Self-Reported Communication Attitudes of Children With Childhood Apraxia of Speech: An Exploratory Study

Abstract

Purpose:

Much of the research literature on childhood apraxia of speech (CAS) has focused on understanding, diagnosing, and treating the impairment, rather than examining its functional effect on children's daily lives. This study focuses on the Personal Factors component of the World Health Organization International Classification of Functioning, Disability and Health–Children and Youth Version Framework. Specifically, the purpose was to examine the self-reported communication attitudes of children with CAS.

Method:

Two validated communication attitude questionnaires were administered to 12 children with CAS enrolled in an intensive speech-focused intervention (age range: 4–10 years old). Children's scores were compared to the questionnaires' typically developing norms. Descriptive analyses explored relationships between communication attitude and CAS severity, caregiver perceptions of communicative participation, frustration ratings during therapy, and change in communication attitude over a brief time.

Results:

Older (ages 6–10 years) but not younger (ages 4–5;11 [years;months]) children with CAS were more likely to have negative attitudes about their speech. No clear relationships were observed between communication attitudes and caregiver perceptions of communicative participation; small positive relationships were observed between communication attitude and frustration during therapy. For the younger children, there was also a relationship with CAS severity. For most children, no change in communication attitude was observed over a brief period, though one child appeared to develop more negative and one appeared to develop more positive attitudes.

Conclusions:

These initial findings suggest that older children with CAS may be at greater risk for negative communication attitudes than their peers without CAS. The findings also highlight the need to include more child self-report measures in research. Further implications for CAS assessment and intervention are discussed.

Supplemental Material:

https://doi.org/10.23641/asha.21834432

Capturing child perspectives can provide a unique understanding of the contextual factors that influence how children experience the world. This is particularly important for children with speech sound disorders (SSDs) who face distinct communicative challenges compared to children without communication disorders. SSDs are a group of disorders that cause difficulty with motor production, perception, or phonological representations of speech sounds or syllables (American Speech-Language-Hearing Association [ASHA], 2007).

Childhood apraxia of speech (CAS) is a motor-based SSD that results from inaccurate motor planning and programming of the speech musculature (ASHA, 2007). Specifically, children with CAS have difficulties timing and orienting their speech movements that are not explained by neuromuscular deficits such as weakness (ASHA, 2007; Murray & Iuzzini-Seigel, 2017). The inability to accurately plan movements results in prosodic and stress errors, vowel and consonant distortions, and inconsistency of errors (ASHA, 2007; Murray et al., 2021; Strand et al., 2013), which may affect intelligibility (e.g., Chenausky et al., 2022; Skoog & Maas, 2020). Estimates of CAS prevalence range from one to 10 children per 1,000 (Shriberg et al., 1997, 2019).

CAS puts children at an increased risk of developing literacy problems (e.g., Boada et al., 2022; Lewis et al., 2004; Miller et al., 2019) and may limit both academic and economic outcomes (e.g., Felsenfeld et al., 1994; Lewis et al., 2004; McCormack et al., 2009; Raitano et al., 2004; Skebo et al., 2013). CAS may also affect social and emotional development (Cassar et al., 2022; Lewis et al., 2021; Murray & Iuzzini-Seigel, 2017; Rusiewicz et al., 2018) and often requires more and longer treatment than other SSD (e.g., Campbell, 1999). Given the number of ways that CAS may pose additional challenges for a child, it is critical for researchers and clinicians to consider the impacts of the speech disorder beyond speech skills alone.

The World Health Organization (WHO) International Classification of Functioning, Disability and Health–Children and Youth Version (ICF-CY; WHO, 2007) provides a framework for understanding the complex array of factors that impact functioning (see also Baylor & Darling-White, 2020, for a reconceptualization of the WHO ICF-CY framework specifically focused on communication disorders, in which communicative participation is centered as the ultimate goal of intervention). This framework recognizes the interconnections between the different components and encourages clinicians to approach a client's care holistically. The health condition identifies the disorder or disease (here, CAS). The body functions/structures component refers to specific bodily structures, functions, and processes that are impaired (speech motor planning, in the case of CAS). Those impaired structures/functions are important for a functional activity (e.g., speaking to communicate a message). The ability to perform the functional activity in turn affects the degree to which someone participates in social, vocational, academic, and personal life activities (e.g., the degree to which a child participates in class activities). In addition, functioning is affected by both environmental factors and personal factors. Environmental factors are those influences that play a significant role in the individual's life beyond personal control (e.g., public policy, caregiver support, and availability of resources for therapy). Personal factors are more unique to the person's life such as race, ethnicity, personality, and motivations (Threats, 2006). They can be divided into unchangeable factors (e.g., race, age) and factors amenable to change (e.g., lifestyle habits, attitudes, and coping styles; Howe, 2008).

CAS research, thus far, has almost exclusively addressed the body structures/functions component of the ICF (e.g., Iuzzini-Seigel & Murray, 2017; Maas et al., 2019; Skoog & Maas, 2020), with many studies aimed at finding diagnostic markers (e.g., Murray et al., 2021) and effective treatments targeting the impairment (e.g., Kearney et al., 2015; Maas et al., 2014). Largely neglected from current research on CAS is consideration of the environmental and personal factors components of the WHO ICF-CY model.

This study focuses on personal factors. Personal factors can help researchers and clinicians understand how different individuals are impacted by the same diagnosis (Threats, 2006). Understanding the perspectives of children with CAS may prove useful for informing treatment approaches and for developing functionally relevant goals (Murray & Iuzzini-Seigel, 2017). In Baylor and Darling-White's (2020) reframing of the WHO ICF-CY model, in which communicative participation is centered as the primary goal of intervention, personal perspectives are considered to be one important contributing factor to improving communicative participation, along with (impairment- and activity-level) communication skills.

This initial foray into the personal factors component in relation to CAS specifically explores the communication attitudes of children with CAS—how children with CAS evaluate their own speech. In addition to potentially playing an important role in improving communicative participation (Baylor & Darling-White, 2020), it is important to consider these perspectives to better understand the impact of CAS on children's well-being. Before introducing the specific questions addressed in this study, we first discuss how to measure children's communication attitudes and emphasize the importance of obtaining insights about communication attitudes from children themselves, rather than from caregivers or other proxies.

Self-Reported Measures and CAS

There is growing recognition of the importance of self-reported outcome measures (Baylor & Darling-White, 2020; McCormack et al., 2019; Page & Yorkston, 2022; Yorkston & Baylor, 2019). Recent studies collected self-reported measures from adolescents (Lewis et al., 2021) and adults with a history of CAS (Cassar et al., 2022). Adolescents with CAS reported more social problems (Lewis et al., 2021), and adults with a history of CAS exhibited higher rates of anxiety compared to the normative data (Cassar et al., 2022). However, to our knowledge, no published studies have examined self-reported measures with children with CAS.

Parent perspectives are often used as a proxy for understanding personal factor-level information for children with communication disorders. This may be due to the perceived and inherent difficulties when collecting child perspectives, including questioning the reliability and validity of a child's response and accounting for extra ethical considerations when working with children (Harcourt, 2011). Rusiewicz et al. (2018) used parent measures to evaluate the daily impact of a CAS diagnosis, with parents being particularly concerned about their child's overall intelligibility and ability to navigate social situations. More generally, McCormack, McLeod, Harrison, and McAllister (2010) asked parents and speech-language pathologists (SLPs) to rate how they felt about their children's speech impairment using an ICF-based questionnaire. Both populations expressed greatest concern regarding their children's interpersonal relationships and verbal communication skills.

Solely collecting parent perspectives, however, assumes they are always accurate. In many cases, parents can provide accurate and detailed accounts for their child. However, several studies have highlighted discrepancies between parent and child reports. For example, parents of children with chronic illnesses accurately rated their child's physical functioning but had more difficulty rating their child's internal symptoms as compared to their children's self-reports (e.g., anxiety and sadness; Eiser & Morse, 2001). A study that compared children with orthopedic problems between the ages of 5 and 18 years and their parents' perspectives on the child's physical functioning, mental health, and general health found that children reported their physical functioning higher than their parents and their mental and general health lower than their parents (Matsumoto et al., 2011). Together, these two studies underscore how parent reports may provide both accurate and inaccurate information and, thus, should be interpreted with caution.

Specific to the field of speech-language pathology, investigations into correlations between parent and child communication perspectives have demonstrated that parents are not always aware of their child's true communicative experience. For example, Vanryckeghem (1995) obtained perspectives about the communication attitudes of children who stutter from these children and their parents and found that parents rated their child's communicative attitude as significantly more negative than the children did. Conversely, parents of children who did not stutter rated their children's communication attitudes more positively than their children.

Moving toward a more holistic approach to client care means addressing and appreciating children's perspectives during therapy (Baylor & Darling-White, 2020). As specified by Article 12 of the United Nations Conventions on the Rights of the Child (Cohen, 1989) and the follow-up General Comment No. 7 (United Nations, 2005), children's perspectives should be included in research that affects them. When collected correctly, children can provide unique, distinct insights into their life experiences (Harcourt, 2011). While children's attitudes may change frequently, reflecting their development, this does not mean that their responses at the given assessment are any less relevant or valuable (Harcourt & Einarsdóttir, 2011). Directly asking children with speech and language disorders how they feel may be an effective way of determining if children are receiving adequate support (Roulstone & McLeod, 2011).

While informal methods of collecting communication attitudes may be warranted during individual treatment, there are benefits to using more formal communicative assessments, particularly during intensive impairment-focused treatment protocols. More formal methods allow for comparison to norms and to various clinical populations and may enable systematic monitoring of change over time. Relatively few standardized, validated, and normed self-report measures exist for understanding the communication attitudes of children with communication disorders. One such measure is the Communication Attitude Test for School-Age Children Who Stutter (CAT; Brutten & Dunham, 1989; Brutten & Vanryckeghem, 2007).

A component of the Behavior Assessment Battery for School-Age Children Who Stutter, the CAT is designed to assess the communication attitudes of children ages 6;0 (years;months) and older (Brutten & Dunham, 1989; Brutten & Vanryckeghem, 2007). It consists of 33 true/false statements such as “I like to talk” and “My parents like how I talk” and returns a score that can be compared against normative data. In a recent systematic review, Jones et al. (2021) evaluated existing instruments to assess the psychosocial impact of stuttering and found the CAT to have the strongest measurement properties compared to other measures of the psychological impacts of stuttering.

The Communication Attitude Test for Preschool and Kindergarten Children Who Stutter (KiddyCAT) was developed as an extension of the CAT to assess how preschool children evaluate their own speech (Vanryckeghem & Brutten, 2007). It asks children (ages 3;0–5;11) 12 yes/no questions such as, “Do you like how you talk?” and “Do words sometimes get stuck in your mouth?” Like the CAT, the KiddyCAT returns a score that can be compared against normative data.

Both the CAT and KiddyCAT have standardized administration procedures and normative data samples. Administration of the CAT and KiddyCAT to children who do and do not stutter have revealed that children who stutter are more likely to have negative communication attitudes (Brutten & Vanryckeghem, 2007). The CAT and KiddyCAT have been administered in several countries, including Australia (McCormack et al., 2019), Italy (Bernardini et al., 2009), Belgium (Vanryckeghem & Brutten, 1992; Vanryckeghem et al., 2015), Japan (Kawai et al., 2012), Pakistan (Vanryckeghem & Mukati, 2006), and Sweden (Johannisson et al., 2009). Across studies, generally the same patterns are observed, with low scores (indicating positive communication attitudes) among children who do not stutter and higher scores (indicating more negative communication attitudes) among children who stutter. This convergence provides some support for the validity of these measures across cultures.

Though developed for stuttering, these assessments have also been successfully administered to other populations, including children with phonological disorder (McCormack et al., 2019), articulation disorder (De Nil & Brutten, 1990), voice disorder (De Nil & Brutten, 1990), and cleft palate (Havstam et al., 2011). These applications indicate that the CAT and KiddyCAT can be successfully administered to populations other than children who stutter.

Despite their widespread use, neither the CAT nor KiddyCAT have been administered to children with CAS. One recent dissertation study did ask four children with CAS directly about their social–emotional status related to communication and suggested more negative self-perceived emotional well-being compared to children without CAS (Swartz, 2022). However, this study used a nonstandardized survey and included only older children (9–16 years old). The CAT and KiddyCAT were chosen for this study because they are standardized, have normative data, are designed for children as young as 3 years old, and have been found to be reliable and valid across a variety of languages, cultures, and clinical populations.

Communication Attitudes and CAS

The purpose of this preliminary study is to examine communication attitudes in children with CAS. In particular, we address the questions below.

Do Communication Attitudes of Children With CAS Differ From Those of Children Without Communication Disorders?

Children with CAS are more likely to experience unsuccessful communication interactions than children without communication disorders and may experience social stigma and teasing/bullying by their peers. Such experiences may lead to the development of negative communication attitudes (e.g., De Nil & Brutten, 1991; Havstam et al., 2011). In support of this idea, significantly more negative communication attitudes relative to peers without communication disorders have been reported for children with cleft palate (Havstam et al., 2011), children who stutter (Clark et al., 2012; De Nil & Brutten, 1990, 1991; Vanryckeghem et al., 2005; Winters & Byrd, 2021), and children with voice disorders (De Nil & Brutten, 1990). These populations likely experience such unsuccessful communication interactions and/or social stigma because of the way they communicate.

However, children with articulation or phonological disorders do not appear to have more negative communication attitudes compared to their peers (De Nil & Brutten, 1990; McCormack et al., 2019). One explanation is that such disorders have a more modest impact on communicative interactions than stuttering or voice disorders (De Nil & Brutten, 1990). An alternative possibility is related to age: In some of these prior studies, the children with articulation or phonological disorders were younger (e.g., mean age 8;8 for children with articulation disorders vs. 10;3 for stuttering children [De Nil & Brutten, 1990]; age 4;7 for children with phonological disorders [McCormack et al., 2019]).

It is possible that negative communication attitudes emerge or increase over time, as children experience more unsuccessful communication attempts and become more aware of their communicative difficulties, especially as children enter school and have increased social interactions (Tarshis et al., 2020). Age effects (increasingly negative communication attitudes with increasing age) have been reported for children who stutter between ages 6 and 15 years (e.g., De Nil & Brutten, 1991; Vanryckeghem & Brutten, 1997; see Guttormsen et al., 2015, for a meta-analysis), although not always for younger children who stutter (Vanryckeghem et al., 2005; but see Clark et al., 2012, for a reverse age effect).

It is unknown whether children with CAS exhibit more negative communication attitudes compared to peers without communication disorders, and if so, whether such differences increase with age. Given parent reports of communication challenges and frustration among children with CAS (Rusiewicz et al., 2018) and the possible effects of such experiences on communication attitudes, we expect more negative communication attitudes in children with CAS compared to peers without communication disorders, in particular in older children (see Swartz, 2022, for data suggesting more negative self-perceived emotional well-being in four older children with CAS).

Do Communication Attitudes of Children With CAS Relate to Impairment Severity?

If the extent (number, seriousness) of communicative breakdowns or other negative communication experiences associated with CAS indeed do play a role in the development of negative communication attitudes, then it can be hypothesized that more severe CAS is associated with more negative communication attitudes, given that children with more severe CAS are more likely to have such experiences. Correlations with impairment severity have been reported for children who stutter older than 6 years (De Nil & Brutten, 1991; Vanryckeghem & Brutten, 1996) though not for children who stutter younger than 6 years (Winters & Byrd, 2021). In addition, McCormack et al. (2019) reported a significant correlation between impairment severity (speech accuracy) and communication attitudes in children with phonological disorders, such that lower accuracy was associated with more negative communication attitudes.

Do Communication Attitudes of Children With CAS Relate to Communicative Participation?

It is reasonable to expect a relationship between communication attitudes and communicative participation, although this relationship is likely complex and influenced by other personal factors such as a child's personality and “grit” (e.g., Duckworth & Quinn, 2009) as well as environmental factors (e.g., the support and attitudes of communication partners). Negative communication attitudes may lead children to be reluctant to participate in communicative situations in an effort to avoid (further) negative communicative interactions. If so, we expect that children with more negative communication attitudes will have more limited communicative participation than children with more positive communication attitudes. No such relationship was found for children with phonological disorder (McCormack et al., 2019), but this has not yet been examined in children with CAS, whose speech communication experiences may differ from children with more predictable speech error patterns such as those with phonological disorders.

Do Communication Attitudes of Children With CAS Relate to Frustration During Therapy?

Children's communication attitudes may influence how they respond to speech therapy. Negative communication attitudes may make children more sensitive and less resilient to the negative feedback typically associated with impairment-focused speech therapy (e.g., correcting speech errors). This may result in greater expressions of frustration during impairment-focused speech therapy. In turn, increased frustration during speech therapy may negatively impact its effectiveness (e.g., due to reduced time spent on therapy activities; reduced motivation), in addition to being an undesirable effect in itself for children's well-being and potentially exacerbating their negative communication attitudes.

Do Communication Attitudes of Children With CAS Change Over a Brief Period?

Although most interventions for CAS do not directly or explicitly target communication attitudes, experiences during speech therapy may affect a child's communication attitude, positively or negatively. Before such influences can be fully addressed, it is important to determine the stability of children's communication attitudes over a brief time. Although the KiddyCAT has strong test–retest reliability for children who stutter and children without communication disorders (Vanryckeghem et al., 2015), the stability of communication attitudes of children with CAS is unknown. This study will take a preliminary look at changes over time, in order to inform future studies with larger sample sizes that can establish stability as well as effects of treatment on communication attitudes of children with CAS.

This Study

Given the potential impact of speech disorders on psychosocial well-being (e.g., McCormack, McLeod, McAllister, & Harrison, 2010; Rusiewicz et al., 2018), the often slow progress and extensive and intensive speech therapy needed for CAS (e.g., Campbell, 1999; Davis & Velleman, 2000; Shriberg et al., 1997), and a recent report that negative psychosocial effects of CAS continue into adulthood (Cassar et al., 2022), it is important to investigate children's communication attitudes. This study is the first to use self-reported measures with children with CAS aged 4–10 years old in order to fill this critical gap in the literature. The CAT and KiddyCAT were administered to children with CAS who were enrolled in a randomized controlled trial of (impairment-focused) speech therapy for CAS, which was conducted as an intensive summer camp. Although not originally and specifically designed for this purpose, this ongoing study afforded the setting and opportunity to explore the research questions (RQs) above and reiterated below:

1. Do communication attitudes of children with CAS differ from those of children without communication disorders based on the tests' normative values?

2. Do communication attitudes of children with CAS relate to impairment severity?

3. Do communication attitudes of children with CAS relate to communicative participation?

4. Do communication attitudes of children with CAS relate to frustration during therapy?

5. Do communication attitudes of children with CAS change over a brief period?

Method

Study Context

Data for this study were collected in the context of a larger study involving a randomized controlled trial of ASSIST (Apraxia of Speech Systematic Integral Stimulation Treatment; Maas et al., 2022, 2019) for children with CAS (ClinicalTrials ID NCT03903120). ASSIST is an impairment-focused speech treatment based on integral stimulation (“watch me, listen to me, say what I say”) and principles of motor learning (Maas et al., 2008). Speech targets are words and phrases selected based on personal functional relevance and contain only sounds in the child's phonetic inventory. Targets are embedded in carrier phrases to facilitate variable practice and focus on prosody. Treatment involves drill-based repetition of targets, with clinician feedback and cues to help the child achieve movement patterns for accurate speech.

In the larger study, children participated in an intensive 4-week virtual summer camp program, during which they received 16 hr of individual ASSIST in 32 sessions of 30 min each. Children were randomly assigned to receive all their individual treatment in the first 2 weeks (immediate massed), the second 2 weeks (delayed massed), or divided over all 4 weeks (distributed; see Figure 1 for a design overview). Group activities and individual ASSIST were delivered via videoconference. Treatment was delivered by trained SLP graduate student clinicians, supervised by experienced and licensed SLPs. Before (T1), at midpoint of (T2), and after (T3) the 4-week period, children attended individual data collection sessions in person in the Speech, Language, and Brain Lab at Temple University. Details and results from the larger clinical trial are not further reported here, except insofar as relevant to this study on child-reported outcome measures (CAT and KiddyCAT), which were administered at T2 and T3. ¹ All study procedures were approved by the Temple University Institutional Review Board (Protocol 25807); informed consent was obtained from at least one parent, and all children provided assent.

Figure 1.

Design of larger study that provides the context and setting for this study. Data collection for this study occurred in Week 4 (T2) and Week 7 (T3). ASSIST = Apraxia of Speech Systematic Integral Stimulation Treatment.

Participants

Twelve children ages 4–10 years with a primary diagnosis of CAS participated in this study. Three expert SLPs provided independent ratings of presence of CAS, using a 3-point scale (0 = no CAS, 1 = possible CAS, 2 = CAS). All expert SLPs had 10+ years of experience working clinically with children with CAS and have been actively involved in CAS research; several have also been actively involved as faculty in Apraxia Kids' Boot Camp for SLPs (https://www.apraxia-kids.org/intensive-training/). Their independent CAS ratings were based on the presence of current consensus features of CAS (ASHA, 2007), including inconsistent consonant and vowel errors, difficulties achieving or transitioning into articulatory configurations, and abnormal prosody. At least one SLP made this judgment on the basis of the live assessment sessions, whereas the remaining SLPs made their judgments from video recordings of the assessment sessions. To be included in the study, children were required to receive an average rating of > 1 across the three expert SLPs and an apraxia score of 1 or 2 on a maximum performance protocol (Thoonen et al., 1996, 1999), the only prospectively validated protocol to date with acceptable sensitivity and specificity (Murray et al., 2021).

Inclusion criteria were (a) English as primary home language per parent report, (b) normal hearing per parent report and/or pure-tone audiometric screening, (c) nonverbal cognition in the typical range based on the Reynolds Intellectual Assessment Scales (Reynolds & Kamphaus, 2003), and (d) verbal output of 50+ words and communicative intent per SLP and parent report. Exclusionary criteria were (a) co-occurring neurobehavioral diagnosis (e.g., autism) per parent report, (b) significant visual impairment per parent report, (c) primary speech diagnosis of dysarthria based on SLP judgment, (d) oral structural–anatomical abnormalities (e.g., cleft palate) based on parent report and SLP judgment from an oral mechanism evaluation, and (e) inability to operate or attend to virtual sessions per parent and SLP judgment.

A battery of tests were administered prior to allocation to groups (before T1) to determine eligibility and to characterize participant profiles (see Table 1). However, these will not be discussed here, except for the Dynamic Evaluation of Motor Speech Skill (DEMSS; Strand & McCauley, 2019), as the DEMSS score represents an index of CAS impairment severity, relevant to RQ2.

Table 1.

Participant information.

Child	Age a	Sex	Cond. b	CAS rating c	DEMSS score d	GFTA e raw	GFTA e SS	DEAP f Phon. SS	DEAP f Incons.	RIAS NIX g	RIAS NV Memory h	EVT i SS	PPVT j SS	CELF k CLS
304	5;5	M	MI	2.00	295	40	72	75	44%	78	56	98	90	93
306	4;2	M	D	2.00	215	107	49	65	80%	83	63	117	93	NA l
310	4;1	M	D	2.00	DNC m	131	40	CNC n	72%	66	75	83	71	NA l
312	5;9	F	MI	1.67	143	109	40	55	64%	47	81	81	76	79
313	4;11	M	MD	2.00	DNC m	78	49	65	84%	67	59	96	86	NA l
324	5;2	M	MD	1.33	357	76	47	65	52%	53	66	98	118	100
301	6;0	M	D	1.00	383	43	63	60	32%	50	50	82	96	75
305	7;0	F	MD	2.00	382	70	40	55	44%	46	49	86	78	75
309	7;10	M	D	1.33	298	73	40	55	68%	42	64	81	85	66
314	10;0	M	D	2.00	401	22	59	NA l	40%	40	38	80	95	72
321	6;4	M	MI	2.00	262	83	40	55	56%	43	49	89	94	93
322	8;3	F	D	1.00	395	10	70	105	36%	42	61	87	86	93

Note. M = male; F = female.

^a Age at T2 in years;months.

^b Condition: D = Distributed; MI = massed immediate (Phase 1); MD = massed delayed (Phase 2).

^c Mean of three expert speech-language pathologists' independent ratings regarding presence of childhood apraxia of speech (CAS): 0 = no CAS; 1 = possible CAS; 2 = CAS.

^d Score on the Dynamic Evaluation of Motor Speech Skill (Strand & McCauley, 2019).

^e Raw score and standard score (SS) on the Goldman-Fristoe Test of Articulation–Third Edition (Goldman & Fristoe, 2015).

^f Diagnostic Evaluation of Articulation and Phonology (Dodd et al., 2006) phonology standard score and inconsistency percentage.

^g Nonverbal intelligence quotient (NIX) T score based on the Reynolds Intellectual Assessment Scales (Reynolds & Kamphaus, 2003).

^h Nonverbal memory T score based on the Reynolds Intellectual Assessment Scales (Reynolds & Kamphaus, 2003).

ⁱ Standard score (SS) on the Expressive Vocabulary Test–Third Edition (Williams, 2019).

^j Standard score (SS) on the Peabody Picture Vocabulary Test–Fifth Edition (Dunn, 2019).

^k Core Language Score (CLS) standard score on the Clinical Evaluation of Language Fundamentals–Fifth Edition (Wiig et al., 2013).

^l Not administered due to age outside normative range.

^m Did not complete (no total score available).

ⁿ Could not compute (score outside of normative range).

Participants were recruited through local community referrals, existing databases of individuals who expressed interest in research opportunities, community outreach and education events, word of mouth, and advertisements in newsletters and websites. Of the 16 children assessed for eligibility, 12 children met the selection criteria.

Materials and Procedure

CAT and KiddyCAT

As described in the introduction, the CAT (Brutten & Vanryckeghem, 2007) and KiddyCAT (Vanryckeghem & Brutten, 2007) have been used to provide insight into how children with a range of speech disorders feel about their speech. Both the CAT and KiddyCAT include statements/questions where an approximately even number of true and false (or yes and no) responses indicate negative communication attitudes to counter possible response biases. Each answer that indicates a negative attitude is awarded 1 point. Thus, higher scores on both assessments indicate more negative communication attitudes. The average KiddyCAT score for children who do not stutter in the normative sample (n = 63) is 1.79 (SD = 1.78), and the average CAT score for children who do not stutter in the normative sample (n = 538) is 6.38 (SD = 5.21). Per each manual, a score of 5 or higher on the KiddyCAT or a score of 17 or higher on the CAT is indicative of a negative communication attitude. However, clinicians and researchers are also advised to consider and investigate slightly lower scores on both assessments during analyses.

At the end of T2 and T3 data collection sessions, trained research assistants, blinded to treatment status, administered the CAT (for children age 6;0+; n = 6) or KiddyCAT (for children ages 4;0–5;11; n = 6). Caregivers were not present during administration but instead observed from a separate room. Per the CAT/KiddyCAT manuals, children were told that there were no right or wrong answers because the questions were about what they thought about their own speech. Before administration, two practice questions were given to establish adequate understanding of the task. In this study, one participant at T2 was thought to have questionable understanding of the concepts “easy” and “difficult.” Per the KiddyCAT manual, this child was given two jars, one that was difficult to open and one that was easy to open. The child was asked to distinguish which was easy to open and which was difficult to open. The child had no trouble understanding this concept, and the KiddyCAT was then administered.

Each question was read aloud by research assistants for both the CAT and KiddyCAT. Participants were required to respond true/false for the CAT and yes/no for the KiddyCAT. For the CAT, right/wrong and yes/no were also deemed acceptable responses in place of true/false. Administration took approximately 5–15 min for the KiddyCAT and 15–30 min for the CAT. Testing took place in a quiet room, and all sessions were audio- and video-recorded for subsequent analysis and reliability check.

Focus on Outcomes of Communication Under Six

The Focus on Outcomes of Communication Under Six (FOCUS-34; Oddson et al., 2019; Thomas-Stonell et al., 2013, 2012) is a caregiver-rated, 34-question assessment of a child's participation across various communicative contexts. It consists of two parts: Part 1 asks about how well a statement applies to the child (e.g., “My child uses words to ask for things”), and Part 2 asks about how much help a child needs to accomplish communicative tasks (e.g., “My child is included in games by other children”). Item scores range from 1 (not at all like my child [Part 1]/cannot do at all [Part 2]) to 7 (exactly like my child [Part 1]/can always do without help [Part 2]).

The same caregiver completed this form at each time point, regardless of the child's age. While this assessment was designed and normed for children under 6 years of age, there is no significant reason to question its validity for older children, and there is precedent for use of the FOCUS-34 with children older than 6 years (e.g., Rusiewicz et al., 2018).

Frustration Ratings

As part of the larger study's safety protocol, potential negative side effects of treatment were monitored, with a particular emphasis on child frustration. Treating clinicians provided a numerical score reflecting the child's frustration level based on their judgment informed by their knowledge of the child and the child's behavior in the session. Because frustration may be expressed differently (e.g., some children may cry, others may refuse to continue, and others may display avoidance behaviors) and because information about frustration was needed quickly in order to make rapid decisions about child safety and well-being, judgment by treating clinicians was deemed adequate to capture these potentially different expressions in an expedient way (Cook, 2022). Possible scores were 0 (no frustration, full compliance with procedures throughout the session), 1 (some frustration, occasional noncompliance), 2 (significant frustration, frequent noncompliance), and 3 (marked frustration, noncompliance during entire session).

Reliability for the frustration ratings in this sample was not formally assessed. However, ratings were provided by the treating clinician at end-of-day debrief meetings where all children were reviewed and where the clinical supervisor was able to express agreement or disagreement. There were no disagreements between the clinician and their supervisor for any of the children.

Data Analysis

CAT and KiddyCAT questionnaires were scored independently by a second scorer (who was not blinded to condition) from session recordings. Each answer was screened for adequate understanding. A few specific questions were excluded due to missing items ² ; for this reason, all scores were converted to percentages for analysis. ³ Scores of the second scorer were compared with those of the examiner who administered the questionnaires to evaluate interrater reliability. Interrater reliability of total scores was assessed via intraclass correlation coefficients (ICCs) with 95% confidence intervals (CIs), calculated with the irr package (Gamer et al., 2012) in R (R Studio Version 1.2.5042), based on a single-measure, absolute agreement, two-way mixed-effects model (McGraw & Wong, 1996a, 1996b). Resulting ICCs were in the good-to-excellent range (Koo & Li, 2016) for all comparisons (CAT T2: ICC = .896, 95% CI [.397, .985]; CAT T3: ICC = .995, 95% CI [.962, .999]; KiddyCAT T2: ICC = .836, 95% CI [.304–.975]; KiddyCAT T3: ICC = .886, 95% CI [.339–.987]).

Discrepancies between scorers were reviewed and resolved by a third independent scorer and were attributed to human error by the administrator. For all subsequent analyses, scores from the second scorer were used because this scorer had the opportunity to rewatch the video to increase confidence in the validity of the child's response.

Analyses

Given the small sample size, no formal inferential statistical analyses were performed (except for RQ1; see below). Instead, we addressed the RQs descriptively.

RQ1 (comparison with normative values) was addressed both descriptively and inferentially. Specifically, children whose scores were more than 1 SD from the normative mean were identified. In addition, as a more stringent test, individual CAT and KiddyCAT total scores (percentages) at each time point were compared to their respective normative values (also converted to percentages for this comparison) using Crawford et al.'s (2010) single-case statistical methods (using SingleBayes_ES.exe ⁴ ) to determine significant differences (α = .05). Because this study represents the first investigation of self-reported communication attitudes in children with CAS, trends (p < .10) are also reported.

For RQ2 (relation with CAS severity), CAT/KiddyCAT scores were plotted against the DEMSS total score for visual inspection. Note that more severe CAS is reflected by “lower” DEMSS scores. Thus, if more severe CAS is associated with more negative communication attitudes, we expect a negative association, as more negative communication attitudes are reflected by “higher” CAT/KiddyCAT scores.

For RQ3 (relation with communicative participation), CAT/KiddyCAT scores at each time point were plotted against the FOCUS-34 total scores from those same time points for visual inspection. If children with more negative communication attitudes are less likely to participate in communicative activities, we expect a negative association, as higher FOCUS-34 scores indicate greater communicative participation.

For RQ4 (relation with frustration during treatment), CAT/KiddyCAT scores were plotted against the mean frustration rating across treatment sessions. If children with negative communication attitudes are more prone to frustration, we expect a positive association.

For RQ5 (change following brief period), descriptive analysis was used given the small and unequal sample sizes in the different treatment conditions. The number of children whose scores increased, remained the same, and decreased numerically were identified.

Results

Results are described below by RQ. Within each RQ, CAT and KiddyCAT results are described separately (see Table 2). For the KiddyCAT, one participant (P306) did not return for T3 testing; thus, analyses involving T3 are based on data from the remaining five children in this age range.

Table 2.

Raw and percentage CAT and KiddyCAT scores at T2 and T3.

Test	Child	Condition	T2		T3		Change
			Score/possible	% Score	Score/possible	% Score
CAT	301	D	18/33	54.5	16/33	48.5	−6.0%
	309	D	13/31	41.9	12/32	37.5	−4.4%
	314	D	7/33	21.2	1/32	3.1	−18.1%
	322	D	10/32	31.3	13/33	39.4	8.1%
	321	M1	12/33	36.4	13/33	39.4	3.0%
	305	M2	9/33	27.3	17/33	51.5	24.2%
KiddyCAT	306	D	2/12	16.7
	310	D	3/11	27.3	3/12	25.0	−2.3%
	304	M1	1/12	8.3	1/12	8.3	0.0%
	312	M1	4/12	33.3	3/12	25.0	−8.3%
	313	M2	2/12	16.7	2/12	16.7	0.0%
	324	M2	0/12	0.0	0/12	0.0	0.0%

Note. CAT = Communication Attitude Test for School-Age Children Who Stutter; KiddyCAT = Communication Attitude Test for Preschool and Kindergarten Children Who Stutter.

RQ1: Comparison to Normative Values

Among the younger children (KiddyCAT), one child fell more than 1 SD above the normative range at T2, but none did at T3 (normative sample mean = 14.92%, SD = 14.83%). One child (P324) fell below the normative range at both time points. ⁵ Among the older children (CAT), three of six children had scores > 1 SD above the normative mean at T2, and five of six children at T3 (normative sample mean = 19.33%, SD = 15.79%). One child showed a lower (more positive) score at T3 compared to the normative sample (see Figure 2).

Figure 2.

CAT (top) and KiddyCAT scores (bottom) at T2 and T3. Horizontal black line represents the normative sample mean; gray shading indicates normative sample standard deviation. Distributed = distributed condition (treatment in Weeks 2–3 and 5–6); MI = massed immediate condition (treatment in Weeks 2–3); MD = massed delayed condition (treatment in Weeks 5–6). Asterisks (*) indicate a significant difference (p < .05), and caret (^) indicates a trend toward a significant difference (p < .10) from the test's normative sample. CAT = Communication Attitude Test for School-Age Children Who Stutter; KiddyCAT = Communication Attitude Test for Preschool and Kindergarten Children Who Stutter.

When analyzed statistically using Crawford et al.'s (2010) single-case statistics, two children demonstrated significantly higher scores on the CAT than the normative sample: P301 at T2 (with a trend at T3) and P305 at T3. No other children differed significantly at either T2 or T3, for either the CAT or the KiddyCAT (see Table 3 for details).

Table 3.

Comparisons of each individual child's percentage score against the test's normative sample using Crawford et al.'s (2010) single-case statistics.

Test	Child	T2				T3
		Score	Z	95% CI	p	Score	Z	95% CI	p
CAT	301	54.5%	2.23	[2.08, 2.38]	.026	48.5%	1.85	[1.71, 1.98]	.065
	305	27.3%	0.51	[0.42, 0.59]	.614	51.5%	2.04	[1.89, 2.18]	.042
	309	41.9%	1.43	[1.31, 1.55]	.154	37.5%	1.15	[1.05, 1.26]	.251
	314	21.2%	0.12	[0.04, 0.20]	.906	3.1%	−1.03	[−1.13, −0.93]	.309
	321	36.4%	1.08	[0.98, 1.18]	.281	39.4%	1.27	[1.16, 1.38]	.205
	322	31.3%	0.76	[0.67, 0.85]	.449	39.4%	1.27	[1.16, 1.38]	.205
KiddyCAT	304	8.3%	−0.45	[−0.70, −0.19]	.659	8.3%	−0.45	[−0.70, −0.19]	.659
	306	16.7%	0.12	[−0.13, 0.37]	.906	N/A
	310	27.3%	0.84	[0.55, 1.12]	.411	25.0%	0.68	[0.40, 0.95]	.503
	312	33.3%	1.24	[0.91, 1.57]	.224	25.0%	0.68	[0.40, 0.95]	.503
	313	16.7%	0.12	[−0.13, 0.37]	.906	16.7%	0.12	[−0.13, 0.37]	.906
	324	0.0%	−1.01	[−1.31, −0.70]	.322	0.0%	−1.01	[−1.31, −0.70]	.322

Note. Included are the Z value, 95% confidence interval (CI), and p values. Bold underline font indicates significant difference from normative sample (p < .05); italic underlined font indicates trend toward significant difference from normative sample (p < .10). CAT = Communication Attitude Test for School-Age Children Who Stutter; KiddyCAT = Communication Attitude Test for Preschool and Kindergarten Children Who Stutter; N/A = not available.

RQ2: Relationship With CAS Severity

Inspection of the scatter plots in Figure 3 reveals no clear relationship between CAT and DEMSS scores at either T2 or T3. However, for the KiddyCAT, a clear and linear relationship in the expected direction is evident for both T2 and T3. Note that these scatter plots are based on small sample sizes (n = 4 for T2, n = 3 for T3) due to missing T3 data for P306 and incomplete DEMSS scores for two other children (P310 and P313).

Figure 3.

Scatter plot of DEMSS scores and CAT (top) and KiddyCAT (bottom) percentages. Blue open squares represent T2 data; filled red circles represent T3 data. Higher DEMSS scores indicate more severe CAS, and higher CAT/KiddyCAT scores indicate more negative attitudes. DEMSS = Dynamic Evaluation of Motor Speech Skill; CAT = Communication Attitude Test for School-Age Children Who Stutter; KiddyCAT = Communication Attitude Test for Preschool and Kindergarten Children Who Stutter.

RQ3: Relationship With Communicative Participation

Inspection of the scatter plot in Figure 4 suggests a possible negative association between CAT and FOCUS-34 scores at T2, reflecting greater communication participation for children who have more positive communication attitudes. However, this relationship was not as clear at T3. For the KiddyCAT, no clear relationship was discernible at either T2 or T3.

Figure 4.

Scatter plot of FOCUS-34 scores and CAT (top) and KiddyCAT (bottom) percentages. Blue open squares represent T2 data; filled red circles represent T3 data. Higher FOCUS-34 scores indicate greater communicative participation, and higher CAT/KiddyCAT scores indicate more negative attitudes. FOCUS-34 = Focus on Outcomes of Communication Under Six, 34-question assessment; CAT = Communication Attitude Test for School-Age Children Who Stutter; KiddyCAT = Communication Attitude Test for Preschool and Kindergarten Children Who Stutter.

RQ4: Relationship With Frustration During Treatment

Visual inspection of the scatter plots in Figure 5 suggests a possible, small positive relationship between CAT score and mean frustration ratings at T2 and T3, reflecting slightly greater frustration during speech therapy for children with more negative communication attitudes. For the KiddyCAT, a similar but clearer relationship was observed at both T2 and T3.

Figure 5.

Scatter plot of mean frustration ratings and CAT (top) and KiddyCAT (bottom) percentages. Blue open squares represent T2 data; filled red circles represent T3 data. Higher frustration ratings indicate greater frustration experienced during treatment, and higher CAT/KiddyCAT scores indicate more negative attitudes. CAT = Communication Attitude Test for School-Age Children Who Stutter; KiddyCAT = Communication Attitude Test for Preschool and Kindergarten Children Who Stutter.

RQ5: Change Following a Brief Period

Inspection of Figure 2 and Table 2 reveals that most older children showed minimal change from T2 to T3. Exceptions were P322 (distributed), who showed a notable decrease over time (−18.1%), and P305 (massed delayed), who showed a notable increase (24.2%). For younger children (KiddyCAT), scores mostly remained the same or decreased slightly across time points; one child (P312, massed immediate) showed a more notable decrease between T2 and T3.

Discussion

This study is the first to examine self-reported communication attitudes of children with CAS. Because of the small sample sizes and the different treatment conditions, the results must be considered preliminary and interpreted with caution. For RQ1–RQ4, we focus our discussion primarily on data from T3, because at this time point, the children are most comparable relative to intervention status (i.e., all children are posttreatment at T3).

Communication Attitudes of Children With CAS

As a first step, we considered whether these children with CAS exhibited more negative attitudes about their speech than children without speech difficulties, as determined by comparing scores to those of the normative samples. Because of the novelty of this research and the potential risk of underestimating the negative impact of CAS, scores that fell outside 1 SD of the normative range were identified as possible negative attitudes.

Among the young children with CAS, most did not feel negatively about their speech. Only one child obtained a score of > 1 SD above the normative mean (at T2 only), and none of the children obtained a score considered to reflect a negative communication attitude (≥ 5 on the KiddyCAT). This is contrary to the findings with young children who stutter (e.g., Clark et al., 2012; Winters & Byrd, 2021; see Guttormsen et al., 2015, for meta-analysis) but on par with previous literature regarding the communication attitudes of young children with phonological disorder (McCormack et al., 2019). If these results are replicated in a larger sample, this might suggest that, despite differences in type of impairment, young children with CAS may be more similar to children with phonological disorders than to children who stutter with respect to communication attitudes, possibly due to fewer or less negative communicative experiences.

In fact, there were several KiddyCAT questions where all children's answers were indicative of positive speech attitudes. Questions 8 and 11 on the KiddyCAT both relate to if children think speaking is “difficult” (Q8) and whether words are “hard” to say (Q11). Answers to these two questions are demonstrative of awareness, or lack thereof, of speech difficulties. At T3, when all children already concluded speech therapy, none of the younger children thought that “talking [was] difficult” or that “words [were] hard for them to say.” Additionally, all participants answered “yes” to the question (Q2), “Do you think that you talk right?”

Two KiddyCAT questions reference how children perceive their speech in relation to other people. At T3, all participants answered “yes” to Q3, “Do mom and dad like how you talk?” This may be an indication that children do not equate caregiver concern regarding speech, as demonstrated by caregivers sending their child to an intensive speech treatment, with their caregivers disliking how they talk. ⁶ Additionally, all children answered “yes” to Q7, “Do you talk well with everybody?” Considering the long-term socioemotional difficulties that adolescents (Lewis et al., 2021) and adults with a history of CAS (Cassar et al., 2022) may experience, it is a positive sign that young children in this study felt that they could talk well with everyone. This also suggests that supportive environments and interventions may prevent the development of negative communication attitudes (cf. Baylor & Darling-White, 2020).

In contrast to the younger children, all except one of the school-age children with CAS exhibited scores more than 1 SD above the normative sample mean at T2, T3, or both. One child obtained a score of more than 1 SD below the normative mean at T3 (P314). Two children showed significantly higher scores at one time point, and those scores were also indicative of a negative communication attitude according to the test manual.

CAT Q18, “Other kids wish they could talk like me,” was the only question in which all participants answered “false,” indicating a negative communication attitude. It is worthwhile to note that this question may have a questionable translation to overall communication attitudes. Thinking that others do not want to talk like them may not necessarily denote a negative communication attitude. Rather, children may be drawing different conclusions about the meaning of the question. In future research using the CAT with children with and without CAS, it would be of interest to understand how children without CAS respond to this question and the underlying reasons for those responses.

Several questions on the CAT highlighted positive communication attitudes held by all participants. At both time points, all children responded “false” to Q1, “I don't talk right,” and “true” to Q7, “I like the way I talk.” These questions somewhat parallel Q3, Q8, and Q11 on the KiddyCAT. They demonstrate that, while the older children in this study were more likely to have negative communication attitudes, there are many aspects of their speech they view positively. Additionally, at T3, all children answered “true” to Q9, “My parents like the way I talk.” This aligns with the younger children's responses to the question, “Do your parents like how you talk?” and is an assuring sign that both older and younger children do not feel their parents think negatively of their talking because of their speech difficulties.

Overall, these findings suggest that older children with CAS are more likely to have negative communication attitudes than children without speech disorders. This suggestion is consistent with findings with other pediatric populations with speech disorders, including children with cleft palates (Havstam et al., 2011), children with voice disorders (e.g., De Nil & Brutten, 1990), and children who stutter (Vanryckeghem & Brutten, 2007; see Guttormsen et al., 2015). These authors have argued that older children may become more aware of speech differences once they enter school and therefore develop more negative communication attitudes. This may apply to children with CAS as well (see Tarshis et al., 2020, for discussion of social considerations relative to CAS). However, it is also worth considering that factors more specific to CAS may play a role in these findings. CAS impairments are unpredictable (unlike articulation errors, for example), which can make children more difficult to understand, can interfere with normal conversations and the development of social competencies (Tarshis et al., 2020), and may lead to longer and more challenging treatment. The relatively low incidence and limited awareness of CAS among the general public may also contribute to lack of understanding and support for children with CAS. The potential negative social consequences of these factors may contribute to the development of negative communication attitudes as children with CAS get older, which may possibly persist into adolescence and adulthood (Cassar et al., 2022; Lewis et al., 2021). These initial findings warrant further study with larger sample sizes to determine contributing and mitigating factors for developing negative communication attitudes.

Communication Attitudes and CAS Severity

When CAT and KiddyCAT assessments were related to impairment severity, findings were mixed. Increasing impairment severity was expected to be associated with negative communication attitudes. However, no such association was evident for the older children. In contrast, the younger children showed a clear, linear relationship in the expected direction, such that children with more severe CAS reported more negative communication attitudes. This finding, though preliminary given the small sample size, is important and deserves further study. If replicated in a larger sample, clinicians and families will be better able to provide necessary support for the child. Finally, the fact that no consistent patterns were observed across age groups underscores the importance of recognizing the complex and changing relationship between body structures/functions and other components of the WHO ICF-CY model and the need to obtain measures that capture these components.

Communication Attitudes and Communicative Participation

Communication attitudes may influence and be influenced by social interactions and the degree to which children participate in communicative activities (e.g., Baylor & Darling-White, 2020; De Nil & Brutten, 1991; Havstam et al., 2011). Furthermore, though designed to assess the Participation component of the ICF, the FOCUS-34 also contains items that reflect aspects of the Personal Factors component (Thomas-Stonell et al., 2012) either directly (e.g., “My child is comfortable when communicating”) or indirectly (e.g., “My child joins in conversations with her/his peers”). As such, we expected to observe a negative relationship between FOCUS-34 score and CAT/KiddyCAT scores. A hint of such a relationship was noted for the older children at T2, but not for T3, nor for the younger children. Furthermore, two children who both showed small increases in communicative participation from T2 to T3 showed opposite changes in communication attitudes, with one showing more negative (P305) and one more positive (P314) communication attitudes at T3 than at T2. Thus, these preliminary findings do not provide support for a relationship between communication attitude and communicative participation.

If replication with a larger sample size is consistent with these observations (i.e., no relationship), this would further support that these assessments measure two distinct constructs and that caregivers' perceptions of participation across contexts cannot be used as indicators of the communication attitudes of children with CAS. It is worth noting that a study with preschool-age children with phonological disorder also failed to find a significant correlation between caregivers' perceptions of participation across contexts and children's communication attitudes (McCormack et al., 2019).

In conjunction, these findings suggest caregiver ratings of child participation are not adequate substitutes for information gained from self-report assessments of communication attitudes (Vanryckeghem, 1995). Anecdotally, some caregivers spontaneously expressed surprise at their child's answers, indicating that they would have answered those same questions very differently for their child yet noting that they believed their child to have answered the questions truthfully. Although anecdotal (as we did not formally ask caregivers to complete the CAT/KiddyCAT on behalf of their child), these observations suggest the importance of self-reported measures rather than relying on caregiver report.

Communication Attitudes and Frustration During Therapy

Children with more negative communication attitudes appeared more likely to express frustration during therapy, as judged by the treating clinician. This relationship was noted for both preschool- and school-age children with CAS but appeared stronger for the younger children. Frustration during sessions could have been due to several reasons (e.g., general unhappiness, tiredness, boredom, and stress) and need not reflect a relationship with communication attitude. To the extent that there is a relationship, the causal direction of this relationship is unclear. On the one hand, more negative communication attitudes could increase a child's sensitivity and reduce their resilience to the corrective feedback that is integral to most impairment-focused interventions such as ASSIST and thereby result in greater frustration. On the other hand, higher levels of frustration during treatment could also have exacerbated negative communication attitudes, for example, by increased awareness of speech difficulties. This is an important consideration, because it may necessitate changes to treatment procedures or intensity to achieve optimal balance between reducing the impairment in speech skills (body function) and fostering positive attitudes about communication (personal factor) in order to maximize participation (Baylor & Darling-White, 2020). Given the small sample size and relatively weak associations for the older children, more research is needed before drawing further conclusions.

Communication Attitudes and Change Over a Brief Period

When younger children's scores were compared across time points, no clear trends emerged. KiddyCAT scores remained essentially the same for all children, regardless of treatment condition (all change < 10%). Whether a child received treatment in the first 2 weeks (M1), last 2 weeks (M2), or throughout all 4 weeks (D) did not appear to impact communication attitudes for younger children. Although this may imply that the ASSIST protocol and camp format did not change how young children viewed their speech, interpretation must await further study of stability (test–retest reliability) in the absence of treatment and comparison of children who did and did not receive treatment.

For most of the older children with CAS, change was minimal (< 10%). However, two children showed greater change. One child (P314), in the distributed condition, showed a more positive communication attitude at T3 than at T2, whereas another child (P305) showed a more negative communication attitude at T3 than at T2. Interestingly, P305 was the only older child in the massed-delayed condition. Although speculative at this time given the small and uneven number of children in each condition, this raises the possibility that the intensive treatment received by P305 between T2 and T3 may have heightened awareness of communication differences and led to a more negative communication attitude. On the other hand, P314 showed a notably more positive communication attitude at T3 than at T2. Anecdotally, in response to some questions, he spontaneously commented that he had improved (e.g., in response to Q24 “Sometimes I have trouble talking,” he said “That was true but now not true”). Thus, in his case, it appears that impairment-focused intervention may have improved his speech skills, which may have increased his confidence and his communication attitude. It is unclear at present whether these changes are meaningful or represent variation unrelated to intervention, and if meaningful, what factors might predict a more positive or more negative communication attitude. P305 also experienced greater levels of frustration during treatment than P314, as judged by the treating clinician. Coincidentally, both P305 and P314 were treated by the same clinician, suggesting that the difference is not obviously related to clinician differences. Whether positive or negative, changes in communication attitude may relate to experiences during impairment-focused intervention (cf. Baylor & Darling-White, 2020). This possibility for impact on communication attitudes has important clinical implications, and further research is needed to determine whether and how the treatment protocol or intensity impacts how children view their speech.

Limitations and Future Directions

As the first investigation into self-reported communication attitudes of children with CAS, there were several limitations to this study. First, this study had a small sample size; the CAT and KiddyCAT were only administered to a total of 12 children with CAS (six per test version), limiting the potential generalizability of these observations. Second, only children with CAS were included in this study, and as such, scores were only compared to established norms and not peers without communication disorders tested under the same circumstances. Third, due to logistical challenges, no communication attitude assessments were administered at the first data collection time point (T1), and the first 2 weeks of therapy for those in the massed-immediate group may have influenced scores at T2.

Despite these limitations, this study indicates that further study of self-reported communication attitudes (and other personal factors, e.g., resilience/grit; Duckworth & Gross, 2014) is warranted, given the preliminary findings of associations with CAS severity in younger children, more negative communication attitudes in older children with CAS, a possible relationship with frustration in speech therapy, and the lack of relationship with a caregiver-rated measure of communicative participation. In addition, the potential impact of impairment-focused intervention on communication attitudes warrants further study.

Future research into communication attitudes should include a larger sample size of children with and without CAS and should obtain self-reported data prior to treatment. In addition, obtaining parallel caregiver-reported measures would be important to determine the validity of caregiver proxies to assess communication attitudes of children with CAS.

Given that this was the first study to administer the CAT and KiddyCAT to children with CAS, several concerns regarding these assessments arose. First, the use of true/false responses on the CAT is unnecessary and can be confusing. While school-age children may be capable of responding to true/false statements, it would be clearer for the CAT to adopt yes/no statements, as on the KiddyCAT. In this study, children often responded with yes/no or right/wrong despite being directed to respond true/false. This is a particularly important consideration for future administration of the CAT to larger samples of children with CAS, as these children are at increased risk for comorbid language disorders (e.g., Iuzzini-Seigel et al., 2022).

Wording on specific CAT questions should also be reconsidered for future studies. For example, some children had difficulty understanding the meaning of Q2: “I don't mind asking the teacher a question in class.” For young school-age children, such language may pose unnecessary comprehension challenges. For example, one participant answered “false, I don't mind,” and another participant responded “true, I don't mind,” but when asked to restate their responses, both clearly meant that they had no reservations about asking the teacher a question in class. Future revisions should consider changing such complex language (e.g., “I'm okay with…” instead of “I don't mind…”).

Future studies should also consider including a qualitative component (e.g., McCormack et al., 2019; McLeod, 2004). Several participants volunteered comments regarding their communication attitudes during data collection, providing context that quantitative assessments cannot provide. For instance, one participant (P314) made various remarks regarding his speaking improvements since the beginning of camp including “I'm working on [speaking] and getting better at it” and remarked that reading out loud in class “was hard but now it's easy.” Incorporating a qualitative component in future studies of children with CAS will help provide reasoning as to why children feel the way they do, which in turn may help identify ways to mitigate negative communication attitudes and foster development of positive attitudes.

Given that several concerns emerged when the KiddyCAT and CAT were administered to children with CAS, future research would benefit from the development of a new self-report speech attitudes measure designed for children with SSDs or adaptation of existing instruments (e.g., McLeod, 2004) for this purpose. This measure should include questions better geared toward SSDs and be adaptable for administration to children with comorbid disorders.

Conclusions

This first study of self-reported communication attitudes of children with CAS begins to fill an important gap in the literature and help researchers and clinicians understand how children with CAS feel about their speech. In this study, scores of most school-age children with CAS were indicative of negative communication attitudes. Communication attitudes of preschool- and school-age children showed no obvious associations with communicative participation but small relationships with frustration during speech therapy. Younger children also revealed a possible relationship between communication attitudes and CAS severity. Finally, while communication attitudes remained stable across the two time points for most children, one child did show a notably more negative communication attitude following 2 weeks of intensive impairment-focused intervention, and another child showed a notably more positive communication attitude. Though preliminary, these findings have important implications for clinical management and environmental support. An SLP who identifies a child with CAS to have negative communication attitudes may more adeptly create relevant treatment goals. Families who are aware of their child's attitudes will be better able to provide psychological support. Including communication attitude measures in future studies is a critical step toward understanding how to best treat and support children with CAS from a holistic perspective.

Data Availability Statement

De-identified CAT and KiddyCAT data used in this study are provided in Supplemental Material S1. De-identified FOCUS-34 and frustration data can be obtained from the corresponding author (Edwin Maas, emaas@temple.edu).

Funding Statement

Preliminary data were presented at the Conference on Motor Speech (Charleston, SC, February 2022) and the Apraxia Kids Research Symposium (Las Vegas, NV, July 2022), and a version of this work was published in a non–peer-reviewed archive related to a Livingstone Undergraduate Research Award to the first author.