Early Intervention With Childhood Stuttering Revisited
By Richard F. Curlee
The value of early intervention in treating a variety of diseases and other medical conditions is well known and has become a largely unquestioned principle of effective clinical management in medicine and other health related disciplines, including speech-language pathology. Indeed, the wisdom of beginning therapy as soon as possible after the onset of stuttering seems so self-evident to many speech-language pathologists that anyone questioning this long standing principle may risk being charged with clinical heresy. At the 1995 ASHA Convention in Orlando, Ehud Yairi and I suggested that the necessity and value of early intervention with two- to four-year-olds who stutter was more myth than fact and went on to review evidence which suggested that delaying therapy for a year or two after stuttering onset may not be harmful. This presentation was followed by an emotionally charged "discussion" that questioned both the intellectual and ethical soundness of these views. An invited restatement of these views and the series of letters to the editor that resulted were published two years later in the American Journal of Speech-Language Pathology. This is neither the first, nor likely the last, controversy to divide clinicians and researchers with strong interests in stuttering. Such controversies are inevitable whenever the validity of widely accepted theories or beliefs about stuttering or its clinical management is questioned.
This paper provides an overview of the evidence pertaining to the necessity of early intervention, the risks of delaying treatment, and the efficacy treatment currently used with this age group for those who are not familiar their discussion elsewhere. I conclude by focusing on a larger, more important issue that seems to have escaped much thoughtful discussion so far. Although most of the views expressed here have been presented jointly by Ehud Yairi and myself in other forums (e.g., Yairi & Curlee, 1995; Curlee & Yairi, 1997, 1999), this paper is solely my responsibility.
Necessity of Early Intervention
Central to the controversy on treatment necessity is accounting for: 1) reports of untreated remissions of stuttering, and 2) the discrepancy between the incidence and prevalence of stuttering. Anecdotal reports of untreated remissions are common, particularly among those of us who were children when therapy was not widely available thoughout the U.S. Estimates of the frequency of untreated remissions have usually relied on recollections of adolescents or adults who used to stutter, parents, or other family members and have ranged from 32% to over 80% (Wingate, 1976). The validity of such estimates is questionable and has been challenged, especially the higher ones (e.g., Ingham, 1983; Martin & Lindamood, 1986; Young, 1975). Their arguments claim that remissions occurring within a few months of onset do not involve "real" cases of stuttering, that common-sense home remedies of parents are responsible for almost all "untreated" remissions, which are, in fact, "treatment," that many untreated remissions are only temporary respites from stuttering, and that fewer than half of the preschoolers stop stuttering before entering grade school.. Certainly, each of these posssibilities may account for some of the untreated remissions reported, just as they may account for some of the treated recoveries of young children that have been reported.
Findings from a handful of longitudinal studies, however, have relied on direct observations of young children who had not been stuttering long, which should be more valid than those of studies relying on memories of the distant past. Without exception, findings from these studies place untreated remissions in the high range (i.e., 65- 89%) with a median of 80% (Andrews & Harris, 1964; Panelli, McFarlane, & Shipley, 1978; Masson, 1999; Ryan, 1990; Yairi & Ambrose, 1992, 1996). In addition, most of these untreated remissions appear to occur within two years of onset. Andrews' (1984) reanalysis of the Newcastle-on-Tyne data (Andrews & Harris, 1964) found that 35-40% of the children stopped stuttering within 6 months of onset, which increased to 65% after 18 months post-onset. Yairi, Ambrose, and Niermann (1993) reported similar rates of untreated remissions during the first 6 months of onset, even among children who had stuttered severely. If stuttering continues for a year, however, only four children in ten are likely to stop if untreated, which drops to fewer than two in ten after five years of stuttering (Andrews, 1984).
Turning now to the discrepancy between the incidence and prevalence of stuttering, most estimates of the lifetime incidence of stuttering cluster around 4.5 to 5.0%, whereas prevalence is substantially lower, usually ranging from 0.5 to 1.0% (Andrews, 1984; Bloodstein, 1987, Mansson, 1999). The difference in the percentage of people ever affected (i.e., incidence) and currently affected (i.e., prevalence) indicates that stuttering is a current problem for only 75-90% of those who ever stutter. The relative contributions of untreated remissions and treated recoveries to this discrepancy are central to the issues under dispute. Bloodstein's (1987) examination of incidence studies also led him to conclude that children are at highest risk for beginning to stutter between their second and fourth birthdays and that 75-90% of all stuttering onsets occur prior to 6 years of age. If fewer than half of the preschoolers who stutter stop before they enter public school as claimed, the prevalence of stuttering ought to be higher in early than in later grades of school, which seems not to be the case. A well-controlled, nationwide study of over 30,000 U.S. children in grades K-12 (Hull, Mielke, Willeford, & Timmons, 1976) found the percentage of children who stuttered varied little across grades and was essentially identical to the median prevalence of stuttering (i.e., 0.7-0.8%) across all age groups (Bloodstein, 1987). Findings summarized earlier from longitudinal studies indicate that the rate of untreated remissions, especially among preschoolers within a year or two of stuttering onset, could easily account for the discrepancy between the incidence and prevalence of stuttering. Thus, converging findings from varioue sources of data indicate that early treatment is not necessary for most young preschoolers within the first year or two of stuttering onset.
Risks of Delaying Treatment
Even if most children stop stuttering, on their own, within the first year or two of onset without having received treatment, a substantial number will continue to stutter. So, does waiting a year or more to begin therapy with young preschoolers make treatment goals more difficult to achieve, substantially increase the amount of treatment needed to achieve those goals, or result in less satisfactory treatment outcomes? As noted earlier, additional untreated remissions continue to occur, but at decreasing rates, for some period of time (Andrews, 1984). Thus, the longer children stutter the higher their risk for continuing to stutter as adults, with or without treatment. In trying to assess the risks of waiting to initiate treatment, information on how treatment outcomes differ for younger and older children might be helpful.
At the 1996 SID-4 Leadership Conference, Roger Ingham examined findings from 10 studies, seven that reported the treatment outcomes of children who stutter and three that followed preschool- and school-age children who did not receive treatment for various periods of time. A portion of his examination compared groups of children who did and did not receive therapy and had stuttered fewer than 15 months when first assessed as well as treated and untreated groups who had stuttered longer. In the early assessment group, 16 of the 22 (73%) untreated children stopped stuttering, and 12 of the 14 (86%) who received treatment recovered. In the groups with longer stuttering histories, only 5 of the 27 (19%) untreated children stopped stuttering, but 16 of the 22 (73%) who received treatment recovered.
Taken at face value, these findings indicate that rates of both untreated stuttering remissions and treated recoveries are higher among younger children who have stuttered for shorter periods of time. There are a number of risks in comparing groups that have been formed by combining data from different studies employing a variety of data collection procedures and interventions with different clinical samples. If the groups that are compared are not carefully matched, or randomly assigned, to control for factors known to be related to the remission of stuttering, such as age, sex, familial history of stuttering, etc., their comparison may not be valid. For example, when older children who have passed through the primary remission period are included, as was true in this data-set, they comprise a subgroup of children who stutter who may have predispositions or habits that increase their risk of chronic stuttering.
Unlike the data Ingham reviewed, Onslow and his colleagues (e.g., Lincoln, Onslow, Lewis, & Wilson, 1996; Onslow, Andrews,& Lincoln, 1994; Onslow, Costa, & Rue, 1990) gathered data from a series of treatment outcome studies with preschool- and school-age children who were selected from the same general clinical population who were administered the same treatment procedures supervised by staff from the same clinic. Therefore, differences in these preschool- and school-age children's findings are more likely to reflect differences in ages and lengths of time the children had stuttered rather than other group difference than would those. The Onslow group reported that stuttering was eliminated, or almost eliminated, in a median of 10.5 clinic visits in all preschoolers who completed their treatment protocol. Children in the school-age group were described as slightly more likely to relapse and needing about one more clinic visit to complete the same treatment protocol. Consequently, there appears to be no major treatment outcome differences between groups of preschool and school-age children. A caveat applicable to both these and the findings examined by Ingham is that children were designated as having "recovered" if they stuttered fewer than 0.5 words per minute in non-clinic situations. Hence, some who "recovered" may still stutter sometimes. The risks of delaying treatment for two- to four-year-olds who stutter have yet to be studied directly. As a result the risks and the factors that may increase or mitigate such risks remain largely unknown, even though it appears that earlier interventions do not always result in better outcomes and longer delays before intervening do not always adversely affect treatment.
Efficacy of Early Treatments
A variety of intervention strategies and treatment procedures for preschoolers who stutter have been described, and high rates of success have been reported without exception. Included are counseling or training parents to reduce environmental and communicative pressures, placing contingencies on fluent and stuttered utterances, training preschoolers to produce speech in ways believed to facilitate fluency, and so on. Stuttering has even been reduced when it was not the target of intervention (Wahler, Sperling, Thomas, Teeter, & Luper, 1970). Two children who stuttered were administered response-contingent stimulation for behavior problems while the frequency of each child's problem behavior and stuttering were tracked in ABAB single-subject designs. Concurrent decreases in each child's problem behavior and stuttering occurred, even though authors had taken care to ensure that inadvertent contingencies were not placed on stuttering. Moreover, improvements in young children's fluency during treatment are usually reported to require fewer treatment sessions, to generalize more frequently, and to last longer, often resulting in permanent elimination of stuttering. Lastly, no adverse effects or reactions to any of these treatments have been reported.
The apparent success of this diverse array of therapies with young children who stutter may indicate that early childhood stuttering is highly susceptible to treatment or that each of the treatment procedures described is robust and highly effective. Alternatively, when every sort of treatment seems to be able to eliminate or significantly reduce children's stuttering, it may indicate that none are necessary to effect such changes. Opposing interpretations of treatment outcome findings such as these are possible when studies' methodological limitations undermine their credibility. To my knowledge, no study of stuttering treatments outcome has met standards widely accepted as necessary for assessing treatment effectiveness (Bloodstein, 1987, pp. 400-406):
- The method must be shown effective with an ample and representative group of stutterers.
- Suitable control groups and control conditions must be used to show that reductions in stuttering are the result of treatment.
- The success of a program of therapy must not be inflated by dropouts.
- The method must be shown to be effective in the hands of essentially any qualified clinician, including those without unusual status, prestige, or force of personality.
I suspect that most clinicians believe that every child should receive treatment as soon as possible after stuttering begins, that withholding treatment for any period of time would be unethical and places the child at high risk for chronic stuttering problems, that therapies for young preschool-age children who stutter are nearly always effective, and even if treatment were unnecessary, it would do no harm (e.g., Starkweather, 1997). This is somewhat puzzling in view of the fact that valid assessments of treatment effectiveness cannot be made unless the high rates of untreated remissions known to occur in this age group and their effect on the therapy provided are adequately controlled. At present, empirical support for the efficacy of treatments used with young children who stutter is weak, and the efficacy of early interventions with children who stutter is essentially unknown.
The More Important Issue
The controversy that Ehud Yairi and I precipitated in 1995 was based on the thesis that additional scientific evidence is needed to support treatment decisions and treatment of early childhood stuttering. We hoped to convince colleagues of the critical need for soundly designed, well-controlled longitudinal studies of the natural patterns of progression and remission that characterize early childhood stuttering and the effects of early and later clinical interventions. Unfortunately, the larger, more important issue - the need for additional systemtic studies of early childhood stuttering -- seems to have been obscured by argumentation and debate that focus on specific areas of data needs, such as the necessity of early intervention, untreated remission rates, effectiveness of current treatments, and effects of delaying professional intervention. Such debate appears to ignore the fact that these issues cannot be resolved appropriately in the absence of credible research findings.
The controversy that has resulted appears to reflect different philosophies about the bases of sound clinical practice - the weight given to systematic, well-controlled empirical research relative to that given clinicians' past training, professional experiences, and current kills -- in guiding this profession's clinical practices for children who stutter. At present, much more weight is being given to our clinical experence and skills because of the dearth of adequately controlled empirical research in this area of study. I am not optimistic that this controversy will be resolved in the near future, for much of the evidence that is needed must be based on findings for which data are yet to be gathered. Until such time that scientific support for these sorts of clinical decisions is available, important clinical decisions that will determine the efficacy of our management of early childhood stuttering will continue to rest largely on opinions and anecdotal evidence that are untempered by controlled studies.
The history of healthcare is repleat with examples of how the best minds of the day are easily and frequently deceived when the treatments that are proscribed are not infomed by information gathered from rigorous scientific research. Clearly, acquiring the scientific data needed to identify intervention strategies and procedures that are necessary and effective will be neither easy nor inexpensive and will pose a major challenge for this profession. In my opinion, this is a challenge that tests the profession's and its members' ethical responsibility and commitment to this clinical population, and it is not yet clear to me that we are up to the test.
References
Andrews, G. (1984). The epidemiology of stuttering. In R. Curlee & W. Perkins (Eds.), Nature and treatment of stuttering: New directions (pp. 1-12). San Diego: College-Hill Press.
Andrews, G., & Harris, M. (1964). The syndrome of stuttering. London: Heinman Medlcal Books.
Bloodstein, O. (1987). A handbook on stuttering (4th ed.). Chicago: National Easter Seal Society.
Curlee, R., & Yairi, E. (1997). Early intervention with early childhood stuttering: A critical examination of the data. American Journal of Speech-Language Pathology, 6(2), 8-18.
Hull, F., Mielke, P., Willeford, J., & Timmons, R. (1976). National speech and hearing survey: Final report. Project No. 50987, Washington, DC: Office of Education, DHEW.
Ingham, R. (1983). Stuttering and spontaneous remission: When will the emperor realize he has no clothes on? In D. Prins & R. Ingham (Eds.), Treatment of stuttering and early childhood:Methods and issues (pp. 113-140). San Diego: College-Hill Press.
Ingham, R. (1996, May). On watching a discipline shoot itself in the foot: Some observations on current trends in stuttering treatment research. Paper presented at the meeting of the American Speech-Language-Hearing Association Special Interest Division 4, Monterey, CA.
Lincoln, M., Onslow, M., Lewis, C., & Wilson, L. (1996). A clinical trial of an operant treatment for school-age children who stutter. American Journal of Speech-Language Pathology, 5(2), 73-85.
Mansson, H. (1990. Personal communication.
Martin, R., & Lindamood, L. (1986). Stuttering and spontaneous recovery: Implications for the speech-language pathologist. Language, Speech, and Hearing Services in Schools, 17, 207-218.
Onslow, M., Andrews, C., & Lincoln, M. (1994). A control/experimental trial of an operant treatment for early stuttering. Journal of Speech and Hearing Research, 37, 1244-1259.
Onslow, M., Costa, L., & Rue, S. (1990). Direct early intervention with stuttering: Some preliminary data. Journal of Speech and Hearing Disorders, 55, 405-426.
Panelli, C., McFarlane, S., & Shipley, K. (1978). Implications of evaluating and intervening with incipient stutterers. Journal of Fluency Disorders, 3, 41-50.
Ryan, B. (1990, November). Development of stuttering, a longitudinal study: Report 4. A paper presented at the American Speech-Language Hearing Association Convention, Seattle, WA.
Starkweather, C. W. (1997). Therapy for younger children. In R. Curlee & G. Siegel (Eds.), Nature and treatment of stuttering: New directions (2nd ed., pp. 257-279). Needham Heights, MA: Allyn& Bacon.
Wahler, R., Sperling, K., Thomas, M., Teeter, N., & Luper, H. (1970). The modification of child-hood stuttering: Some response-response relationships. Journal of Experimental Child Psychology, 9, 411-428.
Wingate, M. (1976). Stuttering:Theory and treatment. New York: Irvington.
Yairi, E., & Ambrose, N. (1992). A longitudinal study of stuttering in children: A preliminary report. Journal of Speech and Hearing Research,35, 755-760.
Yairi, E., & Ambrose, N. (1996). Erratum. Journal of Speech and Hearing Research, 39, 826.
Yairi, E., Ambrose, N., & Niermann, R. (1993). The early months of stuttering: A developmental study. Journal of Speech and Hearing Research, 36, 521-528.
Yairi, E., & Curlee, R. (1995, December). Early intervention in childhood stuttering: Myths and facts. Paper presented at the annual convention of the American Speech-Language-Hearing Association, Orlando, FL.
Young, M. A. (1975). Onset, prevalence, and recovery from stuttering. Journal of Speech and Hearing Disorders, 40, 49-58.
