Journal of Fluency Disorders

28 (2003) 237–245

Evidence-based treatment of stuttering:

IV. Empowerment through evidence-based
treatment practices夽
Mark Onslow∗
Australian Stuttering Research Centre, The University of Sydney,
P.O. Box 170, Lidcombe, NSW 1825, Australia

Received 3 April 2003; received in revised form 24 April 2003; accepted 29 April 2003

Abstract

Assertion-based treatments for stuttering have historically been more popular than
evidence-based treatments. In this paper it is argued that the use of evidence-based treatments
for stuttering is professionally empowering for clinicians, but that the use of assertion-based
treatments is a circular process that inhibits professional development. The arguments in fa-
vor of evidence-based treatment are elaborated under headings of “professional investment,”
“professional development and diversity,” and “optimizing treatment efficacy.”
Educational objectives: The reader will understand and be able (1) to describe the dis-
tinction between assertion-based and evidence-based treatment practices (2) to present a
series of arguments that evidence-based treatment practices are professionally empowering.
© 2003 Elsevier Inc. All rights reserved.

Keywords: Treatment efficacy; Assertion-based treatment; Recovery plots; Randomized controlled

trials

1. Introduction

If anything stands out in the history of our field, it is the popularity of asser-
tion-based treatments for stuttering: a system based on belief or theory that

夽 This paper expands on a presentation made at the 2002 ASHA Annual Convention.
∗ Tel.: +61-2-9351-9767; fax: +61-2-9351-9392.
E-mail address: [email protected] (M. Onslow).

0094-730X/$ – see front matter © 2003 Elsevier Inc. All rights reserved.
doi:10.1016/S0094-730X(03)00041-X
238 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245

certain treatments are effective, rather than on the basis of empirical evidence.
The popularity of this line of thought with our treatment methods for children
was dramatically underscored in a recent edition of the Journal of Fluency
Disorders, which contained a forum concerning the popular multifac-
torial Demands and Capacities treatment model. As Siegel (2000) noted in his
opening of the forum, much of our current treatment for children is influenced
by this model. In the course of the subsequent discussion by clinical figures in
our field (Bernstein Ratner, 2000; Curlee, 2000; Kelly, 2000; Manning, 2000;
Starkweather & Gottwald, 2000; Yaruss, 2000) there was clearly no writer who
felt the treatment should be discarded. It is testament indeed to the popularity
of assertion-based treatment in our profession that its leaders could be so com-
fortable with a treatment for early stuttering for which there has not been pub-
lished a single, prospective trial of any kind that incorporates objective speech
data.
It is clear that health professions worldwide are distancing themselves from
assertion-based treatment justification. In Australia, for example, the governing
health body is the National Health and Medical Research Council. That organiza-
tion has recently specifically rejected expert opinion as a means to justify treat-
ment choice, insisting instead on empirical means (National Health and Medical
Research Council of Australia, 2000). In reality, such political pressures — which
certainly will not abate — are sufficient reason alone for rejecting assertion-based
practice in favor of evidence-based practice. Put simply, health care practices that
do not depend on science for their substance will eventually be disempowered.
They certainly will not receive the level of financial support from governments
that they have in the past.
However, in the case of stuttering treatments, the assertion-based system has
worked so well for so long, for so many, and has so many adherents, that it seems
a little hollow to argue that political pressure alone should lead to its downfall.
Are there reasons other than political ones that can compel its rejection? What will
we benefit if we make the change to an evidence-driven profession? If we do, our
profession will surely endure, but will it be a better one for the change? These are
the matters that are discussed in the following, under three headings: “professional
investment,” “professional development and diversity,” and “optimizing treatment
efficacy.” The paper concludes with an attempt to unify these arguments under the
heading of “professional empowerment.”
First, though, one more piece of prefatory material. The following contains
several references to “outcomes” and “efficacy.” The measurement of outcome and
efficacy is integral to evidence-based practice. However, in the following, there is
no intended implication about the value of any particular measure of outcome or
efficacy — communication, attitude to communication, stuttering severity, speech
rate, speech naturalness, and so on. The topic of what might be suitable measures
of stuttering treatment outcome and efficacy is certainly important, but it is beyond
the scope of the present discussion.
 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245 239

2. Professional investment

One way of thinking about our treatments for stuttering is that they are profes-
sional assets; things of value that are hard won, but worth having. As such, they
are worth investing in. Few of us would not make personal investments during our
lifetimes, so that the assets that emerge from that investment make our lives better.
So, if we would invest in our lives to build personal assets, why would we not
invest in our profession to build professional assets? Considering the major part
of life occupied by a profession, and the time taken to obtain professional qualifi-
cations, surely that warrants an investment for a professional future. And, perhaps
more importantly, surely it is worth an investment for the clinicians of the future
that have not yet joined our profession. A commitment to evidence-based clinical
practice with stuttering is just that. It is an investment in real potential growth
in the fundamental assets of our profession; an investment in the development of
efficacious treatments for our professional lives and for the professional lives of
those who follow us.
At the time of writing, share markets around the world are shaky indeed, and
the potential growth of an investment there is by no means predictable. More
confidence can be placed, however, in the potential for growth in the science un-
derpinning stuttering treatment. That potential growth is clear and it is predictable,
which is ideal for an investment. There is some guarantee of a return perhaps for
the immediate future, but certainly for the long term. This should be true for our
own professional lives over the next decade and for our colleagues of the future
who will replace us. Like all investing, what is required here is vision, and, more
importantly, patience. Patience is particularly important in the case of a profes-
sional investment in evidence-based practice. For there is no guarantee that we
will see the great many benefits of such an investment during our professional
lifetimes.
This is so because new treatments, and improvements on existing treatments,
come slowly. Judging from the work of my colleagues and myself, it appears that
around 10 years is required from the conceptualization of a new treatment to its
availability for the use of clinicians after publication in our scientific journals.
Consider the Lidcombe Program. This treatment was conceptualized during the
mid-1980s, but sufficient evidence to warrant its use in Australian clinics had only
emerged by the mid-1990s (Onslow, 2003). And, it should be added, that was the
time required for the emergence of the most rudimentary outcome data imaginable.
According to the guidelines mentioned previously (National Health and Medical
Research Council, 2000), the strongest evidence comes from a review of many
randomized controlled trials. However, we are a long way yet from such luxury
with our stuttering treatments, as we await the publication of the first random-
ized controlled trial of basic speech pathology procedures for the control of stut-
tered speech. However, the laborious process of assembling scientific evidence
in stuttering treatment has begun in our profession. True, it is slow, and it is for
240 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245

tomorrow. But it is unstoppable now it has started. And as such, it is a good

investment.
What returns can we, or the generations of our future clinicians, look forward to
as returns in an investment in evidence-based clinical practice? Nothing less than
clinical solutions to the problems that confront our clinicians each day in the clinic.
As an example of what might come in the future, consider the contents of some of
our recent journals: Roger Ingham and colleagues’ demonstration of the clinical
potential of a machine driven means to control stuttered speech (Ingham et al.,
2001), and Bray and Kehle’s demonstration of the benefits of self-modelling in
stuttering therapy (Bray & Kehle, 1998, 2001). If those preliminary reports reach
fruition, the benefits will be extraordinary. We will have effective stuttering treat-
ments that can be used by clients with reasonable independence from a clinician.
If nothing else, science is for the future. Not only is its growth unstoppable, its
growth is exciting. Evidence-based clinical practice allows an investment in that
future excitement.

3. Professional development and diversity

It is certainly an option for a clinician to constrain professional life to the pre-

sentation of treatments for stuttering. However, evidence-based clinical practices
provide a means to allow the clinicians of our profession to grow into more than
this if they wish. There are two ways that evidence-based practice can provide
such professional development, and these are considered below.
The first is to be a scientist practitioner. Simply, evidence-based practice can
foster improvements in clinical performance. This cuts at the heart of the problems
with assertion-based clinical practice: It can’t improve clinical performance. If a
clinician believes and asserts that a treatment is effective, that is just about as far
as professional development goes. Assertion-based clinical practice is about the
clinician, and what the clinician believes. It is not about the clients. By definition,
a clinician cannot under achieve in assertion-based clinical practice. Assertion-
based clinical is endlessly circular: (1) the clinician asserts that the treatment is
acceptable, (2) the clinician does the treatment, (3) therefore, the clinical practice
is acceptable. That cycle can continue for an entire career, cocooning the clini-
cian, by definition, against the prospect of ever being wrong or needing to improve
performance in the clinic.
However, an evidence-based clinical practitioner — a scientist practitioner —
can break that cycle. Evidence-based treatment is not about clinicians; it is about
their clients getting better. As such, there is a mechanism for clinicians to know
quickly if they are under performing. Clinical outcomes are measured. And here
is the source of professional growth from evidence-based practice; if a clinician is
under achieving a clinician will know. If clients do not do as well as science says
they should, then the clinician will know. Perhaps, it may be intimidating to be
placed in a position where any clinical inadequacies will be apparent, but a clinician
 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245 241

who has the interests of clients at heart — rather than his or her own beliefs —
would do it gladly. And those who do it are the true advocates of evidence-based
practice. They are scientist practitioners.
There are various ways that the scientist practitioner can use the evidence base
of clinical practice to gauge acceptability of clinical performance. One method
is to surmise from the literature that a certain treatment is best in a certain case
scenario, and then expect clients to get better because the evidence says that this is
what should occur. And then, using speech measures similar to those that appear
in the evidence-based literature, the scientist practitioner determines whether or
not the client is improving as science says the client should. If the client fails to
improve, then clinical problem solving is necessary to fix the problem. There are
plentiful data, for example, to show how quickly children treated with the Lidcombe
Program should improve (for example, Jones, Onslow, Harrison, & Packman, 2000;
Kingston, Huber, Onslow, Packman, & Jones, 2003; Onslow, Harrison, Jones, &
Packman, 2002). In short, children start to reduce their stuttering severity in 3–5
weeks, and 50% of them are not stuttering after 11 clinic visits to the clinician
with their parents.
In the case of treatments for adults, similar data are available. There are some
published scientific reviews of the literature that can assist the scientist practitioner
to actually know how long it should take for clients to get better (for example,
Cordes, 1998; Ingham, 1984). In fact, it is mandatory practice for recent treatment
reports to include treatment time as a dependent variable. For example, O’Brian
and colleagues’ recent report of treatment times with the Camperdown Program
showed a mean of 20 h for control of stuttered speech with this prolonged-speech
technique (O’Brian, Onslow, Cream, & Packman, in press). So, if a clinician takes
a great deal of more than 20 h to achieve the same goal with this treatment, there
may be a need to improve clinical performance.
There is an even more sophisticated guide for the scientist practitioner: a re-
covery plot analysis of caseloads. For example, such plots were presented in the
reports of Jones et al. (2000) and Kingston et al. (2003), totalling 328 children
who were treated with the Lidcombe Program. An advantage of this method is
that it contains no dependent variables. It only graphs time to reach treatment
criteria and the number of clinic sessions to get there. The recovery plot charts
only time to criterion speech performance, whatever that might be. So a recovery
plot can be constructed for every different treatment that might be used in the
clinic.
Recovery plots are a powerful tool for professional development and diversity.
It is common, for example, for the Lidcombe Program clinicians to become con-
cerned when they arrive at the 15th clinic session and the child is not clearly about
to stop stuttering during the next few weeks. The scientist practitioner can also use
this kind of recovery plot to be guided by science in benchmarking with colleagues.
Is the clinician performing as well as colleagues? Or perhaps a clinician has just
entered the workforce, and wishes to compare performance with more experienced
colleagues. A recovery plot can be constructed from as few as 20 cases that have
242 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245

been treated with one method. Then it is possible for a clinician to determine his or
her performance, and that of an evidence-based clinical community, and compare
it to the published benchmark. It is even possible to do a simple statistical test —
the Logrank statistic (Altman, 1991) — to determine whether any two recovery
plots are significantly different from each other.
In fact, it is possible to do more than be a scientist practitioner to develop and di-
versify professional activities. The benchmark data in the two publications referred
to previously — Jones et al. and Kingston et al. — were generated by professional
clinicians in public health systems in Australia and the United Kingdom. They
linked up with researchers and had their data published. They developed profes-
sionally to the point where they were able to convert their day-to-day clinical work
into publications in peer reviewed, international journals. They are clinician re-
searchers, and I have to say that I find them, and others in our literature, a source of
inspiration. Surely they are exemplars of professional development and diversity
in our field, to move from providers of treatments to providers of treatments and
publishers of data in the international literature.
Surely this type of professional development is critical for us as we inch our
way toward knowing how to treat stuttering in clients of all ages, for the speech
clinic should be a prime source of profitable clinical ideas that make it to pub-
lished outcome data. Certainly, this does not mean that every clinician could step
effortlessly into the role of clinical researcher. In fact that is most unlikely to be
the case. A clinician needs to be willing to step into that role, to have the right sort
of clinic, the right sort of caseload, and a clinician needs to be in a position where
productive partnerships can be forged with professional researchers. There must
be clinicians who have the potential to make that ultimate professional develop-
ment and diversity, and, in the long term, our field will be immeasurably better off
for it.

4. Effective treatments

It seems natural for clinicians to want to provide the best possible treatment
to those who trust their health care to them. Now, our profession is backed by
many scientific journals that publish outcome and efficacy data for stuttering treat-
ments. This assertion might seem a bit controversial at first, but really it is lit-
tle more than self-evident: The outcome and efficacy data from those journals
is the evidence base of our work with those who stutter, and that is the place
to find the best treatments for our clients. Simply, to not use our scientific evi-
dence base as a source of stuttering treatment does not provide the best quality of
service.
Can that be correct? Can it be that the extensively popular treatments that
have emerged from the assertion-based literature are not the most efficacious
for our clients? It is a new mindset to accept this. For our past and quite re-
cent history of treatment contains examples of charismatic assertions about
 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245 243

treatment, particularly in the case of early stuttering. The Demands and Capacities
treatment, referred to earlier, is certainly an example of that, as is the work of Wen-
dell Johnson and Charles van Riper, and the currently popular Maguire Program
(http://www.mcguire-freedomsroad.com). But could it be that such assertions are
not the route to the best treatments, and that, ergo, they should be abandoned now,
forever, in favor of an evidence-based approach?
In considering this serious matter, attention needs to be focused on the dis-
tinction between charismatic and effective. The past and present assertions about
stuttering treatment are certainly charismatic. Who, for example, could resist the
notion that the origins and perpetuation of the disorder arise from a myriad of
environmental stressors that the clinician must diagnose and rectify? Not many, if
our history, and the recent Journal of Fluency Disorders forum (see above) is any-
thing to go by. But all that aside, is that treatment effective? That is what clinicians
really want to know as they strive to find what is best for their clients. They want
to know the most efficacious treatments, not necessarily the most charismatic. Can
assertions about treatment practices, for all their charisma, be trusted to specify
the most efficacious treatment for a client? The answer is “no,” for the following
reason.
I think that this history of stuttering treatment shows that assertions about
treatments have followers — believers — and an assertion goes from strength
to strength, like a faith, a critical mass of itself, in direct proportion to the number
of believers. For all that such self-perpetuating movements are charismatic, they
are simply not driven by concerns about efficacy.
And there is the bottom line: The search for the most effective treatments is not
about clinicians or what clinicians believe, it is about our clients. It is not about the
charismatic nature of assertions about how to do treatment, and it is not about the
number of followers of that assertion. Evidence-based practice is about the clients
getting better from their disorder.

5. Professional empowerment

In summary, then, evidence-based treatment is an investment for the future

of our profession, and it provides a means of professional development and di-
versity and ensures that clinicians choose the most effective treatments for their
clients. Another way of saying this is that evidence-based treatment practices
can be professionally empowering for clinicians in their quest to treat those who
stutter. The importance of such professional empowerment to our future can
be seen by considering the opposite of that picture of empowerment: A pro-
fession that has no investment in its future treatment practices, has no means
for the development and diversity of its clinicians, and has stuttering treatment
practices with unknown efficacy. If that is our profession, what do we have?
Without evidence-based practices we have no future. Only an endless and un-
accountable cycle of assertion-based treatments for stuttering, and the handing
244 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245

down of them, essentially unchanged, from one generation of clinicians to the

next.

References

Altman, D. G. (1991). Practical statistics for medical research. New York: Chapman and Hall.
Bernstein Ratner, N. (2000). Performance or capacity, the model still requires definitions and boundaries
it doesn’t have. Journal of Fluency Disorders, 25, 337–348.
Bray, M. A., & Kehle, T. J. (1998). Self-modeling as an intervention for stuttering. School Psychology
Review, 27.
Bray, M. A., & Kehle, T. J. (2001). Long-term follow up of self-modeling as an intervention for
stuttering. School Psychology Review, 30, 135–141.
Cordes, A. (1998). Current status of the stuttering treatment literature. In A. Cordes & R. J. Ingham
(Eds.), Treatment efficacy in stuttering: A search for empirical bases. San Diego, CA: Singular
Publishing Group.
Curlee, R. F. (2000). Demands and capacities versus demands and performance. Journal of Fluency
Disorders, 25, 329–336.
Ingham, R. J. (1984). Stuttering and behavior therapy: Current status and experimental foundations.
San Diego, CA: College-Hill Press.
Ingham, R. J., Kilgo, M., Ingham, J. C., Moglia, R., Belknap, H., & Sanchez, T. (2001). Evaluation of a
stuttering treatment based on reduction of short phonation intervals. Journal of Speech, Language,
and Hearing Research, 44, 1229–1244.
Jones, M., Onslow, M., Harrison, E., & Packman, A. (2000). Treating stuttering in children: Predicting
treatment time in the Lidcombe Program. Journal of Speech, Language, and Hearing Research,
43, 1440–1450.
Kelly, E. (2000). Modeling stuttering etiology: Clarifying levels of description and measurement.
Journal of Fluency Disorders, 25, 359–368.
Kingston, M., Huber, A., Onslow, M., Jones, M., & Packman, A. (2003). Predicting treatment time
with the Lidcombe Program: Replication and meta-analysis. International Journal of Language
and Communication Disorders, 38, 165–177.
Manning, W. H. (2000). Appeal of the Demands and Capacities Model: Conclusions. Journal of Fluency
Disorders, 25, 377–383.
National Health and Medical Research Council of Australia. (2000). How to evaluate the evidence.
Canberra, ACT, Australia: Author. Retrieved February 25, 2000, from http://www.nhmrc.gov.
au/publications/synopses/cp30syn.htm.
O’Brian, S., Onslow, M., Cream, A., & Packman, A. (in press). The Camperdown Program:
Outcomes of a new prolonged-speech treatment model. Journal of Speech, Language, and Hearing
Research.
Onslow, M. (2003). From laboratory to living room. In M. Onslow, A. Packman, & E. Harrison
(Eds.), The Lidcombe Program of early stuttering intervention: A clinician’s guide. Austin, TX:
Pro-Ed.
Onslow, M., Harrison, E., Jones, M., & Packman, A. (2002). Beyond-clinic speech measures during
the Lidcombe Program of early stuttering intervention. Acquiring Knowledge in Speech, Language
and Hearing, 4(2), 82–85.
Siegel, G. M. (2000). Demands and capacities demands and performance? Journal of Fluency
Disorders, 25, 321–327.
Starkweather, C. W., & Gottwald, S. R. (2000). The demands and capacities model: Response to Siegel.
Journal of Fluency Disorders, 25, 369–375.
Yaruss, J. S. (2000). The role of performance in the demands and capacities model. Journal of Fluency
Disorders, 25, 347–358.
 M. Onslow / Journal of Fluency Disorders 28 (2003) 237–245 245

CONTINUING EDUCATION
Evidence-based treatment of stuttering: IV. Empowerment through evidence-
based treatment practices
QUESTIONS
1. Assertion-based treatment practices:
a. have been proven by expert opinion to be correct
b. are the opposite of evidence-based treatment practices
c. guide the development of clinical trials research
d. are argued by the writer to be a good means of professional development
e. are argued by the writer to be linear rather than circular
2. Recovery plots:
a. have never been used in stuttering treatment research
b. can be used with any treatment because they have no dependent variable
c. cannot be tested statistically
d. have appeared in the demands and capacities model research
e. are an effective substitute for a randomized controlled trial
3. A series of randomized controlled trials is:
a. a long way from completion in stuttering research
b. the second highest level of possible evidence for a treatment
c. underway with the Demands and Capacities Model treatment
d. compatible with the notion of assertion-based treatment
e. possible only with treatments for adult stuttering
4. The author argues that evidence-based treatment is:
a. circular
b. of equal value to assertion-based treatment
c. an excellent means for developing treatments for early stuttering but not
advanced stuttering
d. politically in favor
e. self-perpetuating and circular
5. The author argues that the development of empirical treatments for stuttering:
a. can occur only with concurrent development of assertion-based treatments
b. is a slow process
c. can never bring immediate benefits to modern clinicians
d. leads to a cycle of assertion-based treatments
e. relies heavily on statistical methods