speaking; avoidance of words, sounds, or speaking situations; functional difficulties in communicating with others; and problems performing activities necessary for their career. For the most part, the specific outcomes of broad-based treatments designed to address these negative consequences have been less thoroughly evaluated, leading to the present concerns about stuttering treatment outcomes.

Some have suggested that there is a certain degree of risk in selecting treatment strategies that are not clearly supported by empirical evidence. To be sure, there is ample evidence that some practitioners have used and promoted approaches that are less-than-appropriate for treating stuttering or for helping people who stutter (e.g., see review in Bobrick, 1995). Still, it seems reasonable to assume that at least some of these approaches were originally developed and continued to be utilized over time because clinicians and clients found them to be helpful (Yaruss, 1998a). Rather than abandoning these approaches because of the lack of available data, therefore, we can and should attempt to collect the outcome data necessary for clinicians, consumers, and payers to feel confident selecting these treatments.

Unfortunately, evaluating these so-called eclectic treatments can be very difficult. There are several reasons for this difficulty. The purpose of this paper is to briefly summarize some of the authors’ efforts to address two of the primary roadblocks to furthering outcomes research for broad-based treatment programs for adults who stutter. Before beginning this discussion, however, a brief comment about the current status of outcomes research for eclectic stuttering treatment would seem appropriate. In all likelihood, it is the difficulty of the task that is to blame for the lack of substantive efficacy data for eclectic treatment approaches, rather than a lack of interest in the outcomes of treatment on the part of practitioners who have used and developed these techniques, or a lack of interest in scientific inquiry (as some have insinuated). Nevertheless, as the pressure on clinicians to justify their actions has grown, so has the need to overcome the obstacles to documenting the outcomes of these types of treatment approaches. These are the goals to which this paper, and the effort which it describes, are directed.

Defining the goals of treatment. The first step to documenting treatment outcomes is to clearly define the goals of treatment. Although this sounds straightforward, it is not. In part, this is due to the fact that the desired outcomes of treatment differ from client to client, as well as over time, depending upon the individual’s needs (e.g., Manning, 1996). Furthermore, these desired outcomes can address a variety of different aspects of the stuttering disorder, including changes in client’s level of fluency, changes in the client’s speech attitudes, and changes in the client’s willingness and ability to talk in difficult speaking situations (even if this results in increased stuttering). Such outcomes stand in contrast to the typical stated goal of prolonged speech treatment, which is the attainment and maintenance of a specific target frequency of stuttered syllables (e.g., 0% or 2%). With the prolonged speech approach, the client, clinician, and payer know exactly when the client has achieved the goal of treatment. With the eclectic approaches, the multiple goals are constantly changing and sometimes in conflict with one another. Thus, it can be more difficult to describe these goals for treatment outcomes research, and even more difficult to identify a specific point in time at which the treatment can be said to have achieved its desired outcomes.

Yaruss (1998a, b) has suggested that one way to overcome the difficulties of defining the goals of treatment is to utilize a common framework for describing the components of different types of treatment approaches. Such a framework would have to be broad in scope in order to accommodate all of the different types of stuttering treatments. It would also be helpful for this framework to be consistent with frameworks used with other disorders across the broader field of healthcare or rehabilitation to facilitate the communication of treatment goals to other professionals (and, not trivially, to insurance companies).

One framework that appears to hold some promise for facilitating the documentation of stuttering treatment outcomes is the World Health Organization’s (WHO, 1980) three-part classification system for describing the consequences of disorders. The International Classification of Impairments, Disabilities, and Handicaps (ICIDH) has been applied to stuttering by several authors (Conture, 1996; Curlee, 1993; McClean, 1990; Prins, 1991), and has been used as the centerpiece of a recent model for describing the broad scope of the stuttering disorder (Yaruss, 1998b). In brief, the model describes how disorders such as stuttering can be viewed at three levels: impairmentdisabilityhandicap. Impairment refers to the basic structural or functional difficulty the person is experiencing; disability refers to the problems the person may have in completing life tasks due to the impairment; and handicap refers to the disadvantages the person may have in fulfilling expected life roles (such as occupational roles or social roles) due to the impairment and resulting disability.

As applied to stuttering, impairment refers simply to the production of stuttering events (a functional difficulty in producing fluency speech), disability refers to the difficulties the person may have in communicating (e.g., trouble using the phone or talking to others), and handicap refers to the limitations the person may experience in achieving life goals (e.g., getting a desirable job). Importantly, not all speakers who exhibit a stuttering impairment will necessarily experience disability and handicap — whether or not a speaker experiences these consequences is largely a function of how he or she reacts to the stuttering. Thus, Yaruss (1998b) added a separate component highlighting the role of the speaker’s affective, behavioral, and cognitive reactions to the basic ICIDH model. (For a more detailed discussion of the ICIDH as it applies to stuttering, see Yaruss, 1998a, 1998b. Click here to see a schematized version of the model from the Yaruss [1998c] ISAD paper.)

Using this framework, clinicians and their clients can then explicitly define their desired outcomes of treatment at various points in the therapeutic process at each of the four levels of the model (impairment, reactions, disability, and handicap). In addition to facilitating the development of a consistent definition of treatment goals that can be evaluated over time, this common terminology can also facilitate comparison across various treatment programs.

Measuring treatment outcomes. In addition to describing various possible consequences of the stuttering disorder, this model can also be used to facilitate the process of measuring stuttering treatment outcomes. Specifically, the components of the model can be associated with the components of different stuttering treatment approaches. For example, components of treatment that focus on reducing the frequency of stuttering can be said to operate at the level of the impairment; components of treatment that focus on helping the speaker become more comfortable with the fact that he or she stutters can be said to operate at the reactions level; and components of treatment that focus on functional communication in specific speaking situations can be said to operate at the disability level. Treatment goals can then be defined — and evaluated — at each of these levels to describe the outcomes of various treatment approaches.

A primary assumption underlying this approach to treatment outcomes documentation is that measurement should be conducted at each level at which the treatment is designed to operate (Yaruss, 1998a). In other words, treatments that focus primarily at the impairment level (i.e., reducing or eliminating stuttering events) should be assessed at the impairment level (by counting the frequency of stuttering); treatments that focus at the impairment and reactions level should be assessed at both levels (e.g., by counting the frequency of stuttering and assessing changes in speech attitudes); and so on.

Some might suggest that outcomes analysis should be completed at all levels for all treatments. Such thorough documentation would certainly have the benefit of allowing comprehensive comparison of different types of treatment approaches. Such information would certainly be of benefit to consumer groups that wish to evaluate existing treatment approaches for its membership. The pitfall of this approach, however, is that it would be reasonable for some practitioners to be concerned that their treatment was being assessed in a domain that it does not intend to treat. For example, a clinician who utilizes a strict fluency shaping approach might find it inappropriate to question whether his treatment changes the client’s functional communication abilities, since this was not part of the intended scope of treatment. This is similar to how a clinician who uses a broad-based, eclectic treatment approach might feel if his or her treatment were assessed in terms of the frequency of disfluencies, since a very reasonable goal of treatment for some clients with an eclectic approach might be increased participation in social interaction, even if this means increased stuttering. (In actuality, practitioners who utilize broad-based treatment approaches are typically assessed only in terms of their ability to affect the client’s stuttering impairment, rather than the overall impact the stuttering disorder. As will be discussed in more detail below, this is primarily due to the fact that we have not previously had access to reliable instruments for measuring changes in clients’ experience of the stuttering disorder at levels other than the basic impairment.) Accordingly, it would seem more appropriate, for a first attempt at treatment outcomes documentation, to attempt to measure outcomes at each level at which the treatment was focused and intended to bring about changes.

To assess the outcomes of eclectic treatment approaches, then, it is first necessary to define goals of treatment at each level. (A detailed discussion of the goals of different treatment approaches is well beyond the scope of this paper, and will at the very least differ from program to program or client to client.) The next step is to measure and document changes at each level prior to the onset of treatment, throughout treatment, and at several points following the conclusion of treatment. This is where the primary problem lies for treatment outcomes documentation. Although there are well-defined techniques for measuring changes in the stuttering impairment and several instruments available for assessing selected aspects of the speaker’s reactions to stuttering, there are few, if any instruments currently available for assessing changes in the disability or handicap of the stuttering disorder.

In an attempt to address this concern, Yaruss and colleagues (Quesal & Yaruss, 1999; Yaruss & Quesal, submitted; Yaruss, Quesal, King, & Schaffner, 1998; Yaruss et al., submitted) have been working to develop a series of measurement instruments that directly assess the disability and handicap of stuttering. (A preliminary description of the first draft of these instruments was presented at last year’s ISAD conference [Yaruss, 1998c]). Following the example of Frattali (1998), disability is framed in terms of the speaker’s functional communication abilities and handicap is framed in terms of the speaker’s overall quality of life. In addition, Yaruss et al. have developed a new instrument for measuring the speaker’s reactions to stuttering that specifically and separately examines affective, behavioral, and cognitive reactions. Together with existing techniques for assessing the stuttering impairment, these three new measurement instruments (Speaker’s Reactions to Stuttering [SRS]; Functional Communication and Stuttering [FCS]; and Quality of Life and Stuttering [QOL-S]) should greatly expand the ability of clinicians and researchers to document the outcomes of a wide variety of stuttering treatment approaches.

Development of these instruments has involved several steps. First, the specific goals of each of the three instruments were defined, based in part on the ICIDH framework described above, and in part on apparent gaps in available measurement tools. Next, an initial version of each instrument was developed based on input from several researchers and clinicians, combined with input from clients and other people who stutter. These initial versions were distributed to approximately 150 people who stutter, with the help of several colleagues in the National Stuttering Project (now the National Stuttering Association [NSA]). In addition to the three measurement instrument drafts, the initial group of respondents also completed a detailed demographic questionnaire and the S-24 attitudes inventory (Andrews & Cutler, 1974) for comparison purposes. Results from the roughly 60 questionnaires that were returned were analyzed to determine which of the items seemed appropriate and which were in need of revision. This analysis involved detailed review of the distribution of data for each item, as well as examination of correlations among items both within and between the three measurement instruments. Results of this initial evaluation were reported at an ASHA convention (Yaruss et al., 1998), and feedback was solicited from the session participants. Analysis of this feedback and the completed questionnaires resulted in major changes to all three instruments, both in terms of the response mechanism and the specific questions that were asked. These revised versions were then distributed to approximately 200 people who stutter who attended the 1999 NSA Convention. Approximately 75 surveys have been returned thus far, and the results are currently being analyzed in much the same way as the initial data. Based on this analysis of the second version of the instruments, a third version of the measurement tools will be prepared and made available to the stuttering community for review. The third revised versions of the instruments should be available for review at the 1999 ASHA Convention, where the development process will be reviewed in a poster session (Quesal & Yaruss, 1999), and also at the Third World Congress of the International Fluency Association in August, 2000 (Yaruss & Quesal, submitted). Also, data from the NSA respondents have been submitted for presentation at the IFA congress (Yaruss et al., submitted). After feedback from participants in these presentations has been evaluated and suggestions for changes are incorporated into the instruments, the final versions of the instruments will be prepared.

An important component in the development process has been the valuable input provided by numerous researchers, clinicians, and people who stutter about the appropriateness and accuracy of the items and constructs that are evaluated in the questionnaires. This second ISAD conference provides another opportunity for the authors to receive such input. Please feel free to provide your comments through the threaded discussion lists or through email to the first author at: jsyaruss@csd.upmc.edu. The second draft versions of the instruments can be viewed through the following links (use the back button on your browser to return to this document):

Speaker’s Reactions to Stuttering
Functional Communication and Stuttering
Quality of Life and Stuttering

It is hoped that these three new instruments — Speaker’s Reactions to Stuttering, Functional Communication and Stuttering, and Quality of Life and Stuttering — will facilitate the process of documenting and evaluating the outcomes of stuttering treatment by providing information that is not presently available from existing measurement instruments. This is particularly important for eclectic treatment approaches that seek to bring about change in several aspects of the stuttering disorder in addition to a reduction in the production of speech disfluencies. Because these instruments are based on a well-defined and commonly utilized framework, it is also hoped that results will facilitate analysis of a wide variety of treatment approaches that view stuttering from different philosophical perspectives. Such an increase in our knowledge about stuttering treatment should improve the ability of clinicians to select appropriate treatment approaches to help their clients, while helping clients to make more informed decisions about the treatment options that are available to them. With luck, such documentation will also strengthen the case to third-party payers that stuttering therapy is effective and that coverage of such treatment is appropriate. The authors of this paper will welcome any input from readers about how we can further our efforts to improve the documentation of treatment outcomes for people who stutter.

References

Andrews, G., & Cutler, J. (1974). Stuttering therapy: The relation between changes in symptom level and attitudes. Journal of Speech and Hearing Research, 39, 312-319.

Bobrick, B. (1995). Knotted tongues: Stuttering in history and the quest for a cure. New York: Simon & Schuster.

Conture, E.G. (1996). Treatment efficacy: Stuttering. Journal of Speech and Hearing Research, 39, S18-S26.

Cordes, A.K. (1998). Current status of the stuttering treatment literature. In. A.K. Cordes, & R.J. Ingham (Eds.), Toward treatment efficacy in stuttering: A search for empirical bases. Austin, TX: Pro-Ed.

Curlee, R.F. (1993). Evaluating treatment efficacy for adults: Assessment of stuttering disability. Journal of Fluency Disorders, 18, 319-332.

Frattali, C. (1998). Outcome measurement: Definitions, dimensions, and perspectives. In C. Frattali (Ed.), Outcome measurement in speech-language pathology (pp. 1-27). New York: Thieme Medical Publishers.

Manning, W. (1996). Clinical decision making in the diagnosis and treatment of fluency disorders. Albany, NY: Delmar Publishers.

McClean, M.D. (1990). Neuromotor aspects of stuttering: Levels of impairment and disability. In J. Cooper (Ed.), Research needs in stuttering: Roadblocks and future directions (ASHA Reports, 18, 64-71). Rockville, MD: American Speech-Language-Hearing Association.

Prins, D. (1991). Theories of stuttering as event and disorder: Implications for speech production processes. In H.F.M. Peters, W. Hulstijn, & C.W. Starkweather (Eds.), Speech motor control and stuttering (pp. 571-580). Amsterdam: Elsevier Sciences Publishers.

Quesal, R.W., & Yaruss, J.S. (1999, Nov.) Assessing Reactions, Disability, and Handicap in Stuttering: An Update. Poster to be presented at the Annual Convention of the American Speech-Language-Hearing Association, San Francisco, CA.

World Health Organization. (1980). International classification of impairments, disabilities, and handicaps: A manual of classification relating to the consequences of disease. Geneva: World Health Organization.

Yaruss, J.S. (1998a). Treatment outcomes in stuttering: Finding value in clinical data. In A. Cordes & R. Ingham (Eds.), Toward treatment efficacy in stuttering: A search for empirical bases (pp. 213-242). Austin, TX: Pro-Ed.

Yaruss, J.S. (1998b). Describing the consequences of disorders: Stuttering and the International Classification of Impairments, Disabilities, and Handicaps. Journal of Speech, Language, and Hearing Research, 49, 249-257.

Yaruss, J.S. (1998c). Documenting treatment outcomes in stuttering: Measuring impairment, disability, and handicap. Paper published on the internet for the First Annual International Stuttering Awareness Day On-line conference.

Yaruss, J.S., & Quesal, R.W. (submitted). Tools for assessing reactions, disability, and handicap in people who stutter. Mini-seminar submitted for presentation at the 3rd World Congress on Fluency Disorders, Nyborg, Denmark (August, 2000).

Yaruss, J.S., Quesal, R.W., King, A., & Schaffner, M. (1998, Nov.) Assessing reactions, disability, and handicap in people who stutter. Miniseminar presented at the annual convention of the American Speech-Language-Hearing Association, San Antonio, TX.

Yaruss, J. S., Quesal, R.W., Tellis, C., Molt, L., Reeves, L., Caruso, A.J., McClure, J., & Lewis, F. (submitted). The Impact of Stuttering on People Attending the NSA Convention. Mini-seminar submitted for presentation at the 3rd World Congress on Fluency Disorders, Nyborg, Denmark (August, 2000).


September 26, 1999