Skip to main content
NCBI home page
Journal of Speech, Language, and Hearing Research : JSLHR logoLink to Journal of Speech, Language, and Hearing Research : JSLHR
. 2023 Feb 16;66(3):791–803. doi: 10.1044/2022_JSLHR-22-00436

Comorbidity and Severity in Childhood Apraxia of Speech: A Retrospective Chart Review

Karen V Chenausky a,b,✉, Becky Baas c, Ruth Stoeckel c, Taylor Brown c, Jordan R Green a,d, Cassandra Runke e, Lisa Schimmenti e, Heather Clark c
PMCID: PMC10205100  PMID: 36795544

Abstract

Purpose:

The purpose of this study was to investigate comorbidity prevalence and patterns in childhood apraxia of speech (CAS) and their relationship to severity.

Method:

In this retroactive cross-sectional study, medical records for 375 children with CAS (M age = 4;9 [years;months], SD = 2;9) were examined for comorbid conditions. The total number of comorbid conditions and the number of communication-related comorbidities were regressed on CAS severity as rated by speech-language pathologists during diagnosis. The relationship between CAS severity and the presence of four common comorbid conditions was also examined using ordinal or multinomial regressions.

Results:

Overall, 83 children were classified with mild CAS; 35, with moderate CAS; and 257, with severe CAS. Only one child had no comorbidities. The average number of comorbid conditions was 8.4 (SD = 3.4), and the average number of communication-related comorbidities was 5.6 (SD = 2.2). Over 95% of children had comorbid expressive language impairment. Children with comorbid intellectual disability (78.1%), receptive language impairment (72.5%), and nonspeech apraxia (37.3%; including limb, nonspeech oromotor, and oculomotor apraxia) were significantly more likely to have severe CAS than children without these comorbidities. However, children with comorbid autism spectrum disorder (33.6%) were no more likely to have severe CAS than children without autism.

Conclusions:

Comorbidity appears to be the rule, rather than the exception, for children with CAS. Comorbid intellectual disability, receptive language impairment, and nonspeech apraxia confer additional risk for more severe forms of CAS. Findings are limited by being from a convenience sample of participants but inform future models of comorbidity.

Supplemental Material:

https://doi.org/10.23641/asha.22096622

Childhood apraxia of speech (CAS) is a developmental speech motor disorder that affects the planning and programming of speech movements, in the absence of abnormal reflexes or muscle tone (American Speech-Language-Hearing Association [ASHA], 2007). It often results in imprecise, inconsistent, and unintelligible speech but is distinct from other speech disorders (e.g., stuttering, phonological disorder, childhood dysarthria) and from language disorders. To differentially diagnose CAS, a speech-language pathologist (SLP) with specific expertise in pediatric motor speech disorders administers a thorough speech assessment, identifying discriminative features that include, but are not limited to, the three consensus criteria developed by ASHA: (a) lengthened and disrupted coarticulatory transitions between speech sounds, (b) inappropriate prosody, and (c) inconsistent errors (ASHA, 2007; Maas et al., 2012). There is increasing agreement among clinical researchers as to the specific characteristics associated with CAS (Chenausky et al., 2020; Iuzzini-Seigel, Hogan, Guarino, & Green, 2015; Murray et al., 2015), including those that may be more discriminative of the disorder (Strand, 2017, 2020; Strand & McCauley, 2019).

CAS Etiology

CAS can occur on its own as an idiopathic disorder or in the context of other neurodevelopmental disorders (NDDs). As a sole diagnosis, it is estimated to affect one to two children per thousand (Shriberg et al., 1997). However, recent work has shown that CAS is prevalent at much higher rates in NDDs such as galactosemia (Shriberg, Potter, & Strand, 2011), 16p11.2 deletion syndrome (Fedorenko et al., 2016; Mei et al., 2018), FOXP2-related conditions (Morgan et al., 2017), 22q11.2 deletion syndrome (Baylis & Shriberg, 2019), minimally verbal autism (Chenausky et al., 2019), and SETBP1 haploinsufficiency disorder (Morgan et al., 2021). This and related work involving genetic testing of cohorts of children with CAS points to a genetic etiology for CAS, with monogenic pathogenic variants identified in a minority of investigated cases (Kaspi et al., 2022). In fact, many of the genes that have so far been associated with CAS are also associated with other NDDs via molecular pathways that affect brain network development (Hildebrand et al., 2020). This suggests that CAS may more often be part of a wider ranging NDD than an isolated speech impairment and will thus commonly occur with comorbid diagnoses.

Comorbidity and Severity in CAS

Previous studies have identified several neurodevelopmental conditions that commonly co-occur with CAS, such as language impairment (LI; Bornman et al., 2001; Iuzzini-Seigel et al., 2017; Lewis et al., 2004; Murray et al., 2019; Stein et al., 2020), autism spectrum disorder (ASD; Tierney et al., 2015; Velleman et al., 2010), and fine or gross motor impairment (Duchow et al., 2019; Iuzzini-Seigel, 2019; Teverovsky et al., 2009). However, little research exists documenting how these comorbid diagnoses affect the severity and course of CAS (which we here consider the main disorder).

Research in other neurodevelopmental conditions suggests that comorbidity and severity are closely intertwined and may be related to genetic mutational load. For example, in a study including 168 children with a speech sound disorder (SSD) and 69 without SSD, Lewis et al. (2011) found that children with SSD and LI had poorer academic and language outcomes than children with SSD alone, though children with combined SSD and LI did not present with more severe SSD than children without comorbid LI. Lewis et al. (2011) also examined their participants from the point of view of the severity of their SSD by dividing them into five groups, ranging from “moderately severe SSD” to “no SSD.” The moderately severe group scored worse on tests of language and phonological memory, but not on performance IQ, than children in the other groups. The moderately severe group also showed the largest number of comorbid conditions.

Other recent work in autism has shown that differences in ASD severity and the severity of associated intellectual disability in children with both conditions reflect contributions from both de novo and inherited genetic conditions (Robinson et al., 2014). In that study, the greatest degree of language and behavioral impairment was seen in individuals with high rates of loss-of-function genetic variants (i.e., genetic changes that limit or completely impair the function of the encoded protein) and low rates of psychiatric disease in other family members. Similarly, Parenti et al. (2020) found that the number and severity of the clinical signs observed in children with NDDs overall is positively correlated with the number of mutations identified in their genes.

Taken together, the above findings demonstrate that the presence of comorbidities in NDDs like CAS is inextricably linked to severity and suggest that severity is a useful lens through which to examine CAS. That is, investigating the factors that contribute to CAS severity can reveal much about how different comorbid conditions interact. Because comorbid conditions do not occur randomly, understanding patterns of comorbidity co-occurrence and contributors to CAS severity is essential for generating testable hypotheses about the neural substrates and genetic mechanisms of CAS. To provide initial documentation of comorbidity patterns in a group of complex children with CAS and begin to understand how comorbidity and severity may be related in CAS, we examined data from a large group of children with CAS who were seen for clinical services at Mayo Clinic in Rochester, MN, over a period of 8 years. We had two primary research goals:

  1. To describe patterns of comorbid conditions in a large cohort of children with CAS.

  2. To explore the relationship between the number and type of comorbid conditions and CAS severity.

Method

Participants

The work described in this clinical focus article was conducted with approval from the Mayo Clinic Institutional Review Board, and participating children's legal guardians gave informed written permission for their child's participation before data collection began. Participants were 375 children aged 1;0–18;0 (years;months) at first visit who had been seen for services at Mayo Clinic between 2007 and 2015. While children come to Mayo Clinic from across the United States, historically, approximately 4.3% of patients seen for question of CAS are from the same county where Mayo Clinic is located. Thus, the sample is largely geographically representative. Information about CAS and comorbid conditions was compiled manually from children's medical records.

CAS Diagnosis

CAS diagnosis was ascertained in the context of a comprehensive speech and language diagnostic workup when a child was able to produce enough speech to participate in the assessments. This did not always happen at the first visit; some children with known genetic conditions such as Trisomy 21 were seen by the SLPs before a CAS diagnosis could be made. The workup minimally included standardized tests of receptive and expressive language, a 15-min spontaneous language sample, an evaluation of nonspeech oral function and praxis, a standardized assessment of articulation, and a dynamic motor speech evaluation. Assessments were administered by one of three certified SLPs (B. Baas, R. Stoeckel, and E. Strand).

Because of the known challenges in differential diagnosis of articulation disorder, phonological disorder, and CAS, we briefly explain here how these diagnoses were made. Articulation disorder was used when a child showed a consistent misarticulation of one or two late-developing phonemes (e.g., labialized /r/). Phonological disorder was diagnosed when a child showed a more widespread pattern of phonological changes such as the processes included on the Khan–Lewis Phonological Analysis (Khan & Lewis, 2015). CAS was diagnosed based on the presence of signs of CAS.

CAS Severity Ratings

As mentioned, as part of the assessment, SLPs classified participants into one of three categories according to the severity of their CAS: “mild,” “moderate,” and “severe.” Severity was based on Dynamic Evaluation of Motor Speech Skill (DEMSS) scores (when available), standard scores on articulation tests, and observations of the number and frequency of CAS speech characteristics over all of the assessment tasks.

Reliability

Diagnostic and severity rating reliability were assessed in two ways. First, 85 of the children (23%) were seen for follow-up and received an independent confirmed diagnosis of CAS by a second SLP. Also, as previously reported, the second and third authors independently assessed a group of 81 children with SSD for CAS using the DEMSS (Strand et al., 2013). An intraclass correlation coefficient on total DEMSS score, from which diagnosis and severity categories were derived, was .98, indicating excellent reliability. Descriptive information about participants appears in Table 1.

Table 1.

Participant descriptive statistics and severity group differences.

Variable Severity group
Mild Moderate Severe Full sample
(n = 83)
 (n = 35)
 (n = 257)
 (N = 375)
M (SD) M (SD) M (SD) M (SD)
[min–max] [min–max] [min–max] [min–max]
Age at CAS diagnosis a 5;0 (3;0) 4;6 (2;2) 4;9 (2;9) 4;9 (2;9)
[1;5–15;0] [2;2–10;8] [1;0–18;0] [1;0–18;0]
DEMSS score b 109.4 (37.3) 235.1 (106.7)* 188.8 (106.2)
[54–159] [46–425] [46–425]
Absolute no. of comorbid conditions 7.1 (3.4) 7.0 (3.1) 8.7 (3.3)** 8.2 (3.4)
[0–18] [1–15] [1–20] [0–20]
No. of communication-related comorbid conditions 4.7 (2.1) 4.9 (2.2) 6.0 (2.1)*** 5.6 (2.2)
[0–9] [1–11] [1–13] [0–13]
Open in a new tab

Note. min = minimum; max = maximum; CAS = childhood apraxia of speech; DEMSS = Dynamic Evaluation of Motor Speech Skill; n.s. = not significant.

a

Ages are listed as years;months.

b

DEMSS scores for the reliability sample (n = 19) are reported.

*

Analysis of variance (ANOVA) significant. Severe group versus mild group, p = .002.

**

ANOVA significant. Severe group versus moderate group, p = .012; severe group versus mild group, p = .001; mild group versus moderate group, n.s.

***

ANOVA significant. Severe group versus moderate group, p = .012; severe group versus mild group, p < .0005; mild group versus moderate group, n.s.

Measures

Number of Comorbid Conditions

Comorbidities for each child were tallied manually from medical records and compiled into a large data file, which then underwent a data cleaning and organization process. Since these comorbidities were extracted from physicians' notes and did not always constitute actual medical diagnoses, we refer to them as comorbid conditions (this is discussed in more detail in the Limitations and Future Work section). The data cleaning process involved, first, a step where conditions that may have been listed twice for any one participant were deduplicated in order to get an accurate count of the absolute number of comorbid conditions per participant. Then, when different terms were used across participants for what appeared to be the same condition, the alternate terms were replaced with one standard term. For example, some participants were listed as having “17q21.31 deletion syndrome,” whereas others were listed as having “Koolen-de Vries syndrome.” Since these refer to the same disorder, one code was used for both. This step yielded 241 unique comorbid conditions.

Next, these 241 conditions were classified into one of 25 categories, 22 of which were related to potential communication impairments (all are listed in Figure 1). These categories were determined by the overall behavioral domain or organ system that the individual conditions applied to. For example, both alcohol exposure and polysubstance exposure were combined into one category of “substance exposure.” Since communication-related comorbid conditions were considered more likely to be associated with increased CAS severity, the final step was to tally the number of communication-related conditions for each participant and use that figure in subsequent analyses. All 241 unique conditions, organized into the 25 categories, are listed in Supplemental Material S1 along with the number of participants showing each condition.

Figure 1.

Figure 1.

Open in a new tab

Bar chart of communication-related comorbid conditions by prevalence. Non-communication-related categories were vision-related, cardiac-related, and other (non-communication-related).

Fleiss' kappa was used to assess reliability for classification of the 241 unique comorbid conditions into the 25 categories because it is appropriate for nominal data and represents the degree of agreement over and above what is expected by chance. The first and last authors independently classified a subset of 24 of the 241 unique comorbid conditions (approximately 10%). Overall agreement was κ = .864 (very good). Judges agreed on all but three classifications: Attention-deficit/hyperactivity disorder was classified as “mood/behavioral challenges” by one judge and as “other communication-related” by the other, developmental delay was classified as “other communication-related” by one judge and as “intellectual disability” by the other, and seizure disorder was classified as “other (non-communication-related)” by one judge and as “other communication-related” by the other judge. However, since the categories “mood/behavioral challenges,” “other communication-related,” and “intellectual disability” all counted as communication-related categories, only the disagreement about seizure disorder would potentially have affected the count of number of communication-related comorbid conditions. Therefore, classification reliability was deemed sufficient, and the second judge's classifications were used in later analyses. We return to the issue of counting and classifying comorbid conditions from medical records in the Discussion section.

Analytic Strategy

Descriptive statistics (mean, standard deviation; minimum, maximum) for number of comorbid conditions and number of communication-related comorbid conditions are reported for the three severity groups. Note that these two variables did not meet all six assumptions for the use of analyses of variance (ANOVAs). First, there were significant outliers in the data. Each severity category contained a small number (one to three) of outliers in both variables whose values were greater than 1.5 box-lengths from the edges of their boxes on boxplots. We opted to retain the outliers because the distributions of each group were similar, as determined by a visual inspection of the boxplots, and because the results of our analyses were the same with or without the outliers. Second, our dependent variables are count variables, which are not classified as continuous because they can only be nonnegative integers. Instead, count variables are classified as discrete variables. Although many other count variables (e.g., test scores, number of words from a language sample) are treated as continuous variables and used in ANOVAs in developmental research, we opted to employ both parametric and nonparametric analyses (ANOVAs and Kruskal–Wallis H tests, respectively).

Finally, ordinal regressions were performed to determine whether the number or type of comorbid conditions predicted concurrent CAS severity. To employ this type of regression, the assumption of proportional odds for dependent variables must be met, meaning that the “slope” estimate between the mild and moderate severity groups must not be statistically different from that between the moderate and severe groups. This is indicated by a nonsignificant full likelihood ratio test. If the full likelihood ratio test was significant, then a multinomial regression was used, which treats the dependent variable categories as nominal instead of ordinal. Model fit was assessed for all regressions using the −2 log likelihood test, which determines whether the full model (including the predictor) is associated with a lower −2 log likelihood than the intercept-only model (with no predictors). Nagelkerke pseudo R 2 was used to assess how much variance in the outcome variable was associated with each predictor. The odds of a child with CAS plus a specific comorbid condition having moderate or severe CAS was calculated as the reciprocal of the odds (Exp(B)) of a child with CAS without that comorbid condition having moderate or severe CAS. There are no restrictions on the distribution of independent variables in regression models (Habeck & Brickman, 2014), so count variables are acceptable as predictors in regressions. Finally, for both ordinal and multinomial regressions, post hoc chi-square tests for homogeneity that employed a Bonferroni-corrected α = .05/3 = .017 were used to determine whether the proportion of children in the different CAS severity groups was significantly different depending on whether a child had a specific comorbid condition or not. All analyses were performed using SPSS Version 28 (IBM Corp., 2022).

Results

Descriptive Statistics and Comorbidity Rates

Children's average age at initial assessment was 4;9 (SD = 2;9; range: 1;0–18;0). Overall, there were 83 children with mild CAS, 35 children with moderate CAS, and 257 children with severe CAS. Children had a mean of 8.2 (SD = 3.4; range: 0–20) comorbid conditions and a mean of 5.6 (SD = 2.2; range: 0–13) communication-related comorbid conditions. See Table 1 for details.

Of the participants, 95.2% experienced expressive LI. Over three quarters experienced cognitive impairment (78.1%), and almost as many experienced receptive LI (72.5%). Approximately one third (37.3%) experienced nonspeech apraxia (i.e., nonspeech oral apraxia, limb apraxia, or oculomotor apraxia), and a similar proportion experienced ASD (33.6%). Figure 1 displays the percentage of participants with different categories of communication-related comorbid conditions, in order of prevalence, and lists the three additional non-communication-related categories.

Between-Group Differences in Comorbid Conditions

A total of 22.1% of the participants were classified in the mild CAS severity group, 9.3% were in the moderate CAS severity group, and 68.5% were in the severe CAS group. Age did not differ between severity groups, F(2, 372) = 0.512, p = .599. Figures are reported as mean (standard deviation) unless otherwise noted. Mean age was 5;0 (3;0) for the children in the mild group, 4;6 (2;2) for those in the moderate group, and 4;9 (2;9) for those in the severe group. DEMSS scores are available for a subset of participants who comprised the reliability set. Mean DEMSS score for the mild group was 109.4 (37.3), significantly lower than that for the severe group (M = 235.1, SD = 106.7), p = .002. The absolute number of comorbid conditions was significantly different between severity groups, F(2, 372) = 9.761, p < .0005. Mean number of comorbid conditions was 7.1 (3.4) in the mild group, 7.0 (3.1) in the moderate group, and 8.7 (3.3) in the severe group. Post hoc tests showed that the absolute number of comorbid conditions was significantly larger in the severe group than in the mild group (p = .001) and the moderate group (p = .012). These results are illustrated in Figure 2 and confirmed by a Kruskal–Wallis H test, which showed that the median number of comorbidities was significantly larger in the severe group than in the mild and moderate groups, H(2) = 17.613, p < .001.

Figure 2.

Figure 2.

Open in a new tab

Number of comorbid conditions by severity category. Circles represent outlying values (3rd quartile plus 1.5 times the interquartile range). *p = .012 between severe and moderate, **p = .001 between severe and mild.

A one-way ANOVA showed that the mean number of communication-related comorbid conditions was also significantly different between severity groups, F(2, 372) = 14.826, p < .0005. Mean number of communication-related comorbid conditions was 4.7 (2.1) in the mild group, 4.9 (2.2) in the moderate group, and 6.0 (2.1) in the severe group. Again, there were significantly more communication-related comorbid conditions for the severe group (p < .0005 vs. mild, p = .012 vs. moderate), illustrated in Figure 3. Details appear in Table 1. These results were confirmed by a Kruskal–Wallis H test, which showed that the median number of communication-related comorbidities was significantly larger in the severe group than in the mild and moderate groups, H(2) = 27.196, p < .001.

Figure 3.

Figure 3.

Open in a new tab

Number of communication-related comorbid conditions by severity category. Circles represent outlying values (3rd quartile plus 1.5 times the interquartile range). *p = .012 between severe and moderate, **p < .0005 between severe and mild.

Association Between CAS Severity and Number of Comorbid Conditions

Absolute number of comorbid conditions. For the absolute number of comorbid conditions, the assumption of proportional odds was met, indicated by a non-significant full likelihood ratio test, χ2(1) = 1.189, p = .276. The ordinal regression model including the number of comorbid conditions significantly predicted severity group over and above an intercept-only model, as indicated by a significantly higher −2 log likelihood, χ2(1) = 18.654, p < .001. Thus, an increase in the absolute number of comorbid conditions was significantly associated with an increase in the odds of being in the moderate or severe category, with an odds ratio of 1.16 (95% CI [1.08, 1.25]), Wald χ2(1) = 17.179, p < .001. That is, for each additional comorbid condition, the odds of a child being judged with moderate or severe CAS increased by 1.16 times. The absolute number of comorbid conditions accounted for 6% of the variance in CAS severity (Nagelkerke pseudo R 2).

Communication-related comorbid conditions. For the number of communication-related comorbid conditions, there were again proportional odds, χ2(1) = 0.200, p = .655. The ordinal regression model with the number of communication-related comorbid conditions was again a good fit to the observed data, χ2(23) = 24.971, p = .352. An increase in the number of communication-related comorbid conditions was significantly associated with an increase in the odds of being in the moderate or severe category, with an odds ratio of 1.32 (95% CI [1.18, 1.46]), Wald χ2(1) = 25.834, p < .001. The number of communication-related comorbid conditions accounted for 9% of the variance in CAS severity (Nagelkerke pseudo R 2).

Association Between CAS Severity and Specific Comorbid Conditions

CAS + cognitive impairment. A total of 225 children (60%) had CAS + cognitive impairment. Of the 225, 19.1% were in the mild CAS group, 5.3% were in the moderate group, and 75.6% were in the severe group. The assumption of proportional odds was not met, χ2(1) = 7.742, p = .005. Therefore, in this case, a multinomial regression was performed to assess whether cognitive impairment comorbidity was associated with different severity levels of CAS. A −2 log likelihood for the multinomial regression model including cognitive impairment comorbidity as a predictor was significantly lower than that for the intercept-only model, χ2(2) = 17.973, p < .001, indicating that comorbid cognitive impairment was a significant predictor of CAS severity category. Children with CAS + cognitive impairment were 1.76 (= 1/.567) times as likely to have severe CAS than mild CAS and 4.33 (= 1/.231) times as likely to have severe CAS than moderate CAS, compared with children with CAS and no cognitive impairment. Comorbid cognitive impairment accounted for 5.8% of the variance in CAS severity (Nagelkerke pseudo R 2).

A significant chi-square test for homogeneity, χ2(2) = 16.011, p < .0005, indicated a difference in the proportion of children with comorbid cognitive impairment in the three CAS severity groups compared with children without comorbid cognitive impairment. Compared with those with CAS and no cognitive impairment, a significantly smaller proportion of children with CAS + comorbid cognitive impairment had moderate CAS (19.1% vs. 26.7%, p = .001), and a significantly larger proportion had severe CAS (75.6% vs. 58.0%, p < .0005).

CAS + receptive LI. A total of 273 children (73%) had CAS + receptive LI. Of the 273, 16.1% were in the mild severity group, 8.8% were in the moderate group, and 75.1% were in the severe group. The assumption of proportional odds was met, χ2(1) = 1.308, p = .253. A −2 log likelihood for the ordinal regression model including receptive LI comorbidity as a predictor was significantly lower than that for the intercept-only model, χ2(1) = 22.994, p < .001, indicating that comorbid receptive LI did predict CAS severity. The odds of a child with CAS + receptive LI having moderate or severe CAS were 3.125 (= 1/.320) times that of a child with CAS and no receptive LI. Comorbid receptive LI accounted for 7.8% of the variance in CAS severity (Nagelkerke pseudo R 2).

A significant chi-square test for homogeneity, χ2(2) = 23.028, p < .0005, indicated a difference in the proportion of children with comorbid receptive LI in the three CAS severity groups compared with children without comorbid receptive LI. A significantly lower proportion of children with comorbid receptive LI had mild CAS (16.1% vs. 38.2%, p < .0005), and a significantly larger proportion had severe CAS (75.1% vs. 51.0%, p < .0005).

CAS and nonspeech apraxia. A total of 148 children (40%) had CAS + nonspeech apraxia. Of the 148, 12.2% were in the mild group, 7.4% were in the moderate group, and 80.4% were in the severe group. The assumption of proportional odds was met, χ2(1) = 0.034, p = .853. A −2 log likelihood for the ordinal regression model including nonspeech apraxia comorbidity as a predictor was significantly lower than that for the intercept-only model, χ2(1) = 14.044, p < .001, indicating that comorbid nonspeech apraxia did predict CAS severity. The odds of a child with CAS and nonspeech apraxia having moderate or severe CAS were 2.43 (= 1/.411) times the odds of a child with CAS and no nonspeech apraxia. Comorbid nonspeech apraxia accounted for 4.6% of the variance in CAS severity (Nagelkerke pseudo R 2).

A significant chi-square test for homogeneity, χ2(2) = 16.958, p < .0005, indicated a difference in the proportion of children with comorbid nonspeech apraxia in the three CAS severity groups compared with children without comorbid nonspeech apraxia. A significantly lower proportion of children with comorbid nonspeech apraxia had mild CAS (12.2% vs. 28.6%, p < .0005), and a significantly larger proportion had severe CAS (80.4% vs. 60.8%, p < .0005).

CAS and ASD. Finally, a total of 126 children (34%) had CAS + ASD. Of the 126, 9.0% were in the mild group, 9.5% were in the moderate group, and 74.1% were in the severe group. The assumption of proportional odds was met, χ2(1) = 0.609, p = .435. A −2 log likelihood for the ordinal regression model including ASD comorbidity as a predictor was not significantly different from that for the intercept-only model, χ2(1) = 2.108, p = .147, indicating that comorbid ASD did not predict CAS severity. The odds of a child with ASD having moderate or severe CAS were not significantly different from those of a child with CAS and no ASD, and a nonsignificant chi-square test for homogeneity, χ2(2) = 1.056, p = .590, indicated no difference in the proportion of participants with and without comorbid ASD in the three CAS severity groups. Table 2 details regression parameters for all regressions performed. Table 3 summarizes the results of the post hoc chi-square tests for homogeneity.

Table 2.

Regression parameters for models predicting childhood apraxia of speech severity category.

Variable B SE Exp(B) p
Absolute number of comorbid conditions (ordinal regression)
Threshold: Mild severity −0.115 0.295 0.892 .697
Moderate severity 0.387 0.294 1.474 .189
Absolute number of comorbidities 0.149 0.036 1.160 < .001
Number of communication-related comorbid conditions (ordinal regression)
Threshold: Mild 0.180 0.299 1.197 .547
Moderate 0.693 0.301 2.001 .021
Number of communication-related comorbidities 0.274 0.054 1.315 < .001
Cognitive impairment comorbidity (multinomial regression)
Mild severity Intercept −0.791 0.193
Cognitive impairment −0.567 0.257 0.567 .027
Moderate severity Intercept −1.276 0.231
Cognitive impairment −1.467 0.387 0.231 < .001
Receptive language impairment comorbidity (ordinal regression)
Threshold: Mild severity −1.635 0.156 0.195 .001
Moderate severity −1.126 0.141 0.324 < .001
No receptive language impairment −1.139 0.235 0.320 < .001
Nonspeech apraxia comorbidity (ordinal regression)
Threshold: Mild severity −1.855 0.218 0.156 .001
Moderate severity −1.372 0.207 0.253 < .001
No nonspeech apraxia −0.890 0.246 0.411 < .001
Autism spectrum disorder comorbidity (ordinal regression)
Threshold: Mild severity −1.752 0.385 0.173 < .001
Moderate severity −1.271 0.380 0.281 < .001
No autism spectrum disorder −0.546 0.397 0.579 .169
Open in a new tab

Table 3.

Associations between specific communication-related comorbid conditions and childhood apraxia of speech (CAS) severity.

Comorbidity group a Severity group
Mild Moderate Severe Total N
Cognitive impairment 43 12** 170** 225
No cognitive impairment 40 23 87 150
Receptive language impairment 44** 24 205** 273
No receptive language impairment 39 11 52 102
Nonspeech apraxia 18** 11 119** 148
No nonspeech apraxia 65 24 138 227
ASD 24 12 90 126
No ASD 159 23 67 249
Open in a new tab

Note. ASD = autism spectrum disorder.

a

The number of children with each comorbid condition who are in each severity group is reported here; see text for percentages.

**

p < .001 between children with CAS + listed comorbidity and children with CAS but without that comorbidity.

Discussion

We aimed to describe patterns of comorbidity in a large cohort of children with CAS and to explore the relationship of different comorbid conditions to CAS severity. Several results emerged. First, we found that all but one child in our sample carried at least one additional medical or communication-related comorbid condition. This finding suggests that comorbidity may be the norm rather than a rare occurrence for children with CAS. Both the absolute number of comorbid conditions and the number of communication-related comorbid conditions showed the same pattern in predicting that the more concurrent comorbid conditions a child had, the more severe their CAS was likely to be. The consistency in these findings is reassuring given the challenges with extracting information on comorbid conditions from medical records, discussed more below.

The weak association between the number of comorbid conditions (however they are counted) and CAS severity also deserves attention. As mentioned, we used the Nagelkerke pseudo R 2 for this purpose, but this statistic should be interpreted with caution because it is generally lower than similar statistics for linear regression. Still, the relatively low amounts of variance in severity accounted for by cognitive impairment, receptive LI, nonspeech apraxia, or ASD leave the question open of what does account for CAS severity. Research shows that gene dosage (i.e., copy number variants, or the number of genetic deletions or duplications larger than 1,000 base pairs) is related to severity in cognitive ability, for example (Casanova et al., 2018; Huguet et al., 2021; Li et al., 2018). By contrast, shared molecular pathways between the genes that are responsible for different NDDs may be more closely related to the presence of comorbidities (Parenti et al., 2020). That is, cases where there are high numbers of comorbidities may, in fact, be cases of multiple molecular diagnoses, even if only one has been identified. Here, phenotypic variation is influenced not only by the additive effect of different mutations but also by the physical and biochemical interactions between those mutations, a phenomenon referred to as epistasis (Parenti et al., 2020). It is also important to remember that not all comorbidities were genetic in origin; some were acquired, and there are nongenetic etiologies associated with some comorbidities as well.

Expressive LI was by far the most common comorbid condition, affecting approximately 95% of the cohort. As mentioned in the introduction, there are three potential explanations for the high comorbidity between CAS and expressive LI: (a) developmental cascade, (b) shared underlying deficit, and (c) masking due to speech impairment. Under the developmental cascade hypothesis, CAS limits children's oral expression abilities and deprives them of opportunities to learn and practice spoken language skills, thereby giving rise to a delay or disorder (Iverson, 2018). The shared underlying deficit hypothesis proposes that an underlying impairment in procedural learning gives rise to both CAS and receptive LI (Iuzzini-Seigel, 2021), which together create an expressive language deficit. Finally, the masking hypothesis proposes that low speech intelligibility due to CAS may make expressive language appear to be impaired, but when CAS is treated, expressive language ability may be found to be within normal limits (Robin, 1992). Of course, more than one of these mechanisms may be operating simultaneously in some children.

The idea of developmental cascades also relates to why motor impairment was included as a communication-related comorbidity. Iverson (2018), for example, showed that being late in meeting motor milestones such as sitting independently and walking unsupported is associated with expressive language delays, presumably because achieving these skills allows children to interact more deeply with the people and objects around them. A diagnosis of CAS may have a similar effect on expressive language in limiting children's ability to orally express language. Further research is required to disentangle the effects of CAS and receptive LI on expressive LI.

Our results also show that a significant proportion of children with CAS (60%) have cognitive impairment and that more children with CAS + cognitive impairment were in the severe, rather than mild or moderate, CAS group. Since there is no a priori reason that cognitive impairment would produce the speech features associated with CAS, the association of comorbid cognitive impairment with more severe CAS supports recent research demonstrating that the two conditions have shared genetic and neural liabilities (Hildebrand et al., 2020; Kaspi et al., 2022), and more research is also needed to understand how they are related.

The high proportions of children with CAS with comorbid receptive LI and nonspeech apraxia are also consistent with previous findings (Duchow et al., 2019; Iuzzini-Seigel, 2019; Murray et al., 2019; Smith & Goffman, 2004; Teverovsky et al., 2009; Thoonen et al., 1997). As mentioned, CAS and LI may share an underlying deficit in procedural learning (Iuzzini-Seigel, 2021). Comorbid receptive LI was also associated with more severe CAS, suggesting a common risk for these two conditions. There is also a wealth of evidence that language and motor development are related in typical development (Iuzzini-Seigel, Hogan, Rong, & Green, 2015; Nip et al., 2011; Smith & Goffman, 2004) as well as in ASD (Iverson et al., 2019; Mody et al., 2017). Nonspeech apraxia was also common among our sample of children with CAS and conferred additional risk for severe CAS. The correlated liabilities models of comorbidity offer one possible explanation here. Under this view, an increase in liability for one disorder is associated with an increase in liability for another disorder, either because the risk factors for each disorder are correlated or because one disorder causes the other (Pennington et al., 2005).

Our finding that approximately 34% of the children in this cohort also had ASD is generally consistent with previous research documenting that between 12.5% and 60% of children with ASD also meet criteria for CAS (Tierney et al., 2015; Velleman et al., 2010) and with previous research documenting that between 25% and 50% of minimally verbal children with ASD meet criteria for CAS (Chenausky et al., 2019). Comorbid ASD was not, however, associated with greater risk for severe CAS in our cohort. This is a surprising result, given that fine and gross motor delays are common in children with ASD (Iverson et al., 2019; Mody et al., 2017). One possible explanation for the lack of apparent comorbidity between CAS and ASD may be genetic heterogeneity, in that CAS may not be a significant comorbidity for all children with ASD, but only for the minimally verbal segment of the ASD population. This kind of heterogeneity is consistent with findings of high rates of CAS in minimally verbal children with ASD but not in low-verbal children with ASD (Chenausky et al., 2019), as well as with a low prevalence of CAS in the speech of children with ASD whose speech is intelligible (Shriberg, Paul, et al., 2011). The comorbidity of CAS in minimally verbal children with ASD may arise independently or be mediated by the association between CAS and cognitive impairment that we see in the current cohort.

Limitations and Future Work

This study has several limitations, one of which is the use of a retrospective convenience sample of children who had been seen at Mayo Clinic over a period of years. This fact has several implications. On the one hand, a sample of complex children that is “enriched” for comorbidity and severity is arguably a good one in which to explore how these two constructs relate. On the other hand, findings in such a cohort may not generalize to less severe cohorts. Because we do not yet know the prevalence of CAS in children with complex NDDs, we also do not know how representative this or any other cohort of children with CAS is of the whole population. Determining this will require further research.

Relatedly, because the DEMSS was being developed during the same time that these data were being collected, DEMSS scores were only available for a subset of participants, who formed the reliability sample. Also, there is currently no universally accepted way of assessing severity in CAS. Still, the clinicians making the diagnoses and severity judgments are highly experienced clinical scientists whose reliability with each other has been demonstrated in much past work (e.g., Strand et al., 2013). Furthermore, clinician ratings have been demonstrated to be a valid method of assessing severity (Chenausky et al., 2022), a useful fact given the importance of relating genetic dosage to symptom severity. Future work, especially at large centers like Mayo Clinic, should aim to collect genetic data from children with CAS in an effort to determine its genetic basis and relationship with comorbidities. Genetic studies that include severity assessment help us understand the range of phenotypic variation in behaviorally defined disorders and its relationship to specific genetic differences.

Another challenge is that of determining what qualifies as a comorbidity and how to classify them. Different terms for what is likely the same condition often appeared in the medical record, thus motivating the analysis using categories of communication-related conditions. This fact also has several downstream implications. First, because not all comorbid conditions were validated by actual diagnostic procedures, this likely added noise to the data. However, it is reassuring that the absolute-number and communication-related condition analyses showed the same pattern.

Also, there is currently no existing system of determining what constitutes a communication-related comorbidity. Put another way, we are only beginning to understand how different comorbidities affect expressive language development. The classification system employed in this study is a preliminary attempt at describing types of comorbidities that could plausibly affect expressive language development and is designed to provide initial information and begin a conversation, rather than to provide definitive answers. It is intentionally a simple start, counting each entry in the medical record as a separate, equally weighted item. In reality, of course, comorbidities are interrelated in complex ways, and future scholars will be able to investigate this with more sophisticated models. However, the analysis reveals useful information, raises specific questions that should be addressed in future work, and makes suggestions about best phenotyping practices for children with CAS going forward. In order to provide the best phenotyping information, future work should aim to employ consistent diagnostic terms, categories, and criteria. Careful phenotyping for developmental disorders, including the use of gold-standard diagnostic instruments and harmonization of data, should be performed. All of this information will help us understand CAS etiology better and refine future conceptual and computational models of comorbidity.

Another limitation of our sample is that because it included only children whose families sought services, the sample may be biased toward more severe CAS: Families of children with more problems, or more severe problems, are more likely to seek help. As large as the current sample is, this bias also likely limits generalization of findings to the broader population. Again, however, even in this potentially skewed sample, differences in the likelihood of having more severe CAS were observed for different comorbid conditions, lending some credence to the findings.

A final limitation is that, because participants spanned a wide range of ages, severity levels, cognitive abilities, and language skills, a consistent set of assessments was not used. Of course, it is not possible to use the same set of assessments with children who are very verbal and children who are nonverbal or minimally verbal because instruments that are appropriate for one group will be too easy or too hard for the other group. Still, a data collection protocol could be used in which the instruments administered to the more severely affected children form a subset of those used with more verbal children. This way, participants in different severity strata will have some scores from the same instruments, along with scores from additional assessments, depending on the specific clinical question under investigation. This is an additional argument for careful and consistent phenotyping of children with developmental disorders, especially for the purposes of understanding genetic and neural differences.

In addition to using a stratified or modular data collection protocol, natural history studies of CAS and other NDDs should be performed. Longitudinal information generated by such studies would shed light on development in these children, providing information that is important for understanding how different comorbidities interact over development (Rapin, 1996). Examination of whether different behaviorally defined developmental disorders share the same underlying deficits (as, e.g., LI and CAS may share an underlying deficit in procedural learning) will help us understand how these disorders are similar to and different from each other.

Data Availability Statement

De-identified data files will be made available to interested researchers upon reasonable request to the last author.

Supplementary Material

Supplemental Material S1. Table of comorbid conditions, categories, and rates of occurrence.
Click here for additional data file. (832.7KB, pdf)

Acknowledgments

This work was supported by National Institute on Deafness and Other Communication Disorders Grants R00 DC017490, awarded to Karen V. Chenausky; P50 DC018006, awarded to Helen Tager-Flusberg, supporting Karen V. Chenausky and Jordan R. Green; and K24 DC016312, awarded to Jordan R. Green. We thank the children who participated in the studies leading to this clinical focus article and their families for allowing them to participate. We also thank Edythe Strand for her help assessing many of the children and for her helpful comments on a previous version of this clinical focus article.

Funding Statement

This work was supported by National Institute on Deafness and Other Communication Disorders Grants R00 DC017490, awarded to Karen V. Chenausky; P50 DC018006, awarded to Helen Tager-Flusberg, supporting Karen V. Chenausky and Jordan R. Green; and K24 DC016312, awarded to Jordan R. Green.

References

  1. American Speech-Language-Hearing Association. (2007). Childhood apraxia of speech [Technical report] . https://www.asha.org/policy/TR2007-00278/
  2. Baylis, A. L. , & Shriberg, L. D. (2019). Estimates of the prevalence of speech and motor speech disorders in youth with 22q11.2 deletion syndrome. American Journal of Speech-Language Pathology, 28(1), 53–82. https://doi.org/10.1044/2018_AJSLP-18-0037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bornman, J. , Alant, E. , & Melring, E. (2001). The use of a digital voice output device to facilitate language development in a child with developmental apraxia of speech: A case study. Disability and Rehabilitation, 23(14), 623–634. https://doi.org/10.1080/09638280110036517 [DOI] [PubMed] [Google Scholar]
  4. Casanova, E. L. , Gerstner, Z. , Sharp, J. L. , Casanova, M. F. , & Feltus, F. A. (2018). Widespread genotype-phenotype correlations in intellectual disability. Frontiers in Psychiatry, 9, 535–535. https://doi.org/10.3389/fpsyt.2018.00535 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chenausky, K. V. , Brignell, A. , Morgan, A. , Gagné, D. , Norton, A. , Tager-Flusberg, H. , Schlaug, G. , Shield, A. , & Green, J. (2020). Factor analysis of signs of childhood apraxia of speech. Journal of Communication Disorders, 87, 106033. https://doi.org/10.1016/j.jcomdis.2020.106033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Chenausky, K. V. , Brignell, A. , Morgan, A. , & Tager-Flusberg, H. (2019). Motor speech impairment predicts expressive language in minimally verbal, but not low verbal, individuals with autism spectrum disorder. Autism & Developmental Language Impairments, 4. https://doi.org/10.1177/2396941519856333 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chenausky, K. V. , Gagné, D. , Stipancic, K. L. , Shield, A. , & Green, J. R. (2022). The relationship between single-word speech severity and intelligibility in childhood apraxia of speech. Journal of Speech, Language, and Hearing Research, 65(3), 843–857. https://doi.org/10.1044/2021_JSLHR-21-00213 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Duchow, H. , Lindsay, A. , Roth, K. , Schell, S. , Allen, D. , & Boliek, C. A. (2019). The co-occurrence of possible developmental coordination disorder and suspected childhood apraxia of speech. Canadian Journal of Speech-Language Pathology and Audiology, 43(2), 81–93. [Google Scholar]
  9. Fedorenko, E. , Morgan, A. , Murray, E. , Cardinaux, A. , Mei, C. , Tager-Flusberg, H. , Fisher, S. , & Kanwisher, N. (2016). A highly penetrant form of childhood apraxia of speech due to deletion of 16p11.2. European Journal of Human Genetics, 24(2), 302–306. https://doi.org/10.1038/ejhg.2015.149 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Habeck, C. G. , & Brickman, A. M. (2014). A common misunderstanding in psychology: Do we need normally distributed independent or dependent variables for linear regression to work? BioRXiv. https://doi.org/10.1101/305946 [Google Scholar]
  11. Hildebrand, M. , Jackson, V. , Scerri, T. , Van Reyk, O. , Coleman, M. , Brade, R. , Turner, S. , Rigbye, K. , Boys, A. , Barton, S. , Webster, R. , Fahey, M. , Saunders, K. , Parry-Fielder, B. , Paxton, G. , Hayman, M. , Coman, D. , Goel, H. , Baxter, A. , … Morgan, A. (2020). Severe childhood speech disorder: Gene discovery highlights transcriptional dysregulation. Neurology, 9(20), e2148–e2167. https://doi.org/10.1212/WNL.0000000000009441 [DOI] [PubMed] [Google Scholar]
  12. Huguet, G. , Schramm, C. , Douard, E. , Tamer, P. , Main, A. , Monin, P. , England, J. , Jizi, K. , Renne, T. , Poirier, M. , Nowak, S. , Martin, C. O. , Younis, N. , Knoth, I. S. , Jean-Louis, M. , Saci, Z. , Auger, M. , Tihy, F. , Mathonnet, G. , … Jacquemont, S. (2021). Genome-wide analysis of gene dosage in 24,092 individuals estimates that 10,000 genes modulate cognitive ability. Molecular Psychology, 26(6), 2663−2676. https://doi.org/10.1038/s41380-020-00985-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. IBM Corp. (2022). IBM SPSS Statistics for Windows (Version 28.0) . [Google Scholar]
  14. Iuzzini-Seigel, J. (2019). Motor performance in children with childhood apraxia of speech and speech sound disorders. Journal of Speech, Language, and Hearing Research, 62(9), 3220–3233. https://doi.org/10.1044/2019_JSLHR-S-18-0380 [DOI] [PubMed] [Google Scholar]
  15. Iuzzini-Seigel, J. (2021). Procedural learning, grammar, and motor skills in children with childhood apraxia of speech, speech sound disorder, and typically developing speech. Journal of Speech, Language, and Hearing Research, 64(4), 1081–1103. https://doi.org/10.1044/2020_JSLHR-20-00581 [DOI] [PubMed] [Google Scholar]
  16. Iuzzini-Seigel, J. , Hogan, T. P. , & Green, J. R. (2017). Speech inconsistency in children with childhood apraxia of speech, language impairment, and speech delay: Depends on the stimuli. Journal of Speech, Language, and Hearing Research, 60(5), 1194–1210. https://doi.org/10.1044/2016_JSLHR-S-15-0184 [DOI] [PubMed] [Google Scholar]
  17. Iuzzini-Seigel, J. , Hogan, T. P. , Guarino, A. J. , & Green, J. R. (2015). Reliance on auditory feedback in children with childhood apraxia of speech. Journal of Communication Disorders, 54, 32–42. https://doi.org/10.1016/j.jcomdis.2015.01.002 [DOI] [PubMed] [Google Scholar]
  18. Iuzzini-Seigel, J. , Hogan, T. P. , Rong, P. , & Green, J. R. (2015). Longitudinal development of speech motor control: Motor and linguistic factors. Journal of Motor Learning and Development, 3(1), 53–68. https://doi.org/10.1123/jmld.2014-0054 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Iverson, J. (2018). Early motor and communicative development in infants with an older sibling with autism spectrum disorder. Journal of Speech, Language, and Hearing Research, 61(11), 2673–2684. https://doi.org/10.1044/2018_JSLHR-L-RSAUT-18-0035 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Iverson, J. , Shic, F. , Wall, C. , Chawarska, K. , Curtin, S. , Estes, A. , Gardner, J. , Hutman, T. , Landa, R. , Levin, A. , Libertus, K. , Messinger, D. , Nelson, C. , Ozonoff, S. , Sacrey, S. , Sheperd, K. , Stone, W. , Tager-Flusberg, H. , Wolff, J. J. , … Young, G. (2019). Early motor abilities in infants at heightened versus low risk for ASD: A Baby Siblings Research Consortium (BSRC) study. Journal of Abnormal Psychology, 128(1), 69–80. https://doi.org/10.1037/abn0000390 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kaspi, A. , Hildebrand, M. S. , Jackson, V. E. , Braden, R. , van Reyk, O. , Howell, T. , Debono, S. , Lauretta, M. , Morison, L. , Coleman, M. , Webster, R. , Coman, D. , Goel, H. , Wallis, M. , Dabscheck, G. , Downie, L. , Baker, E. K. , Parry-Fielder, B. , Ballard, K. , … Morgan, A. T. (2022). Genetic aetiologies for childhood speech disorder: Novel pathways co-expressed during brain development. Molecular Psychiatry. Advance online publication. https://doi.org/10.1038/s41380-022-01764-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Khan, L. , & Lewis, N. (2015). Khan-Lewis Phonological Analysis (3rd ed.). Pearson. [Google Scholar]
  23. Lewis, B. A. , Avrich, M. A. A. , Freebairn, M. L. A. , Taylor, H. G. , Iyengar, S. K. , & Stein, C. M. (2011). Subtyping children with speech sound disorders by endophenotypes. Topics in Language Disorders, 31(2), 112–127. https://doi.org/10.1097/TLD.0b013e318217b5dd [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lewis, B. A. , Freebairn, L. A. , Hansen, A. J. , Iyengar, S. K. , & Taylor, H. G. (2004). School-age follow-up of children with childhood apraxia of speech. Language, Speech, and Hearing Services in Schools, 35(2), 122–140. https://doi.org/10.1044/0161-1461(2004/014) [DOI] [PubMed] [Google Scholar]
  25. Li, Y. , Jia, X. , Wu, H. , Xun, G. , Ou, J. , Zhang, Q. , Li, H. , Bai, T. , Hu, Z. , Zou, X. , Xia, K. , & Guo, H. (2018). Genotype and phenotype correlations for SHANK3 de novo mutations in neurodevelopmental disorders. American Journal of Medical Genetics, 176(12), 2668–2676. https://doi.org/10.1002/ajmg.a.40666 [DOI] [PubMed] [Google Scholar]
  26. Maas, E. , Butalla, C. E. , & Farinella, K. A. (2012). Feedback frequency in treatment for childhood apraxia of speech. American Journal of Speech-Language Pathology, 21(3), 239–257. https://doi.org/10.1044/1058-0360(2012/11-0119) [DOI] [PubMed] [Google Scholar]
  27. Mei, C. , Fedorenko, E. , Amor, D. J. , Boys, A. , Hoeflin, C. , Carew, P. , Burgess, T. , Fisher, S. , & Morgan, A. (2018). Deep phenotyping of speech and language skills in individuals with 16p11.2 deletion. European Journal of Human Genetics, 26(5), 676–686. https://doi.org/10.1038/s41431-018-0102-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Mody, M. , Shui, A. M. , Nowinski, L. A. , Golas, S. B. , Ferrone, C. , O'Rourke, J. A. , & McDougle, C. J. (2017). Communication deficits and the motor system: Exploring patterns of associations in autism spectrum disorder (ASD). Journal of Autism and Developmental Disorders, 47(1), 155–162. https://doi.org/10.1007/s10803-016-2934-y [DOI] [PubMed] [Google Scholar]
  29. Morgan, A. , Braden, R. , Wong, M. M. , Colin, E. , Amor, D. , Liégeois, F. , Srivastava, S. , Vogel, A. , Bizaoui, V. , Ranguin, K. , Fisher, S. , & van Bon, B. (2021). Speech and language deficits are central to SETBP1 haploinsufficiency disorder. European Journal of Human Genetics, 29(8), 1216–1225. https://doi.org/10.1038/s41431-021-00894-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Morgan, A. , Fisher, S. E. , Scheffer, I. , & Hildebrand, M. (2017). FOXP2-related speech and language disorders. GeneReviews. https://www.ncbi.nlm.nih.gov/books/NBK368474/ [PubMed] [Google Scholar]
  31. Murray, E. , McCabe, P. , Heard, R. , & Ballard, K. J. (2015). Differential diagnosis of children with suspected childhood apraxia of speech. Journal of Speech, Language, and Hearing Research, 58(1), 43–60. https://doi.org/10.1044/2014_JSLHR-S-12-0358 [DOI] [PubMed] [Google Scholar]
  32. Murray, E. , Thomas, D. , & McKechnie, J. (2019). Comorbid morphological disorder apparent in some children aged 4-5 years with childhood apraxia of speech: Findings from standardised testing. Clinical Linguistics & Phonetics, 33(1–2), 42–59. https://doi.org/10.1080/02699206.2018.1513565 [DOI] [PubMed] [Google Scholar]
  33. Nip, I. S. , Green, J. R. , & Marx, D. B. (2011). The co-emergence of cognition, language, and speech motor control in early development: A longitudinal correlation study. Journal of Communication Disorders, 44(2), 149–160. https://doi.org/10.1016/j.jcomdis.2010.08.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Parenti, I. , Rabaneda, L. G. , Schoen, H. , & Novarino, G. (2020). Neurodevelopmental disorders: From genetics to functional pathways. Trends in Neurosciences, 43(8), 608–621. https://doi.org/10.1016/j.tins.2020.05.004 [DOI] [PubMed] [Google Scholar]
  35. Pennington, B. F. , Willcutt, E. , & Rhee, S. H. (2005). Analyzing comorbidity. In Pennington B., Willcutt E., & Rhee S. (Eds.), Advances in child development and behavior (Vol. 33, pp. 263–304). Academic; Press. [DOI] [PubMed] [Google Scholar]
  36. Rapin, I. (1996). Practitioner review: Developmental language disorders: A clinical update. The Journal of Child Psychology and Psychiatry, 37(6), 643–655. https://doi.org/10.1111/j.1469-7610.1996.tb01456.x [DOI] [PubMed] [Google Scholar]
  37. Robin, D. (1992). Developmental apraxia of speech: Just another motor problem. American Journal of Speech-Language Pathology, 1(3), 19–22. https://doi.org/10.1044/1058-0360.0103.19 [Google Scholar]
  38. Robinson, E. B. , Samocha, K. E. , Kosmicki, J. A. , McGrath, L. , Neale, B. M. , Perlis, R. H. , & Daly, M. J. (2014). Autism spectrum disorder severity reflects the average contribution of de novo and familial influences. Proceedings of the National Academy of Sciences, 111(42), 15161–15165. https://doi.org/10.1073/pnas.1409204111 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Shriberg, L. D. , Aram, D. M. , & Kwiatkowski, J. (1997). Developmental apraxia of speech: I. Descriptive and theoretical perspectives. Journal of Speech, Language, and Hearing Research, 40(2), 273–285. https://doi.org/10.1044/jslhr.4002.273 [DOI] [PubMed] [Google Scholar]
  40. Shriberg, L. D. , Paul, R. , Black, L. M. , & Van Santen, J. P. (2011). The hypothesis of apraxia of speech in children with autism spectrum disorder. Journal of Autism and Developmental Disorders, 41(4), 405–426. https://doi.org/10.1007/s10803-010-1117-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Shriberg, L. D. , Potter, N. L. , & Strand, E. A. (2011). Prevalence and phenotype of childhood apraxia of speech in youth with galactosemia. Journal of Speech, Language, and Hearing Research, 54(2), 487–519. https://doi.org/10.1044/1092-4388(2010/10-0068) [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Smith, A. , & Goffman, L. (2004). Interaction of language and motor factors in speech production. In Maasen B., Kent R., Peters H., van Lieshout P., & Hulstijn W. (Eds.), Speech motor control in normal and disordered speech (pp. 225–252). Oxford University Press. [Google Scholar]
  43. Stein, C. M. , Benchek, P. , Miller, G. , Hall, N. B. , Menon, D. , Freebairn, L. , Tag, J. , Vick, J. , Taylor, H. G. , Lewis, B. A. , & Iyengar, S. K. (2020). Feature-driven classification reveals potential comorbid subtypes within childhood apraxia of speech. BMC Pediatrics, 20(1), 519–530. https://doi.org/10.1186/s12887-020-02421-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Strand, E. (2017). Appraising apraxia: When a speech-sound disorder is severe, how do you know if it's childhood apraxia of speech? The ASHA Leader, 22(3), 50–58. https://doi.org/10.1044/leader.FTR2.22032017.50 [Google Scholar]
  45. Strand, E. A. (2020). Dynamic temporal and tactile cueing: A treatment strategy for childhood apraxia of speech. American Journal of Speech-Language Pathology, 29(1), 30–48. https://doi.org/10.1044/2019_AJSLP-19-0005 [DOI] [PubMed] [Google Scholar]
  46. Strand, E. A. , & McCauley, R. J. (2019). Dynamic Evaluation of Motor Speech Skill (DEMSS) manual. Brookes. [Google Scholar]
  47. Strand, E. A. , McCauley, R. J. , Weigand, S. D. , Stoeckel, R. E. , & Baas, B. S. (2013). A motor speech assessment for children with severe speech disorders: Reliability and validity evidence. Journal of Speech, Language, and Hearing Research, 56(2), 505–520. https://doi.org/10.1044/1092-4388(2012/12-0094) [DOI] [PubMed] [Google Scholar]
  48. Teverovsky, E. G. , Bickel, J. O. , & Feldman, H. M. (2009). Functional characteristics of children diagnosed with childhood apraxia of speech. Disability and Rehabilitation, 31(2), 94–102. https://doi.org/10.1080/09638280701795030 [DOI] [PubMed] [Google Scholar]
  49. Thoonen, G. , Maassen, B. , Gabreëls, F. , Schreuder, R. , & de Swart, B. (1997). Towards a standardised assessment procedure for developmental apraxia of speech. International Journal of Language & Communication Disorders, 32(1), 37–60. https://doi.org/10.3109/13682829709021455 [DOI] [PubMed] [Google Scholar]
  50. Tierney, C. , Mayes, S. , Lohs, S. R. , Black, A. , Gisin, E. , & Veglia, M. (2015). How valid is the checklist for autism spectrum disorder when a child has apraxia of speech? Journal of Developmental & Behavioral Pediatrics, 36(8), 569–574. https://doi.org/10.1097/DBP.0000000000000189 [DOI] [PubMed] [Google Scholar]
  51. Velleman, S. , Andrianopoulos, M. , Boucher, M. , Perkins, J. J. , Averback, K. E. , Currier, A. R. , Marsello, M. , Lippe, C. , & van Emmerik, R. (2010). Motor speech disorders in children with autism. In Paul R. & Flipsen P. (Eds.), Speech sound disorders in children (pp. 141–180). Plural. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material S1. Table of comorbid conditions, categories, and rates of occurrence.
Click here for additional data file. (832.7KB, pdf)

Data Availability Statement

De-identified data files will be made available to interested researchers upon reasonable request to the last author.

Articles from Journal of Speech, Language, and Hearing Research : JSLHR are provided here courtesy of American Speech-Language-Hearing Association

RESOURCES