When a parent seeks treatment from a speech-language pathologist (SLP) for a preschool-age child who has started to stutter, the SLP is faced with a difficult decision. "Should we implement teatment immediately or should we wait to see if the child recovers naturally?" While a number of children start to stutter in the third and fourth years of life, it is known that many of them will recover without professionally guided intervention. In times gone by, the typical advice to parents was "Your child will probably grow out of his stuttering, so let's wait and see". At first blush, this seems sensible. After all, why should a child and family go through the demanding therapy process if the child is going to recover anyway?

However, a parent is entitled to ask "Will my child respond as well to treatment if I wait until it seems unlikely that he or she will recover naturally?" This is a reasonable question, because we know that stuttering becomes more entrenched and difficult to treat as children get older.

The SLP, then, needs to balance the possibility that the child will recover naturally with the possibility that the child will not respond as well to treatment if it is delayed for some period. Common sense, and knowing that stuttering is generally more difficult to treat in school-age children, together suggest that treatment should commence sometime in the preschool years (prior to age 6 years). However, parents and SLPs want to know whether waiting for a year or more after the onset of stuttering, even within the preschool years, involves a trade-off in treatment effectiveness.

The current emphasis on evidence-based practice in speech-language pathology means that SLPs will want to take all the available evidence into account when making clinical decisions. And, fortunately, there is now a considerable body of scientific research to guide decisions about whether and when to treat stuttering in preschoolers. The relevant research findings come from two broad areas of enquiry: research into natural recovery, and research into the effectiveness of treatment.

The bulk of the relevant research into natural recovery comes from the Illinois Early Stuttering Project. In this project, children who start to stutter are followed for a number of years to determine how many recover naturally, and whether there are factors that predict that recovery (or lack of it). The findings of this research have been published widely and most SLPs will be familiar with them. Only the findings that are relevant to the issue of when to treat are summarised here.

First of all, the most recent estimate of natural recovery to come from this research is 74%. In other words, around 74% of children who start to stutter will recover without professional help. However, the researchers consider this estimate to be conservative, and the percentage of children who recover may be even higher than this. Time taken to recover naturally varies considerably, ranging from a few months to two or more years. A small percentage of children apparently recover later in the school-age years. It is thought, however, that at least half the children who are destined to recover naturally do so within a year.

Second, there are some factors that predict -- to a greater or lesser extent -- that a child will recover naturally. These include having family members who have recovered from stuttering, and being a girl. Many more girls than boys recover naturally. It is also thought that children who are less proficient producing speech sounds are not as likely to recover as those who are. However, since these children are not necessarily deficient in speech sound production they cannot be identified simply by scores on a test. Time since onset is also a factor to consider. It stands to reason that the longer a child has been stuttering, the less likely he or she is to recover naturally. The other side of the coin is that there are some factors that do not predict recovery. While it is not an intuitively appealing idea, children whose stuttering is more severe are just as likely to recover naturally as children whose stuttering is milder.

In summary, there are factors that predict natural recovery, and lack of it. However, the SLP cannot know whether these apply to an individual child. In other words, when a child is brought to the clinic, unfortunately the SLP is unable to advise the parent in advance whether their child will -- or will not -- recover without treatment.

The second body of research to consider is concerned with treatment responsiveness. In deciding whether and when to treat a child with early stuttering, SLPs need to know whether waiting for a while after the onset of stuttering to see if natural recovery occurs means that the child will not respond as well, or as quickly, to treatment, if it is required. To date, the research relevant to this issue has been conducted with two different treatment programs: the Multiprocess approach and the Lidcombe Program. Research into Multiprocess treatment indicates a significant, albeit modest, relationship between time since onset of stuttering and treatment time. In other words it seems that, in a number of cases at least, children who have been stuttering for longer take more time to complete treatment. According to the researchers, this suggests that it is probably wise to commence treatment as soon as possible.

Research into the Lidcombe Program, however, presents a somewhat different picture. Recent research has shown that there is tendency for preschoolers who stutter to respond more quickly to the Lidcombe Program if they have been stuttering for a year or more. What this means, conservatively, is that delaying treatment with the Lidcombe Program for a year after the onset of stuttering is unlikely to reduce the child's responsiveness to the treatment.

In summary, then, it seems that the answer to the question of whether to wait for some period after the onset of stuttering to see if natural recovery occurs depends on the treatment involved. At least, that is the case for the two treatment programs that have been researched. It is not clear why the two treatments differ in this regard, but it may be because the treatment approaches are so different. The Multiprocess treatment is primarily (although not solely) indirect, while the Lidcombe Program focuses directly on the child's speech. In any event, what is clear from all this research is that while waiting for a year before implementing the Lidcombe Program is unlikely to jeopardise treatment responsiveness, this is not necessarily the case for other treatments.

So far this discussion has centred on research findings. However, there are of course other factors relevant to each child that need to be considered. For example, is the child distressed by his or her stuttering? If that is the case, and the Lidcombe Program is the treatment of choice, is it reasonable to delay intervention for a year after onset, particularly if the child does not recover naturally and continues to stutter during that time? The answer to this is not clear-cut. It may be that steps can be taken to reduce the child's distress during this time, with the parents being reassured that the child is likely to respond just as well to treatment at a later time. In any event, it is recommended that any child for whom treatment is delayed be monitored actively by the SLP not only for signs of natural recovery but also to ensure that the child does not suffer undue stress, frustration or social penalty because of the stuttering.

To return to our basic question, then -- "To treat or not to treat?" -- one thing is clear. It is clear that the answer is not arrived at easily! It would seem that the decision about whether and when to treat needs to be made by the SLP in conjunction with the family, taking into consideration the type of treatment and the circumstances of the individual child and family. One thing about which there seems to be consensus, however, is that treatment should be implemented within the preschool years if it is to be maximally effective. While this means that some children who receive treatment would have recovered without it, this seems a small price to pay for ensuring that treatment effectiveness is maximised for the remaining children who are not destined to recover naturally.

Kingston, M., Huber, A., Onslow, M., Packman, A., & Jones, M. (2003). Predicting treatment time with the Lidcombe Program: Replication and meta-analysis. International Journal of Language and Communication Disorders, 38, 165-177.

Onslow, M., Packman, A., & Harrison, E. (2003). The Lidcombe Program of early stuttering intervention: A clinician's guide. Austin, TX: Pro-Ed.

Starkweather, C.W., & Gottwald, S.R. (1993). A pilot study of relations among specific measures obtained at intake and discharge in a program of prevention and early intervention. American Journal of Speech-Language Pathology, January, 51-58.

Yairi, E., & Ambrose, N. (1999). Early childhood stuttering I: Persistency and recovery rates. Journal of Speech, Language, & Hearing Research, 42, 1097-1112

You can post Questions/comments about the above paper to Ann Packman before October 22, 2003.