articleThis content is more than 5 years old.

Study on pre-school children draws criticism

Prof Sheena Reilly, Elaine Kelman | 14.04.2014

An Australian study recently published in the journal Pediatrics, entitled ‘Natural history of stuttering up to 4 years of age’, has prompted some in the dysfluency world to question its findings and methodology. We asked one of its authors to provide a Prof Sheena Reilly, Dr Eleina Kefalionos and Peta Newellsummary of the study, and a leading Speech and Language Therapist to comment on it.

Professor Sheena Reilly, from the Department of Paediatrics at the University of Melbourne, writes:

Recently we reported findings from a study of early stammering1 that surprised many and continues to be the subject of debate.

We recruited 1,910 infants, aged 8-10 months, to the Early Language in Victoria Study (ELVS). Within ELVS we embedded a study to examine the onset and development of stammering. This study was different from many others in that:

  • The children were recruited at a younger age, that is, before many of the children had started stammering;
  • Information was collected on all participants before they started stammering, as well as at frequent, regular intervals once they started;
  • Participants were recruited from the community (e.g., maternal and child health centres, magazine advertisements), rather than speech therapy clinics where children were seeking treatment;
  • In addition to stammering data, information was collected on the children’s social, emotional and behavioural development;
  • When each child was reported as stammering by their parents, and had it confirmed by one of our speech pathologists, the child was visited at their home on a monthly basis for 12 months.

All parents participating in the original ELVS were eligible and invited to participate in the stammering sub-study. The majority of parents elected to participate. Participating families were provided with a fridge magnet displaying examples of different stammering behaviours and prompted every four months to telephone our research team if they noticed that their child had started stammering. Once a parent telephoned to report that their child was stammering, a 45-minute face-to-face assessment was conducted at the child’s home. For those children confirmed as stammering, monthly home visits then took place for one year. In addition to this, a questionnaire was completed by parents every year around the time of their child’s birthday. At 4 years of age, the child’s language skills (comprehension and expression) and non-verbal cognition were measured. Parental reports of the child’s social, emotional and behavioural development and quality of life were also obtained.

The main findings were as follows:

  • Childhood stammering was more than we expected; 11.2% of children were confirmed as stammering by 4 years of age;
  • Being a twin, being male and having a mother with a higher level of education were all associated with stammering onset. However, this doesn’t mean that these factors together will predict stammering onset.

By 4 years of age:

  • Children who had stammering onset had better language development and non-verbal skills than non-stammering children;
  • The negative social, emotional and behavioural effects commonly reported to be associated with stammering, were not evident;
  • Children who stammered were not more shy or withdrawn compared to the non-stammering group;
  • Children who stammered had better health-related quality of life compared to the non-stammering group.
  • Only 6.3% of children recovered from stammering in the first 12 months after onset. This recovery rate was lower than had previously been reported. Recovery rates within the first 12 months after onset were higher for boys, for children who did not repeat whole words at onset and for children who had a lower stammering severity at onset.


We concluded that stammering seemed to be more common in the pre-school years than was previously thought. We were surprised that the negative consequences associated with stammering were not apparent in the majority of children by 4 years of age.

Current best practice recommends waiting 12 months before starting treatment unless the child is distressed, parents are concerned, or the child becomes reluctant to communicate.  This recommendation is based on research conducted about the Lidcombe Programme, which is the only speech and language therapy treatment for early stammering that is supported by randomised controlled trials2, 3. The evidence indicates that waiting 12 months after onset before commencing treatment may actually improve a child's response to treatment. However, this period involves ‘watchful waiting’, which requires parents to monitor fluctuations in their child’s stammering severity as well as changes to their child’s response to stammering.

Given that (a) so few children recovered from stammering in the first year after onset; and (b) there were no detectable negative outcomes by 4 years of age, we suggested that treatment for some children may be delayed for slightly longer than 12 months to allow more time for natural recovery to occur. Limited resources are available to manage early childhood stammering; therefore we argued that these resources should be allocated to those children who do not recover naturally and/or those who experience negative outcomes.

Given the low rates of recovery reported in our study, we were unable to determine what predicts which children will recover from stammering, but this will be the focus of our research in the future.

1. Reilly, S., Onslow, M., Packman, A., Cini, E., Ukoumune, O. C., Bavin, E. L., Prior, M., Eadie, P., Block, S., & Wake, M. (2013). Natural history of stuttering to 4 years of age: A prospective community-based study. Pediatrics, 132(3), 460-467.
2. Jones, M., Onslow, M., Packman, A., Williams, S., Ormond, T., Schwarz, I., & Gebski, V. (2005). Randomised controlled trial of the Lidcombe Programme of early stuttering intervention. British Medical Journal, 331, 659-661.
3. Lewis, C., Packman, A., Onslow, M., Simpson, J.A., Jones, M. (2008). A phase II trial of telehealth delivery of the Lidcombe Program of Early Stuttering Intervention. American Journal of Speech Language Pathology, 17, 139-149.

Commentary by Elaine Kelman, Head Speech and Language Therapist at The Michael Palin Centre:

Elaine KelmanThis ongoing study of 1,910 children in Melbourne, Australia, caused a significant stir when the investigators published their paper last year, reporting their findings on 142 children who stammer up to the age of 4 years.

The popular press responded in typical fashion, selecting snippets of information from which they created dramatic and misleading headlines, such as ‘Pre-schoolers’ Stuttering Not Harmful’ (USA Today) and ‘Children who stutter do not suffer disadvantage at school’ (Daily Mail). This was extremely unfortunate, given that the study’s most significant finding, namely that many more children experience stammering and fewer of those stop within a year than was previously thought, became somewhat buried. There followed a flurry of exchanges, beginning with the Stuttering Foundation of America’s ‘A Blunder from Down Under’ electronic correspondence between the authors and commentators Joseph Donaher and Ellen Kelly in Pediatrics (read at; the Michael Palin Centre’s ‘Good news? Bad news? There is such a thing as bad publicity’ (; verbal presentations on the Stutter Talk podcast (; and most recently an article by Ehud Yairi in the Stuttering Foundation’s winter newsletter ( In this article, I will attempt to outline the key areas of debate that have arisen.

The primary concern among professionals arose from the potential unintended consequence of the above media headlines, that “parents may be discouraged from seeking advice, doctors will assure them that the child will be fine and the opportunity for early intervention will be lost” (Kelman).

There was also concern that the findings indicated that children in this sample "showed little evidence of harm to their mental health, temperament, or psychosocial health-related quality of life". Donaher and Kelly questioned the authors’ interpretation of this data and pointed out the importance of clinicians continuing to evaluate and address the psychological wellbeing and emotional reactions of a child who stammers. There is no question that stammering can have an impact on these areas from an early age in some children and this is often the reason why parents seek therapy for their child. One of the limitations of large group studies is that the results of a minority can be lost in the process of exploring the group average.


It is also important to note that the population reported on by Reilly and colleagues is not a clinical population. It represents a population of all children who start to stammer, rather than those who stammer severely or are concerned enough to seek therapy. So for therapists, it is important to understand that the results and recommendations of this study do not necessarily transfer to the child in the clinic and therefore Donaher and Kelly are right to state that the impact of the stammer should continue to be considered.

Yairi celebrated the excellent use of a “good-size” longitudinal sample, representing the general population of an area, starting in very early childhood and employing multiple variables. He expressed concerns about previous studies that were not referred to, pointing out that the findings regarding incidence, the children’s superior language skills and their temperament characteristics had been reported in articles dating back to 1957. In response, the authors stated that this was due to insufficient space and the different nature of this study.

The 12 months recommendation

There have also been concerns about the authors’ suggested guidelines for intervention.  Firstly, relating to the recommendation to delay treatment for 12 months unless the child is distressed, there is parental concern, or if the child becomes reluctant to communicate. Some of those responding to the article and ensuing publicity were worried that parents and other professionals will focus on the ‘wait and see’ aspect of this recommendation, rather than the proviso, to be watchful and commence intervention if there are certain indicators. The second concern was the recommendation that when therapy is indicated it should be the Lidcombe Programme, which is not the only evidence-based approach and is not the only option. In the discussion that followed, the authors emphasised their original point that these recommendations arose from the original research into the Lidcombe Programme, not from this epidemiological study.

This is a longitudinal study which will help us to understand more about the development of stammering over time; the factors that contribute to the disorder; and the impact that it has for different age groups. It is exciting that this new research stimulates debate which subsequently encourages accountability and open-mindedness and above all moves us forward as we seek to understand, support and ultimately help to improve the lives of children who stammer.

From Speaking Out Spring 2014, pp.14-15