A theory building critical realist evaluation of an integrated cognitive-behavioural fluency enhancing stuttering treatment for school-age children

Journal of Fluency Disorders 82 (2024) 106076
Contents lists available at ScienceDirect
Journal of Fluency Disorders

journal homepage: www.elsevier.com/locate/jfludis
A theory building critical realist evaluation of an integrated

cognitive-behavioural fluency enhancing stuttering treatment for
school-age children. Part 1: Development of a preliminary
program theory from expert speech-language pathologist data.
Michelle C. Swift a, c, *, 1 , Marilyn Langevin b, 2
a
UniSA Allied Health and Human Performance, University of South Australia, GPO Box 2471, Adelaide, SA 5001, Australia
b
Department of Communication Sciences and Disorders, Faculty of Rehabilitation Medicine, University of Alberta, 8205 114 Street, 2–70 Corbett
Hall, Edmonton, AB T6G 2G4, Canada
c
College of Nursing and Health Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia
A R T I C L E I N F O A B S T R A C T
Keywords: Purpose: This study initiated a program of research that aims to develop a program theory un-
Critical realist evaluation derlying integrated cognitive-behavioural fluency enhancing stuttering treatments for school-age
Evidence-based practice children. This research asks, what in the treatment program works (or does not work), for whom,
Stuttering
in what contexts, and why.
Integrated stuttering treatment
Methods: Using a critical realist evaluation approach, seven speech-language pathologists (SLPs)
School-aged children
with extensive experience in treating children who stutter were asked about barriers and facili-
tators of optimal treatment outcomes within the context of the Comprehensive Stuttering Pro-
gram - School-aged Children (CSP-SC). From these data discrete resource mechanisms, contexts,
within child reasoning mechanisms, and outcomes were derived and a preliminary program
theory was proposed.
Results: Facilitating and impeding child physiology, treatment and SLP resource mechanisms,
family and school contexts, and within-child mechanisms were identified. Facilitating mecha-
nisms included motivation, personality/psychological characteristics, understanding and trust of
the treatment process, experience of speaking with less effort, and self-efficacy. Impeding
mechanisms included reduced motivation, impeding personality/psychological characteristics,
lack of buy-in, and, for some children, a prohibitive cost of effort in using learned strategies.
Conclusion: A preliminary program theory was hypothesized which will be further developed in
future analysis of data obtained from children and parents who participated in the CSP-SC at the
same centre from which the SLPs came. Subsequent research with new cohorts of SLPs, children,
and parents from other treatment programs and centres will be needed to establish the gener-
alizability of the program theory generated in this program of research
* Corresponding author at: UniSA Allied Health and Human Performance, University of South Australia, GPO Box 2471, Adelaide, SA 5001,
Australia.
E-mail addresses: michelle.swift@unisa.edu.au (M.C. Swift), marilyn.langevin@ualberta.ca (M. Langevin).
1
ORCID ID: orcid.org/0000–0002-0741–608X
2
ORCID ID: orcid.org/0000–0002-6153–4570
https://doi.org/10.1016/j.jfludis.2024.106076
Received 28 March 2024; Received in revised form 1 August 2024; Accepted 18 August 2024
Available online 24 August 2024
0094-730X/© 2024 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
M.C. Swift and M. Langevin Journal of Fluency Disorders 82 (2024) 106076
1. Introduction
Stuttering is characterized by overt physical behaviours (e.g., repetitions, sound prolongations and complete blockages of speech)
and covert features (e.g., avoiding words and speaking situations, experiencing speech related anxiety and a sense of loss of control of
speaking). Stuttering has the potential to profoundly disrupt communication and have significant negative psychological, emotional,
and social impacts that, if not treated appropriately, can last into adulthood. Negative social interactions start as young as 3 to 4 years
of age (Ezrati-Vinacour et al., 2001; Langevin et al., 2009; Langevin et al., 2010) and continue through school-age years into adulthood
(Hugh-Jones & Smith, 1999). Between 61 % and 73 % of school-age children who stutter (CWS) in the United States, Canada, Japan,
and Poland report being bullied more often than typically developing children, and, in the case of Polish children, more often than
children with other speech and language disorders (Wȩsierska et al., 2023). Impacts of bullying include high degrees of anxiety,
depression, hopelessness, suicidal ideation, and poor social adjustment (Langevin, 2015). It is thus imperative that stuttering in-
terventions be holistic. That is, treatments need to address the overt and covert features of stuttering and its psycho-social impacts
(Johnson et al., 2023a; Johnson et al., 2023b; Langevin & Kully, 2003; Langevin et al., 2007).
In designing treatment programs, speech-language pathologists (SLPs) must engage principles of evidence-based practice (EBP).
SLPs must bring together external evidence from the scientific literature, clinical reasoning from their own data and observations,
knowledge obtained from clinical expertise/expert opinion in the field, and learnings from clients and caregivers (ASHA, n.d.). Equally
important to the conduct of EBP is empirical evidence supporting the underlying theory upon which a program is based (Bernstein
Ratner, 2018).
Evidence from the scientific literature. In contrast to a growing body of research regarding the effectiveness of treatments for adults
and pre-school children, much less attention has been given to treatments for school-aged children and adolescents (Laiho et al., 2022).
A recent meta-analysis (Johnson et al., 2023b) found evidence for the effectiveness of treatments that use speech-restructuring in
isolation and in combination with cognitive behavioural and operant methods. Two recent studies with larger participant numbers
found evidence for speech-restructuring fluency shaping techniques using computer-based biofeedback with 6–9 year olds (Euler et al.,
2021) and stuttering modification treatment with 7–11 year olds (Kohmäscher et al., 2023). Both studies showed reductions in
stuttering frequency, and its impacts as measured by the OASES-S (Yaruss & Quesal, 2016).
The Comprehensive Stuttering Program for school age children (CSP-SC; Kully & Boberg, 1991; Langevin et al., 2007) is a holistic,
stuttering treatment program that integrates speech restructuring (i.e., fluency shaping or fluency enhancing techniques) and stut-
tering modification strategies with cognitive-behavioural strategies, self-management and environmental management strategies to
address the covert features of stuttering and its social impacts. The CSP-SC consists of an establishment phase during which
cognitive-behavioural strategies, self-management and environmental management strategies and fluency skills are taught, a transfer
phase during which these skills are practiced in real-life situations, followed by a maintenance phase during which the client continues
to practice and use the taught skills in their home environment (see supplementary material A). The CSP-SC was developed for children
aged 7 to 12 years. This is a broad age range with great differences in cognitive, emotional, social, and academic development. Thus,
treatment goals and programming are adjusted to meet the needs of the child and family at the time of initial therapy; programming
and goals change over time as the child matures and/or as needs and desires for treatment changes.
As with other integrated cognitive-behavioural fluency enhancing treatment programs for school-age CWS, the locus of change lies
within the child and parents are collaborators in the delivery of treatment. As such the CSP-SC also has a parent training component.
The current evidence base for the CSP-SC consists of a single group pre-post-treatment design with 10 clients at 8 – 18 months follow-
up (Kully & Boberg, 1991) and two pre-post-treatment design case studies with 5 children, with follow-up data obtained at 3 to 19
months post-treatment (Langevin & Kully, 2003; Langevin et al., 2007).
Clinical knowledge, clinical reasoning, and underlying program theory. SLPs can gather knowledge from existing tutorials that draw
from clinical expertise/expert opinion in treating school-aged CWS (e.g., Yaruss et al., 2012). However, clinical reasoning is not a fixed
attribute, but a skill that develops over time (Norman et al., 2007). Clinical reasoning develops from the implicit formulation, testing
and refining of hypotheses about which treatment works for what client in what context, coupled with critical reflection about the
accuracy of these hypotheses (Marcum, 2012). The issue is that in many cases this implicit reasoning and critical reflection is not
shared. In order to promote EBP across the speech-language pathology field, it is important that this expert knowledge be documented
and tested to see if the implicit hypotheses stand up to empirical testing. Equally important is the need to address the knowledge gap of
program theories that underlie stuttering treatment programs for school-aged CWS.
Realist evaluation. Realist evaluation is a program evaluation approach that codifies clinical reasoning and elucidates an underlying
program theory that explains why a program may or may not achieve its intended outcomes (Pawson & Tilley, 1997; Rameses, 2017).
It asks, “how interventions work, for whom (i.e., what subsets of the population), under what circumstances, contexts or conditions,
and why?”. It differs from RCT’s and other quantitative group designs that ask, “does a program work?” and consider context as a
source of bias. In realist evaluation context is “inextricably enmeshed with the mechanisms through which programmes work”
(Greenhalgh & Manzano, 2022, p. 590). The underlying premise of realist evaluation is that outcomes are influenced by multiple
interacting factors, that it is the action of participants in an intervention that makes a program work, and that stakeholders operate
within open ever-changing external social systems (Rameses, 2017).
Program outcomes or patterns of outcome regularities (O) are explained in the interaction of contexts (C) and mechanisms (M).
Mechanisms refer to a participant’s reasoning and the resources in a program that trigger such reasoning. Mechanisms lie at the heart
of generative causation; they are the unobservable underlying causal processes that determine whether and how a program works.
Mechanisms may not always be activated; they are “causal tendencies” for which activation depends on a supportive context (Pawson
and Tilley, 1997;Rameses, 2017). Outcome patterns are conjectured in CMO linkages or configurations that derive from program
2
theory evaluations. CMO configurations are hypotheses that attempt to explain how mechanisms operate in different contexts to
generate different outcomes; as such, they can be empirically tested. CMO configurations are not linear mathematical equations that
lead to a given outcome. Instead, they suggest how the likelihood of an outcome may be achieved and, as such, embrace the complexity
of the interaction between resources, multiple mechanisms, and contexts in complex interventions. Realist evaluation can be con-
ducted using realist meta-syntheses of relevant publications in the scientific literature, or using participant data, or both.
Realist evaluation is increasingly being used in the health sciences (e.g., Lemire et al., 2020, Nielsen et al., 2021). To date the use of
realist evaluation in speech-language pathology has been limited. Klatte and six colleagues (Klatte et al., 2020) used their expert
knowledge to develop a preliminary program theory of SLP and parent collaborative practices when working with children with speech
and language disorders. Langner and Fukkink (2023) developed four CMO configurations that documented how interprofessional
collaborative practices between SLPs and educators supported professional and child outcomes in early education and care settings.
Although not using realist evaluation, Caughter and Dunsmuir (2017) identified potential mechanisms of change from interviews with
seven school-age CWS who participated in an integrated treatment for stuttering.
Swift et al. (2017) and Hersh et al. (2022) proposed that realist evaluation could be used to support the translation of research into
EBP. Given the absence of comprehensive underlying program theories that inform EBP in the delivery of treatments for school-aged
CWS, our program of research begins with a program theory building evaluation. This theory building evaluation will investigate what
in an integrated cognitive-behavioural fluency enhancing treatment program (integrated treatment) for school-age CWS works, for
whom, in what contexts, how, and why. Consistent with realist methodology as described in Swift et al. (2017), multiple data sources
and various types of data are being used to construct a program theory. In the first phase of the research program, qualitative interview
data were obtained from SLPs and analysed in this study to develop a preliminary program theory. In the next study, qualitative and
quantitative data obtained from children and parents who participated in the CSP-SC will be analysed and the preliminary program
theory will be further developed. Thereafter findings will be further tested and the program theory refined with new cohorts of SLPs,
children, parents, and other stakeholders from different centres and different integrated treatment programs.
Using realist evaluation responds to the call for empirical evidence of the underlying theory of stuttering interventions (Bernstein
Ratner, 2018). It is suited for the evaluation of integrated treatments for school-age CWS because it embraces the complexities of the
interaction among aspects of the treatment program, the children who receive it, and the SLPs and parents who provide the resources
and the context; all operate in open, ever changing social systems.
Realist evaluation is not a one size fits all approach. Operationalization requires researcher reflexivity and creativity (Dalkin et al.,
2015). Definitions and locations of mechanisms and contexts in realist research depend on the question the researcher is asking; a
phenomenon can play a role as a mechanism in one study or a context in another (Greenhalgh & Manzano, 2022). The difference likely
depends on the focus of the study. It can also depend on whether or not intervention components are disaggregated from reasoning
mechanisms.
The purpose of this study was to identify discrete C, M, and Os embedded in the SLP data and begin construction of a preliminary
program theory of what in an integrated treatment program for school-age CWS works, for whom, in what contexts, how, and why.
2. Methods
2.1. Methodology
Principles of realist evaluation guided data collection and analyses. We consulted SLPs who delivered the CSP-SC and other
stuttering treatments and asked questions that aimed to discover their perspectives on barriers and facilitators of favourable and less
favourable treatment outcomes for CWS. Stakeholders such as SLPs are considered particularly adept at recognising the contexts that
can lead to certain outcomes which may not be as noticeable to the children and their families (Swift et al., 2017). Ethical approval was
granted by The University of Alberta Research Ethics Board (Study ID: Pro00051316).
2.2. Participants
Purposive sampling was used to recruit speech-language pathologists (SLPs) who delivered the CSP-SC (Langevin et al., 2007) at
the Institute for Stuttering Treatment and Research (ISTAR). Recruitment was an active process, occurring simultaneously with initial
data analysis by the first author. Recruitment was limited to ISTAR for this first critical realist investigation because we wanted the
sample to be homogeneous in terms of participants’ expertise in delivering integrated treatment in the form of the CSP-SC.
Regarding the number of participants needed to reach information redundancy or theoretical saturation (Sandelowski, 1995), we
were cognizant of the saturation research of Guest et al. (2006) that showed that 94% of the high frequency codes had been identified
within the first six interviews and 97% were identified after twelve interviews. Similarly, in their saturation study, Hennink et al.
(2017) found that 84% of codes were identified by the 6th interview and 91% by the 9th interview. Given the exploratory nature of this
first investigation of a program theory of integrated stuttering treatment, we were not seeking maximum variation in sampling but
rather wanted our sample to remain homogenous to ensure that we had an information rich sample (Sandelowski, 1995). In our
sample, it became apparent that we were approaching saturation in the 6th interview; the first author noted in her reflexive journal
that “…it really does feel as though nothing new was raised in this interview. The responses of SP6 supported those made by previous re-
spondents”. Interestingly, during coding three novel themes were identified in the 6th interview; one of which also appeared in the 7th
interview. However, only one novel theme emerged from the 7th interview. Thus, in our study, 94% of themes were present in the first
five interviews and 98% after six interviews. This finding is not surprising given that our sample, like that of Guest et al. (2006) was
3
homogeneous.
2.3. Data collection
Participants were interviewed individually by the first author either at their place of work or their home. The interviewer had 10
years’ experience in working with children who stutter and their families but did not have specific experience with the CSP-SC.
Following each interview, the interviewer noted observations and emerging themes in a reflexive journal.
Interviews followed a semi-structured format with an interview guide used as a starting point for questioning. As part of concurrent
data collection and analysis, changes were made to the interview guide to further explore ideas coming from earlier interviews.
Following each interview, the first author kept fieldnotes and a reflexive journal outlining emerging themes and potential CMO
configurations from the interview.
The interview guide was designed to capture the contexts and mechanisms that, according to the observations and beliefs of the
SLPs, lead to favourable and less favourable outcomes for the children. SLPs were asked to tell the interviewer about clients who they
thought were successful or less successful in the CSP-SC program, to talk about the facilitators and barriers that lead to those outcomes,
and to define successful outcomes. They were asked if the clients to which they referred were typical of others who had similar
outcomes. Although the questions were asked in relation to the CSP-SC, the SLPs responded about clients they had treated using a range
of different treatment approaches.
All interviews were audio-recorded and transcribed verbatim. One interview transcript was transcribed by the first author, the
remainder were transcribed by research assistants. To ensure descriptive validity (Maxwell, 1992), transcripts were then reviewed by
different research assistants who listened to the recordings and noted any disagreements. Percent word agreement ranged from 99.l –
99.6% across all transcripts. During write-up, fillers were removed from quotes for ease of reading.
2.4. Realist thematic analysis
Coherent with realist principles we used an iterative retroductive approach to analyze data. That is, we moved between inductive
and deductive processes to identify causal powers and mechanisms, linking data to program theory and thereafter linking program
theory to formal theory (Nichol et al., 2023; Rameses, 2017; Wiltshire & Ronkainen, 2021).
Using NVivo 12 software, interviews were initially coded by the first author and two research assistants who were not involved in
the transcription process. Codes were developed inductively from the data within the categories of Contexts, Mechanisms, and Out-
comes. Favourable outcomes were derived from “Definitions of Success” provided by the participants. Mechanisms were coded as
either facilitating or impeding based on the outcome to which the SLP was referring. Once four interviews had been inductively coded,
the first author combined codes to formulate a codebook. The final three interviews were deductively coded into the codebook with
new inductive codes added if required. The initial four interviews were checked for the new codes. Consistent with realist philosophy
the codebook was not viewed as “fixed”; it continued to evolve as data were simultaneously obtained and analyzed.
Following this first coding, it became clear that certain treatment program components and SLP characteristics identified by
participants as being important did not fit with the emerging categories of within-child mechanisms or contextual factors and were
better understood as resource mechanisms that interact with within-child reasoning mechanisms. Similarly, the child’s innate phys-
iological capacity for motor speech change did not fit within the category of reasoning mechanisms. Operationalizing discrete contexts,
mechanisms, outcomes has its challenges (e.g., Dalkin et al., 2015; Frykman et al., 2017; Lemire et al., 2020; Nielsen et al., 2021).
Given that the locus of change in an integrated treatment is within the child, we adopted the approach proposed by Dalkin et al. (2015)
and disaggregated reasoning mechanisms located within the child from resource mechanisms located in the treatment program and the
SLPs (i.e., SLP skills and capacities). We also disaggregated innate motor speech capacity for speech change located in the child’s motor
speech physiology and coded it as a resource mechanism. Thus, data were re-coded into Resource Mechanisms (RM), Contexts (C),
within-child reasoning Mechanisms (M), and Outcomes (O).
2.5. Rigour: credibility and quality of thematic findings and preliminary program theory
The second author independently completed a realist thematic analysis of the entire data set using retroductive processes. The
results were then compared to the original coding by the first author. Similar to Wiltshire and Ronkainen (2021) we reviewed each
other’s themes, comparing and cross checking them for accuracy by returning to the transcripts and assembling supportive quotes for
each theme. Like Wiltshire and Ronkainen (2021), we did not quantify our level of agreement. Rather, we came to agreement through
consensus. In doing so only one new theme was constructed. Thereafter, a finalized code book was updated for use in future research.
All RM, C, M and O codes were identified with a numeric code preceded by “F” if it facilitated a favourable outcome and “I” if it
impeded a favourable outcome. For example, FFC1 refers to “Engaged parents/family”, whereas IFC1 refers to “Lack of Family
Engagement”. FO1 refers to “Child enjoys talking”. References to outcomes were coded as FO if favourable and LFO if less favourable.
Thereafter RMCMO linkages embedded in the data were noted. From these data a preliminary overall program theory was developed.
To assess rigour in terms of validity, ethics and accessibility (Porter, 2007) we used the TAPUPASM framework (Pawson et al.,
2003; Ryan & Rutty, 2019) that is coherent with the ontology and epistemology of critical realism (Ryan & Rutty, 2019). To further
assess quality, we used the Rameses (2017) quality standards for realist evaluation for evaluators and peer-reviewers (see supple-
mentary material B); TAPUPASM and RAMESES II were specifically developed for evaluating the rigour and quality of critical realist
findings and reporting. Further, although there is debate as to the logicality of quantifying themes in qualitative research (Braun &
4
Clarke, 2021), Wiltshire and Ronkainen (2021) suggest that it may still be useful to do so in realist thematic analysis given that
frequently occurring themes can be seen as “demi-regularities”. We thus determined the frequency with which participants contributed
to the themes (supplemental material C). In summary, of the 64 themes, 18 themes (28%) were found in the interviews of all 7
participants, 17 (27%) were in six interviews, 12 (19%) were in five interviews, 7 (11%) were in four interviews, 3 (5%) were in three
interviews, 5 (8%) were found in two interviews, and 2 themes (3%) were found in one interview. Overall, 85% of the major themes
were mentioned by at least half of the participants.
3. Results
3.1. Participants
Participants were seven female SLPs who worked at ISTAR who had experience delivering the CSP-SC (Langevin et al., 2007); all
had experience using a range of other stuttering treatments including Palin Parent-Child Interaction Therapy (Nicholas & Kelman,
2020), the Lidcombe Program (Onslow et al., 2023), the ISTAR Comprehensive Stuttering Program for teens and adults (Langevin &
Kully, 2012), and the SpeechEasy device (Janus Development Group, Inc, n.d.). One participant had over 25 years in providing
stuttering therapy for children who stutter and their families. Two participants had over 10 years’ experience. Three participants had
between 2 to 5 years’ experience and one participant had one year of experience. Two of the participants had previously worked as
SLPs for the Ministry of Health and as a school-based SLP. All SLPs completed a rigorous training program at ISTAR prior to being
employed to deliver the CSP-SC and all were dedicated to providing stuttering treatment.
3.2. RM, C, M, O codes
Data analysis revealed 25 resource mechanisms, 16 contexts, 9 mechanisms and 13 outcomes. The results start with outcomes
because they represent the targets in treatment to which the RM, C, and Ms contribute. Quotes that show linkages are identified as such
in brackets after the quote. Further example quotes are included as supplementary material D.
3.3. Outcomes
As defined by the SLPs, successful child related outcomes include a child who enjoys talking (FO1), is able to approach new
challenges (FO2), can problem solve their own issues (FO3), and is open and accepting of their stuttering and acknowledges stuttering
when appropriate (FO4). The child is able to use fluency skills and cognitive-behavioural strategies at a high level, including taking
responsibility for their own skill practice and seeking out a parent to practice when required (FO5). The child maintains low levels of
stuttering over a period of years, keeping up maintenance activities as required to do so (FO6), and is able to achieve fluency when they
want to be fluent (FO7). Successful outcomes also include a child who participates regardless of stuttering (FO8): “fluent speech is part of
what we’re working on but… I love it when parents call me up and they say, ‘Oh they’re just so comfortable talking about stuttering (FM5) and
it’s not stopping them from doing anything’ (FO4, FO8)” [Linkage M+O]. Regarding the family, successful outcomes were that family life
was less stressful and more enjoyable following therapy (FO9): “a lot of parents that are successful in the maintenance phase (FO) … go
home to make adjustments to their everyday life (FFC1) … [maybe] take down the pace a little bit [and] their children are a little bit less stressed
(FO9)” [Linkage C+O].
Less favourable outcomes included short- or long-term loss of fluency (i.e., increased overt stuttering) (LFO1), and increased covert
stuttering and increased avoidances (LFO2). SLPs perceived that less favourable outcomes were related to discontinued or insufficient
practice (LFO3), and not using skills outside of the clinic (LFO4).
3.4. Resource mechanisms
3.4.1. SLP skills and abilities

Participants indicated that important SLP skills and abilities included establishing trusting relationships with the child and family
(FRMS1), having an intuitive understanding of the child’s and family’s needs (FRMS2), understanding that there are differences among
families (FRMS3), and having deep work satisfaction (FRMS4). SLPs need to know when to refer for further psychological support
(FRMS5), they need to collaborate with school personnel (FRMS6), and they need to have good ability to model fluency skills (FRMS7).
SLPs also need to be able to empower parents and children to problem solve issues arising during treatment and during the mainte-
nance period (FRMS8).
3.4.2. Components of a treatment program

Participants thought group work was an important component in treatment programs (FRMT1). Group work gives “kids … a chance
to meet other kids” who stutter. It provides an opportunity for children to share ideas about various aspects of communication, including
dealing with avoidances and teasing and bullying. Group work enables the children to have more fun and feel like “they’re participating
in something worthwhile”. Working in a group format also provides naturalistic opportunities to support each other in practicing fluency
skills and engage in real life transfers. One participant talked about the “cool kid” effect. That is, “if you’ve got a cool kid in the group and
they’re trying hard … some of the other quieter kids will follow them and look to them as a bit of a leader”. On a cautionary note, however: “if
the cool kid isn’t having any of it then that also affects the group sometimes” (IRMT2).
5
Given that avoidances (see IM3; a child that does not buy-in and is inclined to avoid) “have the potential to play a very big role if not
addressed”, SLPs thought that having a talk in which “avoidances and what avoiding might look like and being brave and how to overcome
those avoidances” is essential (FRMT2) [Linkage RM+M]. The availability of an onsite psychologist as well as scheduled individual
sessions with an SLP to help deal with attitudinal and emotional issues were considered helpful, as was having a former client (FRMT3)
share their experiences with the children. SLPs noted that these talks sometimes enabled children to share experiences and perspectives
of which SLPs were not previously aware.
Achieving stability in using fluency skills (FRMT4) was believed to be facilitated by doing a “warm-up” and “lots of [fluency skill]
practice prior to transfers”. The transfer phase (FRMT5) also enables SLPs to provide children and parents with supported challenges so
that parents and children experience difficulties and learn to problem solve such difficulties in real life situations. Some participants
raised concerns about the amount of time spent learning the fluency skills in comparison to the time spent discussing avoidances and
other cognitive behavioural issues (IRMT1), fearing that this might give the perception that being fluent was more important than
participating regardless of stuttering.
Parent training (FRMT6) was thought to be fundamental to the child’s success as were group parent meetings (FRMT7) in which
parents have the opportunity to share and learn from each other. Parents should also be encouraged to engage in check-ins with their
SLP during the maintenance phase of therapy (FRMT11) to help them learn to problem solve issues that arise.
Using external incentives judiciously (FRMT8) during skill acquisition was thought to aid motivation (FM1), however one
participant cautioned “external motivation… can be a big barrier to maintenance” (LFO) because “if the kids don’t see any value in what
they’re doing they’re not going to do it (IM1) and an external motivator only works for so long until it loses its magic…” [Linkage M+O]. Other
essential treatment components included individualizing treatment to meet the needs of each child/child-family duo (FRMT9), and
ensuring the treatment is enjoyable for the children (FRMT10) “because no … child … is gonna spend a good chunk of time doing something
that they don’t enjoy doing” (IM1) [Linkage RM+M].
3.4.3. Child physiology and age

A child’s innate physiology can either facilitate (FRMC1) or impede (IRMC1) fluency-related outcomes (e.g., achieving fluency
when the child wants to [FO7] or maintaining low levels of stuttering over a period of years [FO6]). Some children have more difficulty
accessing stable speech due to a compromised or less flexible motor speech system, a motor speech disorder, or other comorbid speech-
language, attentional, behavioural or cognitive difficulties (IRMC1). Whereas, “if they have the physical ability to change how they speak
more easily, then the therapy, in terms of the learning and the motor practice is gonna go faster” (FRMC1). Similarly, children with reduced
physiological awareness of moments of stuttering (IRMC2) have more difficulty achieving favourable fluency-related outcomes than
those who have capacity for self-monitoring. Comorbidities do not mitigate ability to benefit from and ability to use fluency skills and
CBT strategies, however, such skills may need to be modified to meet the child’s current capacities.
The age or developmental stage of the child and overall maturity could either impede (IRMC3) or facilitate (FRMC2) optimal
outcomes. SLPs reported that “tweens” “sort of 9, 10 and 11 can be a little tricky” (IRMC3) because they are seeking independence but do
not have the maturity to be completely independent. Conversely, “teens in some ways can be almost a little bit easier than younger ones”
(FRMC2) due to the greater ability to negotiate goals, participate in more complex cognitive-behavioural activities, and take re-
sponsibility for their own practice.
3.5. Contexts
3.5.1. Family engagement

Having a family that is engaged in the treatment process (FFC1) facilitates achievement of optimal outcomes. Engaged families are
those who buy into and trust in the programming. They are comfortable modelling fluency skills and facilitating the child’s use of
cognitive behavioural skills. Engaged parents are willing to make a lifestyle change to create a fluency enhancing environment. They
are able to adapt in response to their child’s changing responses. Parent physical and mental health are also important for family
engagement.
Families who are less engaged (IFC1) include those who lack buy-in and are not comfortable to model fluency skills. Less engaged
families may have demanding schedules that tax the child. Parents may also become less engaged over time due to satiation with the
demands of treatment or in response to negative child responses to parents’ initiatives; “usually the people who don’t do so well with the
program treat it as a lot of work and it’s something extra that has to be tacked on to a very busy day” [Linkage C+O]. Some parents may have
health issues that limit their ability to facilitate their child’s success.
3.5.2. Parent-child relationship

The relationship between the parent and the child is thought to be a key facilitator (FFC2) or impediment (IFC2) to successful
outcomes. The parent/child relationship is facilitative when the parent has an intuitive understanding of the child’s needs and how the
child works best. Facilitative relationships include having a strong open working relationship with the child, working together as a
team, having a good match in temperament, and parent nurturing of a child’s emerging independence while still maintaining over-
sight. Finally, the parent/child relationship is facilitative when parents know how to manage a sensitive child but less facilitative when
they have difficulty managing a sensitive child (IFC2).
Tension-filled parent-child interactions (IFC2) that negatively affect the quality of the working relationship can emanate from
conflicting personalities and power struggles with the child. Excessive focus on fluency and unrealistic expectations of consistent
fluency skill use or near zero stuttering can stress the parent-child working relationship as can giving and too much corrective
6
feedback. Stress in the parent/child relationship can also emanate from the child’s sensitivity or the child’s negative response to praise.
In addition, with younger children the parent/child relationship can be stressed by the need for parents to be the primary driver for
beyond-clinic practice.
Parents who are highly anxious about the effect of stuttering on their child when the child is not concerned can negatively affect the
parent-child relationship. Sometimes the anxiety of a parent can come from their own negative experiences as a stuttering child. In
some cases, the treatment approach may not be a good match for the parent-child duo in which case “you can try different things. If you
see the CSP is not working, you can definitely try other approaches … it really depends on the duo”.
3.5.3. Parental understanding of program

Given that parent involvement in the treatment program is essential, an incomplete understanding of parental involvement can
impede optimal outcomes (IFC3). In contrast, having a good understanding of parent involvement and having at least one parent
become the “child’s therapist” is facilitative (FFC3).
Children who experience favourable outcomes have parents who understand the program components (FFC4). These parents
understand that fluency is only one goal and that improved communication skills and reduced avoidances are equally important
(FFC5). Outcomes are facilitated when parents understand the importance of maintenance, the risk factors in maintenance and that
consistency in delivering treatment and maintenance is important. These parents have also become good problem solvers and they
know how to sequence activities from easier to more challenging, incorporate self-monitoring/self-evaluation in practice activities,
and monitor the child’s independent practice. These parents are creative in designing practice activities, know how to switch up
activities and involve the child in designing practice activities using materials/activities that tap into the child’s interests, and
“incorporate the skills into daily life”. They also understand that they have a role in seeking school-based SLP and teacher support for the
child at school.
Conversely, less favourable outcomes are associated with children of parents who have an incomplete understanding of the pro-
gram components (IFC4). Specifically, these parents are “not finely sequencing [practice activities] enough so it’s too big of a jump for the
child” or are not providing the scaffolding that the child needs. Some parents are less able to problem-solve as required "just not maybe
understanding the importance of the problem-solving process in the maintenance". Less favourable outcomes are also associated with parents
not having their children incorporate continued real-life transfer practice during maintenance and doing “what the parent chooses rather
than doing what the child chooses. We want practice to be fun. If it’s an activity that the child doesn’t want to do at all they’re not going to want
to do the practice (IM1)” [Linkage C+M].
3.5.4. Parental expectations of fluency

More favourable outcomes were associated with parents who have realistic expectations about fluency and fluency skill use (FFC5)
and know when to push their child and when to back off. Less favourable outcomes were associated with situations in which parents
“put more weight on the fluency side of things” (IFC5).
3.5.5. Change in adult support

Treatment outcomes can be negatively affected when there is a change in the primary adult supporting therapy (IFC6), as the new
parent or caregiver “doesn’t get the same level and depth of knowledge as the parent who attended, so it can be hard for the parent who
attended to convey that to the other parent.”
3.5.6. Demands on parents of children with co-morbidities

Parents of children with comorbidities have increased demands for parent involvement and support. Less favourable outcomes for
children with comorbidities were associated with parents for whom dealing with stuttering over-taxed family capacities (IFC7): “I’m
thinking of my kids with multiple diagnoses (IRMC1) that they just need more support to be able to do the skills, so if mom or dad isn’t doing the
skill (IFC1) then the child’s not doing it (LFO3)” [Linkage RM+C+O]. For some families, stuttering was perceived as the least important
of the issues.
Participants also talked about families who were able to put supports in place for children with comorbidities (FFC6), for whom
outcomes were more favourable: “They had a little boy who had multiple needs (IRMC1) … but they really took to heart how slowing down the
pace … walking more slowly (FFC6) helped him to be more fluent, and more calm, and more in control (FO7)” [Linkage RM+C+O].
3.5.7. School environment

The school environment has the potential to both impede or facilitate favourable outcomes depending on the mechanisms activated
in the child and the outcomes being studied. For some children “there’s not really an impact at school where they’re not really being bugged
or their friends don’t really care, they just play with them (FSC1/ISC1)”. In the case of participatory outcomes, having an environment that
results in the child not seeing stuttering as a problem is facilitative (FSC1). However, the SLPs said this can impede fluency outcomes if
the child does not see the need to practice and use skills at school (ISC1). In other cases, there may be negative impacts of stuttering at
school such as teasing or being asked to participate in class in a way that does not facilitate fluency or participation (ISC2). While
negative, teasing and bullying can also motivate the child to engage in therapy and use learned speech techniques and strategies
(FSC2): “Unfortunately, those children that have experienced that [teasing and bullying] already typically come to therapy with a lot of
motivation to work on their fluency skills” [Linkage C+M].
7
3.6. Mechanisms
3.6.1. Motivation
The child’s level of motivation (FM1/IM1) for participating in therapy and using techniques and CBT strategies influences out-
comes. Favourable outcomes occur when there is “some intrinsic, internal motivation from the child” and the child wants therapy for
themselves. The motivated child is keen to learn, actively participates in therapy/maintenance activities and chooses to practice.
Favourable outcomes also occur when a supportive parent or caregiver stimulates the child’s motivation by involving the child in
designing practice activities and tapping into the child’s areas of interest so “they can be involved in choosing activities and things to make
it fun” (FM1, FFC4) [Linkage C+M]. The child feeds off the parent’s motivation and engagement and willingly participates in therapy
and maintenance activities.
Conversely, less favourable outcomes occur when a child has reduced motivation (IM1) to participate in the treatment and
maintenance. While external motivators (FRMT8) such as points or stickers might have worked initially to facilitate motivation, if
there is no internal motivation, the external motivators will lose their motivating powers (IM1). Children’s motivation can also be
reduced if they pick up on their parent’s lack of engagement (IFC1) or they themselves are not ready for therapy. A lack of motivation
also may result from previous therapy that has not been successful or develop over time as the child becomes tired and frustrated with
ongoing therapy.
3.6.2. Personality and psychological characteristics

Aspects of the child’s personality and psychological characteristics can either facilitate (FM2) or impede (IM2) outcomes.
Favourable outcomes were associated with children who had a non-perfectionistic personality and were open and willing to try new
things. In contrast, less favourable outcomes are associated with “those really sensitive children who are so aware… [who] tend to develop
perfectionism [for whom] …it’s the end of the world if they can’t quite get a particular skill”.
Outcomes are facilitated when a child is not overly stressed or worried about stuttering, has the cognitive flexibility to cope with
unexpected life events and is receptive to parent support. Conversely, anxious children and those who put on a facade of not being
stressed but actually have many covert features of stuttering are less likely to be successful. Resistance to parent support and not being
open to feedback also mediate treatment outcomes.
3.6.3. Child buy-in to therapy process

The degree to which a child buys into the therapy process (IM3/FM3) will contribute to whether or not there is a favourable or less
favourable outcome from treatment. When a child understands and trusts the process of therapy and maintenance (FM3), favourable
outcomes are more likely. Such children openly use fluency skills and CBT strategies, are independently able to find their own optimal
rate of speech to support speech fluency, and understand the importance of incorporating practice into their everyday lives. These
children take responsibility for practising and they have trust in their SLPs and parents.
Children who lack buy-in to the process of therapy and maintenance (IM3) have less favourable fluency-related outcomes.These
children may think that they do not need therapy, or may be disappointed that maintenance work is required after the therapy
program. Children may also find it difficult to practice due to outside pressures and demands on their time and may not want to take
responsibility for their practice. Some children dislike the sound of fluency skills, being “unsure about using the skills or feeling
embarrassed about how it sound[s]” and thus do not want to use them. At times these children will prefer to avoid speaking rather than
use fluency skills or stutter.
3.6.4. Effort and control

For some children with severe overt stuttering, therapy techniques make their speech less effortful (FM4) resulting in increased
carry-over, although the interaction with intrinsic motivation (FM1) can be hard to separate:
… the kids that I’m thinking of were the most severe that I’ve seen [and] they were also ones who said to their parents that they wanted to
come (FM1) so it’s kind of hard to tease the two out. Did they carry over better (FO) because they wanted to be here, or because they’re
more severe and it was less effort to use the skills to get their words out overall (FM4)? (SP4) [Linkage M+O].
For other children with less severe overt stuttering, it may require “more effort for them to do skills than to just do their normal talking”
and the cost of the effort to use therapy techniques is not worth it (IM4). Similarly, some children just want to speak spontaneously
without using fluency skills. These mechanisms were discussed within the context of less favourable fluency-related outcomes and as
such were coded as impeding; however, it is clear these mechanisms would be facilitative if they were causal to the “Child enjoys
talking” (FO1) or “Participates regardless of stuttering” (FO8) outcomes.
3.6.5. Self-efficacy
Favourable outcomes are associated with the child’s sense of self-efficacy (FM5), especially when the child goes into the main-
tenance period feeling empowered and confident. Children with a strong sense of self-efficacy are willing to say what they want to say
whether or not they stutter. Positive memories of the group (FRMT1) and positive interactions with other children in the program
result in collective efficacy, the shared perception of capacity to be successful (Vassilev et al., 2014) that in turn leads to continued
self-efficacy.
8
3.7. RMCMO linkages
Given that these findings do not include parent-child data, RMCMO configurations have not been hypothesized. Rather, the
linkages embedded in the quotes were highlighted throughout the results section and in supplementary material D. Examples of the
RMCMO Linkages that SP3 made for children who are more likely to have favourable and less favourable outcomes are shown in
supplementary material E.
3.8. Preliminary program theory
We propose a preliminary program theory (Fig. 1) that presents all constituent factors operating within, and thus being influenced
by, an open social system. Constituent factors consist of the facilitating or impeding within child mechanisms that, in interaction with
facilitating or impeding resource mechanisms and contexts, generate favourable or less favourable outcomes. Resource mechanisms
are located within contextual factors to indicate how inextricably linked they are with contexts in providing resources and support to
the child. The proposed preliminary theory incorporates a feedback loop that relies on specific resource mechanisms of treatment
maintenance checks and SLP ability to empower problem solving that facilitate the child/family getting back on track after a period of
less favourable outcomes. The feedback loop also ties favourable outcomes to the aforementioned resource mechanisms given that they
also enable continued support and facilitation of favourable outcomes. The bidirectional arrows between mechanisms and the resource
mechanisms/contexts reflect the influences of maturational development in the child that result in ongoing changes in the child’s
reasoning and, hence, the child’s changing needs for SLP and treatment resource mechanisms and contextual support.
4. Discussion
We begin with a discussion of the proposed preliminary program theory and RMCMO linkages embedded in the data and move to
considerations of the constituent factors as they relate to relevant social theories, existing realist research, EBP, and the need for future
research.
4.1. Preliminary program theory and RMCMO linkages
The proposed preliminary program theory elucidates the interaction of RM, C, M and Os that combine to generate outcomes for
Fig. 1. Proposed preliminary program theory for an integrated cognitive-behavioural fluency enhancing treatment for school-aged children who
stutter. Note, RM = Resource Mechanisms, C = Contexts, M = Mechanisms, and O = Outcomes. Codes preceded by “F” facilitated favourable
outcomes. Codes preceded by an “I” impeded favourable outcomes.
9
CWS in an integrated treatment program. It highlights that all operatives/stakeholders operate within open, ever changing social
systems. Of particular impact is the feedback loop between mechanisms and contexts/resource mechanisms due to maturation of the
school-age child.
Given that theoretical validity in realist evaluation is addressed by the ability to which theory functions as an explanation of how,
why, and for whom a program works (Maxwell, 1992), the validity of the preliminary program theory proposed here is supported by its
similarity to the findings of Caughter and Dunsmuir (2017) in their thematic map of child-identified mechanisms of change in a
different school-aged stuttering treatment, and Klatte et al.’s (2020) preliminary program theory of SLP and parent collaborative
practices when working with children with speech and language disorders. These similarities are further discussed below.
It is interesting to note that some of the constituents and semblances of the proposed program theory were discussed by Kully and
Boberg (1991). That is, they identified child, SLP, parent, and other (environmental) factors that influenced the achievement of the
stated program objectives. The factors that they identified that were found in this study and are still relevant to current practice
include: a child’s comorbidities and severity of stuttering (coded as RMs in this study), parent buy-in and engagement, and negative
environmental factors such as teasing (coded as Cs), and a child’s motivation, perhaps the most foundational mechanism.
In this study, examples of RMCMO linkages embedded in the data were presented but configurations that lead to generalizable
patterns of outcomes were not hypothesized. This is because future analyses of child and parent participant data are needed to
complete the foundational information on which such hypotheses should be made. Future research will confirm, refute, and refine the
constituent factors, program theory, and RMCMO linkages found in this study. Thereafter configurations will be hypothesized for
testing with new cohorts.
4.2. Constituent factors: outcomes
Favourable outcomes in this study were derived from SLPs definitions of success. SLPs referred to participatory outcomes (enjoys
talking, approaches new challenges, participates regardless of stuttering), speech related outcomes (uses fluency skills, maintains low
levels of stuttering, achieves fluency when fluency is wanted) and changes in thinking and independence (uses CBT skills, indepen-
dently problem solves). Consistent with the World Health Organization (WHO), International Classification of Functioning, Disability
and Health–Children and Youth Version (ICF-CY; WHO, 2007) and with the wider stuttering field (e.g., Yaruss et al., 2012), SLPs
viewed participatory outcomes as being most important. However, they also cautioned that given that a larger amount of time in
therapy is spent on learning speech restructuring/fluency enhancing and stuttering modification skills, there is potential for parents
and children to perceive that achieving fluency is more important than participation. Doing so may lead to increased parental emphasis
on using fluency skills, child disappointment in their speech and use of covert stuttering behaviours to reduce overt stuttering.
Throughout therapy SLPs should discuss with the children and their parents that the primary goal is participation and that the fluency
skills are tools that can be used to reduce or modify moments of stuttering when the child wants to do so rather than an end goal in and
of themselves. They should also explicitly discuss what outcomes the family is looking for and continue to do so throughout the
therapeutic process.
Participatory outcomes in this study are also consistent with those of children in Willis et al. (2018) (i.e., outcomes of achievement,
aspirations, and enjoyment in engaging in activities not previously thought possible), Cooke and Millard (2018) who conducted a
Delphi study with 9–13 year olds to determine their own views of the most important outcomes from stuttering therapy, and Caughter
and Dunsmuir (2017). Both Cooke and Millard (2018) and Caughter and Dunsmuir (2017) also affirm the importance of outcomes of
increased fluency and independence found in the present study. Some commentators contend that telling a child that it is OK to stutter
at the same time as teaching them fluency skills to increase their fluency sends a “double message” (Sisskin, 2023). We posit that these
foci in treatment are not mutually exclusive, particularly in the school setting in which children can experience bullying due to their
speech. Providing children with speech tools to speak with less stuttering can help them increase, or maintain increased, participation.
At the same time, work on self-advocacy and self-acceptance can enable them to accept their stuttering and themselves as a CWS and do
so with a focus on participation regardless of whether or not they stutter. The child may choose to participate while stuttering or use
speech tools to participate with less stuttering but the key is that the child has the choice. While the authors actively support reducing
stuttering stigma in schools (e.g., Langevin, 2000, 2015; Langevin & Prasad, 2012) and agree that the onus should not be on CWS to
change their speech in order to be accepted, society is not yet a welcoming place for all CWS. Integrated treatments provide CWS with a
range of tools, cognitive and behavioural, to navigate and successfully participate in this world.
In contrast to the focus on participation in favourable outcomes, the primary focus in less favourable outcomes was related to lost
speech related gains. It is likely that this is a result of the interview methodology. Whereas SLPs were specifically asked to define what
successful therapy meant to them, they were not asked to define unsuccessful therapy, resulting in less nuanced information being
shared relating to less favourable outcomes.
4.3. Constituent factors: resource mechanisms
4.3.1. Treatment components

Consistent with stuttering experts from around the world (Yaruss et al., 2012), SLPs considered it important for CWS to meet other
children who stutter and realise that they are not alone in their journeys. This is consistent with child-identified mechanisms elicited by
Caughter and Dunsmuir (2017) and Willis et al. (2018). The entire treatment need not be provided in a group, so this aspect can be
incorporated into a variety of therapy models. However, SLPs also cautioned that group formation must be done with attention to the
mix of personalities and potential leaders in the group. Bandura (1969) proposes that children learn from social models and are more
10
likely to model the behaviour of those who have characteristics similar to themselves (e.g., other CWS) and who have higher status in
the group (e.g., “the cool kid”). In line with this, SLPs reported positive outcomes from group activities in terms of social and
fluency-skill use models, however, if the “cool kid” was not engaged this could derail the treatment for some of the children.
Conversely, this could be a prime opportunity for children to be supported in the development of resilience in the face of adversity.
4.3.2. Child attributes

In contrast to Willis et al. (2018) who found that a child’s level of disability did not impact reported outcomes achieved, findings of
this study showed that children were considered to have a harder time achieving optimal outcomes when they had less capacity for
speech motor change or had comorbidities. Although these children were considered to have higher risk of less favourable outcomes, it
is important to note that their systems were not viewed as entirely inflexible. With the right amount of parental support and SLP
program modification, these children could still achieve favourable outcomes across both overt stuttering and participatory domains;
indeed this has been demonstrated in previous research (Harasym & Langevin, 2012).
4.3.3. SLP capacities

In general, and consistent with findings in this study, realist evaluations in the allied health space have found that having speci-
alised health professionals run programs is important (e.g., Klatte et al., 2020; Tennant et al., 2020; Willis et al., 2018). In particular,
although there are differences in the way in which SLP factors were coded across studies, SLP capacities found in this study, including
the ability to develop trusting relationships, understand family needs and differences among families, and empower problem solving,
were to varying degrees also found in the collaborative practice research of Klatte et al. (2020) and Langner and Fukkink (2023) and
research into common factors in psychotherapy outcomes (Wampold, 2015). In addition, as noted in this study and Klatte et al.,
(2020), having deep work satisfaction is important for SLPs’ ability to facilitate optimal outcomes; work satisfaction underpins the way
SLPs approach therapy and in turn their ability to support families.
The importance of the SLP’s capacity to take time to collaborate with other professionals and to work with family members was also
emphasized in Klatte et al. (2020). An integrated program such as the CPS-SC relies on collaborative practice between the SLP and the
child and the parent. Thus the SLP requires capabilities, attitudes and beliefs that support collaboration, such as mutual cooperation
and trust and understanding of roles (Langner & Fukkink, 2023).
It is critical to note that the within child mechanisms of trust in the therapy (FM3) or lack thereof (IM3) found in this study interact
with the SLP’s ability to form a positive therapeutic relationship. Within psychological interventions, the therapeutic alliance has been
found to have a medium effect-size in relation to the therapeutic outcome (Wampold, 2015). CWS agree that the therapist is important
in their experience of treatment (Caughter & Dunsmuir, 2017). Previous realistic evaluation studies have similarly posited that having
a safe context (Willis et al., 2018) and trust between stakeholders (Langner & Fukkink, 2023) is important for favourable outcomes.
4.4. Constituent factors: contexts
Contextual factors identified in this study are consistent with the distinctions of observable features (e.g., family, school) and
relational and dynamic forces (e.g., parent-child relationship factors) revealed in a 2022 review of realist studies (Greenhalgh &
Manzano, 2022) and mirror the systemic supports identified by Caughter and Dunsmuir (2017). The existence and understanding of
dynamic forces at play in integrated treatments for school-age CWS is fundamental to EBP and the synergy needed to achieve optimal
outcomes. Such features also make tangible the understanding of unobservable phenomena that are at the heart of realist evaluation
research. These distinctions will be taken forward into analyses of contextual factors in child and parent data.
4.4.1. Interaction of contexts

As Bronfenbrenner’s Ecological Systems Theory (Bronfenbrenner, 1977) states, a child develops within a series of embedded
contexts that interact with each other. At the microsystem level, a stuttering child has a family, attends speech therapy, attends school
and participates in extracurricular activities. All of these contexts then interact with each other at the mesosystems level. The SLPs in
the present study highlighted the complexity of these interactions. Interactions with peers and teachers at school could both facilitate
or impede the child’s continued use of fluency skills or participatory risk-taking. Extracurricular activities could result in a lack of time
to practice therapy techniques but also could increase a child’s confidence and self-efficacy, leading to increased engagement in the
therapy. Parents’ engagement and understanding of the treatment directly influenced the child’s own motivation to engage with the
therapy.
4.4.2. Pivotal role of parents

Findings in this study highlight the pivotal role of parents and family in either facilitating or impeding child outcomes. Themes in
this study regarding the influence on the children of parents’ own opinions about stuttering and their engagement with therapy are
similar to the link between parents’ and participants’ emotional factors found in Caughter and Dunsmuir (2017). It is clear that in-
tegrated treatments such as the CSP-SC depend on collaborative practice, that is, a team approach involving the child, parents and SLP
is necessary. Reinforcing this, the facilitative contexts of parent/family in this study (i.e., engagement, parent-child relationship,
understanding of involvement in therapy and understanding of treatment components) were similar to outcome factors in Klatte et al.
(2020) (i.e., understanding the child’s needs and how to support development, develop confidence in supporting child’s development,
and taking the lead in adapting activities). There were also similarities in impeding contexts found in this study and less favourable
outcomes in Klatte et al. (2020) (e.g., the burden of parents’ responsibility and frustration in therapy).
11
Findings indicate that parents need to be supported to learn techniques themselves and empowered to problem-solve their child’s
ongoing therapy journey post-treatment. This is not just the view of the SLPs but also a desired outcome of school-aged children who
stutter (Cooke & Millard, 2018). Parent training is a critical resource mechanism in integrated treatments that enables parents to
develop the skills needed to support their child and to fully understand their expected role in intervention. Work with parents is part of
many treatments for school-aged stuttering (e.g., Andrews et al., 2016; Euler et al., 2021; Koushik et al., 2009; Langevin et al., 2007;
Yaruss et al., 2012), thus it is important for treating SLPs to routinely check parental understanding of their role in the treatment
process. Of particular interest is the finding that parent group discussions enabled parents to share and learn from each other in a way
that was different from the one-to-one SLP-parent meetings.
4.5. Constituent factors: mechanisms
As shown in the proposed preliminary program theory and consistent with the realist tenet of generative mechanisms (Pawson &
Tilley, 1997), mechanisms considered operative by SLPs in this study reflect children’s reasoning, perceptions, beliefs, and experiences
with stuttering treatment; they also speak to the importance of a child’s agency and internal locus of control. Facilitative mechanisms
such as motivation, understanding and trusting the process of therapy/maintenance and having self-efficacy can all be encapsulated
within the concept of internal locus of control (Rotter, 1966). When these mechanisms are activated, the children believe that they are
able to make a difference to their experience of stuttering and as such are more likely to engage in activities that lead to favourable
outcomes. Conversely, a child who has reduced motivation, perhaps due to previous unsuccessful therapy, and/or a lack of buy-in to
the process of therapy may not believe that they are able to achieve meaningful changes.
According to SLPs, choice and enjoying treatment activities contribute to a child’s motivation to participate in treatment. This is
consistent with clients in Willis et al. (2018) who spoke of the importance of choice and fun. Providing clients with choices gives them a
sense of agency and internal locus of control. Stuttering has been described from an internal perspective as a loss of control over speech
(Perkins, 1990). When a speaker feels a loss of control and believes that they have no agency to regain that control, it leads to less
favourable outcomes. For some children having the choice to use fluency skills to reduce stuttering when they want to do so bolsters
their internal locus of control (Caughter & Dunsmuir, 2017). Conversely, perfectionistic children who have unrealistic standards about
their speech fluency may perceive it as a failure when they stutter despite using fluency skills, reducing their sense of agency.
Dweck’s (2000) work on growth versus fixed mindset and performance versus mastery orientation is relevant here. Children who
focus on the final outcome (performance orientation) and who believe that a person either can or cannot do something (fixed mindset)
may be less likely to achieve favourable outcomes than those who focus on developing a new skill (mastery orientation and belief that a
person can learn to do things that they currently are not capable of achieving (growth mindset). SLPs thus have a role to play in helping
children with a fixed mindset and performance orientation to shift towards a growth mindset and mastery orientation. As suggested by
SP5, providing opportunities for supported failure during transfer activities is one way to achieve this. Children and their families need
to learn that it is OK to not achieve what they set out to achieve on the first try and develop agency in making changes that move them
towards their goals. While teaching problem-solving is already a CSP-SC treatment component, children and parents with perfec-
tionistic personality traits will need additional support to do this during treatment.
4.6. Clinical implications
Preliminary clinical implications have been highlighted throughout the preceding sections. An important point is that impeding
factors do not automatically contraindicate therapy. The mix of factors is important. For example, while the SLPs viewed a sensitive
child personality as impeding successful outcomes, they also reported that these children could achieve success if the parent-child
relationship was such that the parent could support the child to engage with the therapy and transfer tasks. Awareness of impeding
factors is important for counselling clients about potential outcomes and gaining informed consent for treatment. Awareness also
enables the treating SLP to directly address these factors during treatment in order to minimize their impact.
4.7. Strengths, limitations and future research
Strengths of this study include the a priori dual focus on gathering data that investigated intended and unintended outcomes, the
range in SLPs’ years of experience and level of expertise, and the multiple measures used to ensure rigour. Strengths also include
similarities with findings in the studies of Caughter and Dunsmuir (2017), who were investigating mechanisms of change in school-age
CWS following an integrated treatment program that has similarities and differences to the CSP-SC, and Yaruss et al. (2012), who
explored opinions of experts in treating school-age CWS.
A limitation to the generalizability of findings to other cohorts of SLPs and other integrated stuttering treatments for school-age
CWS is that the sample was homogeneous; all participants came from the same treatment centre, all were highly specialized in
providing stuttering treatments, and all were highly specialized in providing the CSP-SC. A further limitation is that the proposed
preliminary program theory has been constructed solely from expert knowledge of SLPs. As noted earlier, the input of the opinions and
experiences of children and parents who participated in the CSP-SC is needed to further develop the program theory. Thereafter
maximum variation sampling will be needed to test the findings of this first program of research with new cohorts of participants at
multiple treatment centres.
12
5. Conclusion
This study proposed a preliminary program theory that contributes to an emerging body of knowledge about what in an integrated
treatment for school-age CWS works, for whom, how, and why. This study makes evident that realist evaluation can be used to inform
translational research programs for treatments with an emerging or well-established efficacy base.
Funding sources
This work was supported by the Women’s and Children’s Health Research Institute [1270], Flinders University [Outside Study
Program], and an anonymous donor to the Institute of Stuttering Treatment and Research [partial funding of A/Prof Langevin’s po-
sition]. Funders had no involvement in study design, analysis or drafting of the manuscript.
Data statement
Data is not available for sharing. Participants were assured that raw data would remain confidential.
CRediT authorship contribution statement
Michelle C. Swift: Writing – review & editing, Writing – original draft, Visualization, Validation, Project administration, Meth-
odology, Investigation, Funding acquisition, Formal analysis, Data curation. Marilyn Langevin: Writing – review & editing, Writing –
original draft, Validation, Supervision, Resources, Project administration, Methodology, Funding acquisition, Formal analysis, Data
curation, Conceptualization.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.
Data Availability
The authors do not have permission to share data.
Acknowledgements
The authors would like to thank the research assistants who transcribed, cross-checked transcriptions and contributed to the data
coding, and the participants for sharing their knowledge. We also thank Alex M. Clark for his mentoring in critical realist methodology
during the early stages of this project.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jfludis.2024.106076.
References
Andrews, C., O’Brian, S., Onslow, M., Packman, A., Menzies, R., & Lowe, R. (2016). Phase II trial of a syllable-timed speech treatment for school-age children who
stutter. Journal of Fluency Disorders, 48, 44–55. https://doi.org/10.1016/j.jfludis.2016.06.001
ASHA). (n.d.) Evidence-based practice (EBP). retrieved December 1, 2023 from 〈https://www.asha.org/research/ebp/〉.
Bandura, A. (1969). Social-Learning theory of identificatory processes. In D. A. Goslin (Ed.), Handbook of Socialization Theory and Research (pp. 213–262). Rand
McNally & Company.
Bernstein Ratner, N. (2018). Selecting treatments and monitoring outcomes: The circle of evidence-based practice and client-centered care in treating a preschool
child who stutters. Language, Speech and Hearing Services in Schools, 49(1), 13–22. https://doi.org/10.1044/2017_LSHSS-17-0015
Braun, V., & Clarke, V. (2021). To saturate or not to saturate? Questioning data saturation as a useful concept for thematic analysis and sample size rationales.
Qualitative Research in Sport, Exercise and Health, 13(2), 201–216. https://doi.org/10.1080/2159676X.2019.1704846
Bronfenbrenner, U. (1977). Toward an experimental ecology of human development. American Psychologist, 513–531 (July).
Caughter, S., & Dunsmuir, S. (2017). An exploration of the mechanisms of change following an integrated group intervention for stuttering, as perceived by school-
aged children who stutter (CWS). Journal of Fluency Disorders, 51, 8–23. https://doi.org/10.1016/j.jfludis.2016.10.003
Cooke, K., & Millard, S. K. (2018). The most important therapy outcomes for school-aged children who stutter: An exploratory study. American Journal of Speech-
Language Pathology, 27, 1152–1163. https://doi.org/10.1044/2018_AJSLP-ODC11-17-0195
Dalkin, S. M., Greenhalgh, J., Jones, D., Cunningham, B., & Lhussier, M. (2015). What’s in a mechanism? Development of a key concept in realist evaluation.
Implementation Science, 10(49). https://doi.org/10.1186/s13012-015-0237-x
Dweck, C. S. (2000). Self theories: Their role in motivation, personality, and development. Philadelphia: Psychology Press,.
Euler, H. A., Merkel, A., Hente, K., Neef, N., Wolff von Gudenberg, A., & Neumann, K. (2021). Speech restructuring group treatment for 6-to-9-year-old children who
stutter: A therapeutic trial. Journal of Communication Disorders, 89, Article 106073. https://doi.org/10.1016/j.jcomdis.2020.106073
13
Ezrati-Vinacour, R., Platzky, R., & Yairi, E. (2001). The young child’s awareness of stuttering-like disfluency. Journal of Speech, Language, and Hearing Research, 44,
368–380.
Frykman, M., von Thiele Schwarz, U., Muntlin Athlin, A., Hasson, H., & Mazzocato, P. (2017). The work is never ending: uncovering teamwork sustainability using
realistic evaluation. Journal of Health Organization and Management, 31(1), 64–81. https://doi.org/10.1108/JHOM-01-2016-0020
Greenhalgh, J., & Manzano, A. (2022). Understanding ‘context’ in realist evaluation and synthesis. International Journal of Social Research Methodology, 25(5),
583–595. https://doi.org/10.1080/13645579.2021.1918484
Guest, G., Bunce, A., & Johnson, L. (2006). How many interviews are enough? An experiment with data saturation and variability. Field Methods, 18(1), 59–82.
https://doi.org/10.1177/1525822X05279903
Harasym, J., & Langevin, M. (2012). Stuttering treatment for a school-age child with Down syndrome: A descriptive case study. Journal of Fluency Disorders, 37,
253–262.
Hennink, M. M., Kaiser, B. N., & Marconi, V. C. (2017). Code saturation versus meaning saturation: how many interviews are enough? Qualitative Health Research, 27
(4), 591–608. https://doi.org/10.1177/1049732316665344
Hersh, D., Azul, D., Carroll, C., Lyons, R., Mc Menamin, R., & Skeat, J. (2022). New perspectives, theory, method, and practice: Qualitative research and innovation in
speech-language pathology. International Journal of Speech-Language Pathology, 24, 449–459. https://doi.org/10.1080/17549507.2022.2029942
Hugh-Jones, S., & Smith, P. K. (1999). Self-reports of short- and long-term effects of bullying on children who stammer. British Journal of Educational Psychology, 69,
141–158. https://doi.org/10.1348/000709999157626
Janus Development Group, Inc. (n.d.). SpeechEasy. Available: 〈https://speecheasy.com/〉 [Accessed 11 December 2023].
Johnson, G., Onslow, M., Horton, S., & Kefalianos, E. (2023a). Psychosocial features of stuttering for school-age children: A systematic review. International Journal of
Language & Communication Disorders, 58(5), 1829–1845. https://doi.org/10.1111/1460-6984.12887
Johnson, G., Onslow, M., Horton, S., & Kefalianos, E. (2023b). Reduced stuttering for school-age children: A systematic review. Journal of Fluency Disorders, 78, Article
106015. https://doi.org/10.1016/j.jfludis.2023.106015
Klatte, I. S., Lyons, R., Davies, K., Harding, S., Marshall, J., McKean, C., & Roulstone, S. (2020). Collaboration between parents and SLTs produces optimal outcomes
for children attending speech and language therapy: Gathering the evidence. International Journal of Language and Communication Disorders, 55(4), 618–628.
https://doi.org/10.1111/1460-6984.12538
Kohmäscher, A., Primaßin, A., Heiler, S., Da Costa Avelar, P., Franken, M.-C., & Heim, S. (2023). Effectiveness of stuttering modification treatment in school-age
children who stutter: A randomized clinical trial. Journal of Speech, Language, and Hearing Research, 66, 4191–4205. https://doi.org/10.1044/2023_JSLHR-23-
00224
Koushik, S., Shenker, R., & Onslow, M. (2009). Follow-up of 6-10 year-old stuttering children after Lidcombe Program treatment: A phase I trial. Journal of Fluency
Disorders, 34, 279–290.
Kully, D., & Boberg, E. (1991). Therapy for school age children. In W. Perkins (Ed.), Seminars in Speech and Language. Stuttering: challenges of therapy (pp. 291–300).
New York: Thieme Medical Publishers.
Laiho, A., Elovaara, H., Kaisamatti, K., Luhtalampi, K., Talaskivi, L., Pohja, S., Routamo-Jaatela, K., & Vuorio, E. (2022). Stuttering interventions for children,
adolescents, and adults: A systematic review as a part of clinical guidelines. Journal of Communication Disorders, 99, Article 106242. https://doi.org/10.1016/j.
jcomdis.2022.106242
Langevin, M. (2000). Teasing and bullying: Unacceptable behaviour: The TAB program. Edmonton, AB: Institute for Stuttering Treatment and Research,.
Langevin, M. (2015). Bullying experienced by youth who stutter: The problem and potential intervention strategies. In K. O. St. Louis (Ed.), Stuttering meets stereotype,
stigma, and discrimination: an overview of attitude research (pp. 71–90). West Virginia University.
Langevin, M., & Kully, D. (2003). Evidence-based treatment of stuttering: III. Evidence-based practice in a clinical setting. Journal of Fluency Disorders, 28, 219–236.
https://doi.org/10.1016/S0094-730X(03)00040-8
Langevin, M., Kully, D. A., & Ross-Harold, B. (2007). The Comprehensive Stuttering Program for School-age Children with strategies for managing teasing and
bullying. In E. G. Conture, & R. F. Curlee (Eds.), Stuttering and related disorders of fluency (3rd ed..,, pp. 131–149). New York: Thieme.
Langevin, M., & Kully, D. (2012). The Comprehensive Stuttering Program and its evidence base. In M. Onslow, & S. Jelčić Jakšić (Eds.), The science and practice of
stuttering treatment: a symposium (pp. 115–129). Oxford: Wiley-Blackwell.
Langevin, M., Packman, A., & Onslow, M. (2009). Peer responses to stuttering in the preschool setting. American Journal of Speech-Language Pathology, 18, 264–278.
https://doi.org/10.1044/1058-0360(2009/07-0087)
Langevin, M., Packman, A., & Onslow, M. (2010). Parent perceptions of the impact of stuttering on their preschoolers and themselves. Journal of Communication
Disorders, 43, 407–423. https://doi.org/10.1016/j.jcomdis.2010.05.003
Langevin, M., & Prasad, N. G. N. (2012). A stuttering education and bullying awareness and prevention resource: A feasibility study. Language, Speech and Hearing
Services in Schools, 43(3), 344–358. https://doi.org/10.1044/0161-1461(2012/11-0031)
Langner, J., & Fukkink, R. G. (2023). A realist synthesis of international collaborative practices in early intervention for children with speech, language and
communication needs. International Journal of Language & Communication Disorders, 58, 516–541.
Lemire, S., Kwako, A., Nielsen, S. B., Christie, C. A., Donaldson, S. I., & Leeuw, F. L. (2020). What is this thing called a mechanism? Findings from a review of realist
evaluations. New Directions for Evaluation, 167, 73–86.
Marcum, J. A. (2012). An integrated model of clinical reasoning: Dual-process theory of cognition and metacognition. Journal of Evaluation in Clinical Practice, 18(5),
954–961.
Maxwell, J. A. (1992). Understanding and validity in qualitative research. Harvard Educational Review, 62(3), 279–300.
Nichol, A. J., Hastings, C., & Elder-Vass, D. (2023). Putting philosophy to work: developing the conceptual architecture of research projects. Journal of Critical Realism,
22(3), 364–383. https://doi.org/10.1080/14767430.2023.2217054
Nicholas, A., & Kelman, E. (2020). Palin Parent-Child Interaction Therapy for Early Childhood Stammering. New York, NY: Routledge,.
Nielsen, S. B., Lemire, S., & Tangsig, S. (2021). Unpacking context in realist evaluations: Findings from a comprehensive review. Evaluation, 28(1), 91–112. https://
doi.org/10.1177/13563890211053032
Norman, G., Young, M., & Brooks, L. (2007). Non-analytical models of clinical reasoning: the role of experience. Medical Education, 41(12), 1140–1145. https://doi.
org/10.1111/j.1365-2923.2007.02914.x
Onslow, M., Webber, M., Harrison, E., Arnott, S., Bridgman, K., Carey, B., Sheedy, S., O’Brian, S., MacMillan, V., Lloyd, W., & Hearne, A. (2023). The Lidcombe
Program Treatment Guide. Lidcombe Program Trainers Consortium. 〈https://www.uts.edu.au/sites/default/files/2023-01/Lidcombe%20Program%20Treatment%
20Guide%202022%20v1.4.%202023-01-26.pdf〉.
Pawson, R., Boaz, A., Grayson, L., Long, A. F., & Barnes, C. (2003). Types and Quality of Knowledge in Social Care. Social Care Institute for Excellence: London, UK.
Pawson, R., & Tilley, N. (1997). Realistic evaluation. London, UK: Sage,.
Perkins, W. H. (1990). What is stuttering? Journal of Speech and Hearing Disorders, 55, 370–382.
Porter, S. (2007). Validity, trustworthiness and rigour: Reasserting realism in qualitative research. Journal of Advanced Nursing, 60(1), 79–86. https://doi.org/
10.1111/j.1365-2648.2007.04360.x
Rameses, I. I. (2017). The RAMESES II Project. Quality standards for realist evaluation: For evaluators and peer-reviewers (Available at 〈https://www.ramesesproject.org/
media/RE_Quality_Standards_for_evaluators_and_peer_reviewers.pdf〉).
Rotter, J. B. (1966). Generalized expectancies for internal versus external control of reinforcement. Psychological Monographs: General and Applied, 80(1), 1–28.
https://doi.org/10.1037/h0092976
Ryan, G., & Rutty, J. (2019). Philosophy and quality? TAPUPASM as an approach to rigour in critical realist research. Nurse Researcher, 27(1), 33–40. https://doi.org/
10.7748/nr.2019.e1590
Sandelowski, M. (1995). Focus on qualitative methods. Sample size in qualitative research. Research in Nursing & Health, 18, 179–183.
14
Sisskin, V. (2023). Disfluency-affirming therapy for young people who stutter: Unpacking ableism in the therapy room. Language, Speech, and Hearing Services in
Schools, 54, 114–119.
Swift, M. C., Langevin, M., & Clark, A. M. (2017). Using critical realistic evaluation to support translation of research into clinical practice. International Journal of
Speech-Language Pathology, 19(3), 335–343. https://doi.org/10.1080/17549507.2017.1309067
Tennant, E., Miller, E., Costantino, K., De Souza, D., Coupland, H., Fotheringham, P., & Eastwood, J. (2020). A critical realist evaluation of an integrated care project
for vulnerable families in Sydney, Australia. BMC Health Services Research, 20, 995. https://doi.org/10.1186/s12913-020-05818-x
Vassilev, I., Rogers, A., Kennedy, A., & Koetsenruijter, J. (2014). The influence of social networks on self-management support: A metasynthesis. BMC Public Health,
14, 719. 〈http://www.biomedcentral.com/1471-2458/14/719〉.
Wampold, B. E. (2015). How important are the common factors in psychotherapy? An update. World Psychiatry, 14(3), 270–277. https://doi.org/10.1002/wps.20238
Willis, C. E., Reid, S., Elliott, C., Rosenberg, M., Nyquist, A., Jahsen, R., & Girdler, S. (2018). A realist evaluation of a physical activity participation intervention for
children and youth with disabilities: What works, for whom, in what circumstances, and how? BMC Pediatrics, 18, 113. https://doi.org/10.1186/s12887-018-
1089-8
Wiltshire, G., & Ronkainen, N. (2021). A realist approach to thematic analysis: making sense of qualitative data through experiential, inferential and dispoitional
themes. Journal of Critical Realism, 20(2), 159–180. https://doi.org/10.1080/14767430.2021.1894909
World Health Organization (WHO) (2007). International Classification of Functioning, Disability and Health: Children & Youth Version. World Health Organization,
Geneva, Switzerland. Accessed: https://iris.who.int/bitstream/handle/10665/43737/9789241547321_eng.pdf.
Wȩsierska, M., Langevin, M., Wȩsierska, K., (2023). Bullying experienced by children who stutter, coping responses and preventive work. Retrieved December 3, 2023
〈https://www.logolab.edu.pl/wp-content/uploads/2023/06/Chapter-11–.pdf〉.
Yaruss, J. S., Coleman, C. E., & Quesal, R. W. (2012). Stuttering in school-age children: A comprehensive approach to treatment. Language, Speech and Hearing Services
in Schools, 43(4), 536–548. https://doi.org/10.1044/0161-1461(2012/11-0044)
Yaruss, J. S., & Quesal, R. W. (2016). OASES: Overall assessment of the speaker’s experience of stuttering: technical-scoring manual. Stuttering Therapy Resources.
Dr Michelle Swift is a Senior Lecturer in Speech Pathology at the University of South Australia and Certified Practising Speech Pathologist working predominantly with
stuttering and cluttering. Data for this research was collected while on an Outside Studies Program with Flinders University.
Dr. Marilyn Langevin is an Adjunct Associate Professor in the Department of Communication Sciences and Disorders, Faculty of Rehabilitation Medicine at the Uni-
versity of Alberta.
15

A theory building critical realist evaluation of an integrated cognitive-behavioural fluency enhancing stuttering treatment for school-age children

Uploaded by

Copyright:

Available Formats

A theory building critical realist evaluation of an integrated cognitive-behavioural fluency enhancing stuttering treatment for school-age children

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A theory building critical realist evaluation of an integrated cognitive-behavioural fluency enhancing stuttering treatment for school-age children

Uploaded by

Copyright:

Available Formats

Journal of Fluency Disorders 82 (2024) 106076

Contents lists available at ScienceDirect

Journal of Fluency Disorders

A theory building critical realist evaluation of an integrated

2.3. Data collection

2.4. Realist thematic analysis

3.2. RM, C, M, O codes

3.4. Resource mechanisms

3.4.1. SLP skills and abilities

3.4.2. Components of a treatment program

3.4.3. Child physiology and age

3.5.1. Family engagement

3.5.2. Parent-child relationship

3.5.3. Parental understanding of program

3.5.4. Parental expectations of fluency

3.5.5. Change in adult support

3.5.6. Demands on parents of children with co-morbidities

3.5.7. School environment

3.6.2. Personality and psychological characteristics

3.6.3. Child buy-in to therapy process

3.6.4. Effort and control

3.7. RMCMO linkages

3.8. Preliminary program theory

4.1. Preliminary program theory and RMCMO linkages

4.2. Constituent factors: outcomes

4.3. Constituent factors: resource mechanisms

4.3.1. Treatment components

4.3.2. Child attributes

4.3.3. SLP capacities

4.4. Constituent factors: contexts

4.4.1. Interaction of contexts

4.4.2. Pivotal role of parents

4.5. Constituent factors: mechanisms

4.6. Clinical implications

4.7. Strengths, limitations and future research

CRediT authorship contribution statement

Declaration of Competing Interest

The authors do not have permission to share data.

Appendix A. Supporting information

You might also like