5 school-aged children chatting in a corridor

Stuttering: what do we mean by ‘recovery’?

We’ve been asking ourselves an important question recently: What, exactly, does ‘recovery’ from stuttering mean? 

We work with toddlers, preschoolers, primary school-aged children, teenagers, and adults who stutter. We work with clients with mild stuttering; and others with severe stuttering. We work with clients who stutter consistently with all people in all situations, others who stutter only in some situations (e.g. when reading, or talking on the phone, or during presentations), and yet others who stutter only occasionally (e.g. when tired, angry or nervous). 

We listen to our clients, read the peer-reviewed research, and reflect on our clinical data. And we still struggle to answer the most basic of questions: 

When can we say a client has recovered from stuttering?

Fortunately, a group of researchers has been looking at this question recently in the context of early stuttering (Einarsdóttir et al., 2020 – see below for full citation).  

Before we get to their interesting findings, let’s revisit some of what the evidence tells us (and doesn’t tell us) about stuttering and recovery:

Not everyone recovers

Stuttering usually starts in early childhood. Most children recover. But a sizeable minority don’t. 

No cure for persistent stuttering

For children (and teenagers and adults) who do not recover in the preschool years, stuttering can be persistent – even with good, evidence-based treatment and improvements in fluency. To date, there is no cure for persistent stuttering (although there are some very smart people working hard on it). Treatment goals often involve learning coping strategies to manage stuttering and its effects.

Predictors of recovery

Can we predict who will recover and who won’t? Some researchers have identified a few factors that seem to increase the chance of recovery, including being a girl. More recent research suggests that children who start to stutter at earlier ages are more likely to recover; and that children who have been stuttering for less than 18 months are more likely to recover than those who stutter for longer than 18 months. Children who receive treatment within 15 months of stuttering onset are more likely to recover than those who receive treatment later. Children with advanced speech, language and literacy skills may be more likely to recover than those with delayed or disordered speech or language (e.g. Einarsdóttir et al., 2020.)

In truth, we don’t know whether a particular child will recover. This is why we recommend early treatment for all young children who stutter. 

Some reported recovery rates in published research might be inflated

As we’ll discuss below, recovery rates depend a great deal on your definition of ‘recovery’. But there are a few other factors that probably inflate some of the published findings: 

(a) Many families drop out of research studies and therapy before finishing

  • One frequently cited study reported a recovery rate of of 79.2%, but 33.2% of the participants in that study dropped out (Andrew & Harris, 1964), making it hard to interpret. We see high drop out rates in many stuttering studies, most notably in some of the Westmead Program studies.
  • Many families drop out of speech therapy before completing treatment. Sometimes, it’s because they are satisfied with their child’s fluency – they decide that the child is recovered or at least ‘fluent enough’ or that stuttering is no longer affecting the child is a material way. But, at other times, it’s because they don’t think the therapy is working, or that the costs of continuing with therapy exceed the benefits.

If we only look at people who finish treatments, we may think we are more successful in treating early stuttering than we actually are.

(b) Unrepresentative participants used in some (not all) studies

Many (but not all) peer-reviewed studies of children who stutter exclude children with speech-language disorders, attention disorders, Autism Spectrum Disorders, and intellectual disabilities. Many of my clients who stutter have speech and language challenges as well. If advanced speech and language skills are possible predictors of recovery, aren’t we at risk of reporting inflated recovery rates because we are excluding children with delayed or disordered speech? 

There are other problems with inclusion criteria used in some studies, too. Many studies, for example, exclude bilingual children. My case load includes many children who are learning more than one language. Are recovery rates different for bilingual children? We don’t know enough about this yet.

(c) Limited long-term follow up of ‘recovered’ children into adulthood

Few studies follow people who stutter from preschool to adulthood, middle-age and late ages. Children who complete preschool treatments and are deemed to have ‘recovered’ may relapse or otherwise start stuttering at some later point. We see lots of older clients who report they were treated successfully as preschoolers.

Some clients learn to mask their symptoms successfully

Low levels of stuttering can be hard to distinguish from the speech of people who do not stutter. If a client stutters so mildly that no one can tell, are they ‘recovered’? If not, why not?

Many people who stutter find effective ways to self-manage their symptoms. For example, some of my clients:

  • gravitate to jobs, hobbies and interests that don’t require much speech;
  • replace or work around potentially difficult speaking situations with other kinds of communication:
    • using an app to order food rather than the phone;
    • using picture or text-based social media, messaging and blogging platforms like Instagram, Messenger, WhatsApp, Discord or Medium to express themselves;
    • using gestures, gifs, infographics and other visuals to supplement their words in presentations;
  • reduce the social pressure of speaking by pre-scripting or rehearsing important speeches, or pre-recording video responses, rather than presenting them live; or
  • work around particular difficulties by substituting words they find difficult to say with synonyms, or simplifying the complexity of their sentences to reduce the length of each utterance. 

Adults who stutter are at a greater risk of developing social anxiety and other mental health issues. But many people who stutter do not have social anxiety or other problems and lead very full, healthy lives. Do we count clients who stutter but mask their stuttering symptoms and participate in life effectively ‘recovered’? If not, why not?

Does the definition of ‘recovery’ depend on the goal of stuttering therapy?

Is the goal of therapy to get rid of the stuttering – to ‘fix the impairment’ –  or to help clients manage the effects stuttering on their participation and mental health and maximise quality of life outcomes? Do we measure recovery differently depending on the goal?

For preschoolers who stutter, there are very good, evidence-based treatments, including the Lidcombe Program and the Westmead Program. The stated aim of these programs is to eliminate stuttering. 

Once a child starts school, stuttering gets much harder to treat and is rarely eliminated altogether. The treatment focus switches away from the elimination of stuttering to helping children self-manage their stuttering in situations that matter most to them, like talking in class and in the playground, giving speeches, performing in school plays or in sports. 

For teenagers and adults, speech restructuring or prolonged speech treatments, like the Camperdown Program have the highest level of evidence. But, again, they don’t work for everyone. For example, in some situations, some clients would rather stutter than use speech restructuring techniques.

Over the years, we’ve observed that stuttering affects people in very different ways. For some clients, it is a really big deal that has a profound, negative effect on their mental health, social participation, and school or work goals. For other clients, stuttering seems to have little obvious effect on their lives. 

Stuttering severity alone is not always the main issue. Some people seem to be fairly untroubled by even quite severe stuttering. For others, very mild stuttering can have a big, negative effect on their quality of life and health.  

Is recovery determined by the reduction of stuttering severity or symptoms, by the quality of the client’s life despite the effects of stuttering, or a combination of factors?

The answer must be different for different clients.

What does ‘recovery’ from early stuttering really mean? 

In a recent study, Einarsdóttir and her colleagues reviewed 23 longitudinal studies to examine the persistence and/or recovery rates in children who stutter. They found a mean recovery rate of 58.7% across studies. But they also found reported recovery rates ranging from 6.3% to 94%! 

One reason for this startling inconsistency across studies is that different stuttering researchers defined ‘recovery’ very differently. Some:

  • used speech measures, like percentage of syllables stuttered and/or a subjective rating system (e.g. from 0-9, 0-7, or 1-10);
  • relied on expert opinion from experienced speech pathologists; 
  • relied on self-reports from families and from children themselves; and
  • used a combination of measures.  

In defining recovery, some researchers used stringent measures – e.g. requiring no stuttering whatsoever measured by experienced clinicians across multiple speech tasks. Others allowed some stuttering (e.g. less than 3% of syllables stuttered, or severity ratings of 2 or less on a rating scale of 0-7). Others used purely subjective measures like family reports, or didn’t specify their definition of recovery at all. Far too few studies asked the child to report whether they felt like they had recovered. 

All this inconsistency makes it very difficult to give families good information about the ‘true’ recovery rate or even what recovery might look like. To help improve the evidence-base, Einarsdóttir and her colleagues studied 38 monolingual children who stuttered, measuring stuttering on two occasions: (1) when they were aged 2-5 years; and (2) seven years later, when they were 9-13 years of age. They applied four different criteria for the definition of ‘recovery’, based on previous research:

(A) A child has a stuttering frequency of less than 3% syllables stuttered in video recorded speech samples.

(B) A child meets all four of the following criteria: (i) two speech pathologists and parents agree that the child no longer stutters; (ii) parents had to be able to report when the stuttering stopped and confirmed that the child had not stuttered for the previous 12 months; (iii) stuttering severity was rated as ‘borderline or no stuttering’ by speech pathologists based on the speech sample; and (iv) the average percentage of stuttered syllables was 3% or less. 

(C) A child meets all the criteria in (B) + the child self-reported that he was no longer stuttering.

(D) A child meets all the criteria in (C) and no stuttered speech at all was present in the speech sample.

They found that all children became more fluent with age. But the recovery rate varied greatly depending on the criteria used. If:

  • criteria (A) were applied, 94.7% recovered;
  • criteria (B) were applied, 71.1% recovered;
  • criteria (C) were applied, 55.3% recovered; and
  • criteria (D) were applied, only 13.2% recovered (5/38 children).

So the rate of recovery ranged from 13.2%-94.7%, depending on the criteria applied – broadly consistent with the results of their analysis of previous longitudinal studies. The most common criteria applied in previous research is (B). But, interestingly, by including children’s self-reports on whether they felt they had recovered (Criteria C), far more children were classified as not-recovered. 

A few caveats to this study: It was fairly small – 38 children (28 male, 10 female). All participants lived in Iceland. The study excluded bilingual children and children where ‘there was evidence of severe language impairment’ or signs of ‘neurological disorder’. The first author was not blinded to which clients identified as recovered. Selection bias may have affected recruitment. The children were only assessed at two points, and in clinical contexts (mitigated by obtaining multiple speech samples on different tasks). Importantly, some children were treated during the study term; and others were not, so no conclusion could be made on whether treatment may have affected outcomes.

Clinical Bottom line

When we talk about ‘recovery’ from stuttering, we need to be clear on exactly what we mean; clear about what the evidence tells us about recovery rates for early and persistent stuttering (including what we don’t know, and some of the limitations of the published studies); and clear about the goals of stuttering therapy.

Recovery means different things to different people – including to different people who stutter. For many clients, requiring completely fluent speech as a criteria for recovery is unrealistic and potentially demotivating and demoralising. Ultimately, the most important view is the client’s, recognising that there is as yet no cure for persistent stuttering.

Related articles:

Principal source: Einarsdóttir, J.T, Crowe, K., Kristinsson, S.H., & Másdóttir, T. (2020). The recovery rate of early stuttering. Journal of Fluency Disorders, 64, 105764.

Man wearing glasses and a suit, standing in front of a bay

Hi there, I’m David Kinnane.

Principal Speech Pathologist, Banter Speech & Language

Our talented team of certified practising speech pathologists provide unhurried, personalised and evidence-based speech pathology care to children and adults in the Inner West of Sydney and beyond, both in our clinic and via telehealth.

David Kinnane
Speech-Language Pathologist. Lawyer. Father. Reader. Writer. Speaker.

Share This

Copy Link to Clipboard

Copy