A new paradigm has emerged in health care emphasizing the importance of scientific evidence in guiding clinical decision making. Evidence-based practice (EBP) increasingly is regarded as fundamental to ethical practice. In this age of accountability and managed care, the future of various treatment approaches may well depend upon the extent to which their effectiveness can be demonstrated empirically. The potential benefits of becoming an evidence-based discipline are to produce better clinical education and training, more cost-effective practice, more knowledge about difficult or unusual cases, and perhaps most importantly, better treatment for each individual who seeks our help.
A widely used definition of EBP is offered by David Sackett and colleagues: "Evidence-based medicine is the integration of best research evidence with clinical expertise and patient values" (Sackett et al, 2000, p.1). The three elements of this definition are reflected in Christine Dollaghan's assertion that "...EBP offers us a framework and a set of tools by which we can systematically improve in our efforts to be better clinicians, colleagues, advocates, and investigators— not by ignoring clinical experience and patient preference but rather by considering these against a background of the highest quality scientific evidence that can be found" (2004, p. 4). Thus, EBP supports a strong client-centered approach to intervention.
To engage in EBP, a clinician should:
- ask a clear, focused question
- find the best evidence
- critically appraise the evidence and determine if it can be used with the client
- integrate the evidence with clinical judgment and client values and circumstance
- evaluate performance
(These steps are taken from http://www.cochrane.uottawa.ca/presentations.asp, A Primer on Evidence-Based Clinical Practice.)
While some sources relate step 5 to evaluation of the practitioner's performance of the EBP process, others regard it as the fundamentally important step of gathering data to determine whether the selected treatment is indeed effective for the client.
Numerous documents relating to the EPB process are available on the ASHA Web site along with links to tutorials and continuing education opportunities. An article by Jan Ingham (2003) also describes and illustrates these steps as they apply to treatment of stuttering.
Sorting the Evidence
Not all sources of evidence are equal, and much of the skill in EBP involves sorting the strong from the weak evidence. Various systems have been developed for classifying strength of evidence, with hierarchies generally listing sources with greatest to least strength (i.e., least to most potential for bias). An example of such a hierarchy appears on ASHA's Web site.
Most hierarchies regard systematic reviews of randomized controlled trials (RCTs—an experiment where individuals are randomly assigned to treatment or control conditions or to different treatments) as the strongest level of evidence. However, in some hierarchies, information from well-controlled single-subject experimental designs (with appropriate replications) also are regarded as strong sources of evidence. For instance, guidelines established in clinical psychology consider rigorous multiple single-case experiments along with RCTs as "Category I" evidence (Chambless & Ollendick, 2001). Given the prevalence of single-subject experimental designs in our research literature, such studies, if well designed and adequately replicated, may offer a valuable contribution to our evidence base.
Opinions vary as to the relevance of RCTs to our field. Some describe these designs as the "gold standard" (e.g., Ingham, 2003) whereas others question their value for all treatments in the field (e.g., Bernstein Ratner, 2005). Reilly et al (2004) have suggested that given the complexity and low incidence of some communication disorders, other scientific models of investigation may be more appropriate for some treatment questions. Such models await development.
The Evidence Base in Stuttering Treatment
Not surprisingly, appraising the evidence base in fluency disorders is difficult and controversial. Like many other disciplines, fluency has a paucity of well-controlled assessment and treatment studies. Contributing to this problem is lack of agreement on outcome measures, varied definitions of success and the heterogeneity of our clients. Further, of the few group studies in peer-reviewed publications, the majority have investigated behavioral or integrated treatment approaches. As Bernstein Ratner (2005) noted, stuttering modification studies are under-represented in the scientific literature. This is a curious situation given the long-standing popularity of these methods.
In spite of this circumstance, developments are encouraging. One such advance relates to the highly systematic development of a direct intervention for early stuttering. The Lidcombe Program has an extensive research base. Results of an RCT that compares the Lidcombe Program to a control group recently were published in the British Medical Journal (Jones et al., 2005). Although the program awaits replication by a fully independent team of investigators, it remains the only program for early stuttering subjected to rigorous testing. This of course does not mean that no other treatment program is effective. However, it does mean that effectiveness of the Lidcombe Program at a group level has been firmly established.
The evidence base for treatment of school-age children and adults also is growing. Controlled trials comparing intensive smooth speech treatment with intensive EMG feedback and home-based smooth speech have been conducted (Craig et al., 1996). Adult group studies evaluating long-term effects of treatment using repeated measures are also accumulating (e.g., Boberg & Kully, 1994; Howie et al., 1982; Langevin & Boberg, 1993; O' Brian et al., 2003).
The existing research base supports the conclusion that treatment programs exist that can effectively treat stuttering. Much more research is required to determine the functional components of treatment programs, comparative effectiveness of different treatments, and the treatment best suited for a particular client. Further, there will always be clients who present with unusual profiles and who bear no resemblance to the populations in group studies. For such clients, more relevant evidence may come from single subject experimental studies, case studies, and, in some instances, expert opinion.
But no matter the strength of an evidence base underlying a treatment, or the extent to which a client matches a subject population, the key factor determining the validity of the approach is the client's individual response to treatment (step 5, "evaluate performance" in the process of EBP above). The principles that apply to ensuring validity of measures used in formal studies apply also to measures used to determine effectiveness for individuals. Measures need to relate to the treatment goals, be repeated over time, be representative of clients' performance, and be valid and reliable. Langevin and Kully (2003) illustrate the process of evaluating data to determine the effectiveness of a treatment program for an individual and for research purposes.
EBP in Speech-Language Pathology
Although EBP has had a slow start in speech-language pathology, it is receiving increasing support from speech-language pathology organizations in the U.S. and other countries. For instance, ASHA has established EBP as a focused initiative and undertaken a range of strategies to facilitate and encourage its use in speech-language pathology and audiology, including the integration of EBP principles into the revised scope of practice documents.
In Canada, the Canadian Association of Speech-Language Pathologists and Audiologists (CASLPA) was among the first affiliate organizations of the Canadian Cochrane Network and Centre (CCN/C—a center that works to promote evidence-based health care in Canada) and, through its Affiliate Representative on the advisory board, has had a voice in the direction of the network and raised the profile of EBP among CASLPA members (Orange, 2004).
In Australia, the School of Human Communication Sciences at La Trobe University has established a curriculum based on problem-based learning principles, a teaching approach consistent with EBP processes. An excellent book, Evidence-based Practice in Speech Pathology, was recently published (Reilly, Douglas, & Oates, 2004) and the national journal, Advances in Speech-Language Pathology, featured a series of articles on EBP (6/2, 2004).
In the U.K., The Royal College of Speech and Language Therapists established clinical guidelines that draw upon existing research and has incorporated aims related to evidence-based practice into its current strategic plan.
Publications relating to the application of EBP in fluency disorders have begun to appear. A series of articles was recently published in the Journal of Fluency Disorders (28/3, 2003) and a chapter on the evidence base for the treatment of stuttering (Block, 2004) was included in the aforementioned book by Reilly and colleagues (2004). Although concepts relating to EBP are not new to our discipline, we face the same challenges as other disciplines in fully implementing them into practice. Successful implementation will require joint effort and commitment from organizations, clinicians and researchers alike.
Deborah Kully is co-founder and executive director of the Institute for Stuttering Treatment and Research (ISTAR), Faculty of Rehabilitation Medicine, University of Alberta. She has sought to apply the principles of EBP to clinical practice for many years. Contact her at firstname.lastname@example.org.
Marilyn Langevin is clinical director at the Institute for Stuttering Treatment and Research, Faculty of Rehabilitation Medicine, University of Alberta. Her research interests currently center on EBP and the social impact of stuttering on children. Contact her at email@example.com.
The authors would like to thank Lucia Barbosa of Brazil, Johanna Einarsdottir of Iceland, and Rosemarie Hayhow of the U.K., for sharing their experiences with EBP with us. Their input assisted us in preparing this article.