Phonological disorder and CAS: Characteristics, goals and treatment

Chapter 6
Phonological disorder and CAS: Characteristics, goals and treatment

Chapter 6 addresses a range of practical issues in the dynamic assessment, and differential diagnosis of children with moderate and severe SSD, including childhood apraxia of speech (CAS) or suspected CAS, sometimes abbreviated as sCAS. The chapter also contains suggestions for and discussion of intervention goals, approaches and techniques for these children. It starts with the proposition (Velleman, 2005) that phonological disorder and CAS have at least six inter-related characteristics in common that, when it gets right down to it in everyday practice, we find ourselves treating symptomatically, while still taking the primary diagnosis into account. In this sense, we ‘treat the symptoms and not the label’.

The four contributors to this chapter are Chantelle Highman (A41) on staying in touch with the juried literature, Karen Froud and Reem Khamis-Dakwar (A42) on the neural underpinnings of CAS, and Amy E. Skinder-Meredith (A43) on rating a child’s speech characteristics.

Treat the symptoms, not the label

As a motor speech disorder, CAS is a discrete diagnostic subtype of childhood (paediatric) SSD. There is a conservative consensus view that it is best characterised as a symptom complex rather than as a unitary disorder and that it may affect, to varying degrees, some combination of: non-speech motor behaviours; speech motor behaviours; production of speech sounds and structures (word and syllable shapes); prosody; language; metalinguistic/phonemic awareness; and literacy (ASHA, 2007b). Maassen (2002) reflected on a finding of Shriberg, Aram, and Kwiatkowski (1997c) that, whereas late onset of speech and slow development are usual in CAS, neither a typical phonological developmental pathway nor a characteristic phonological profile for children with sCAS has been found, and that CAS has no phonological characteristics that are uniquely its own.

It is also generally agreed that many ‘CAS characteristics’ can be found in children with other subtypes of SSD, and that CAS and phonological disorder can co-occur in an individual child. Little wonder then, that Velleman and Strand (1994) declared that this, ‘may result in a variety of motor, phonologic, linguistic or neurologic signs or symptoms and in fact inconsistency among symptoms may be expected as typical’.

Box 6.1 Testing for and observations of phonological disorder

Characteristics		Testing for Phonological Disorder
1. Static speech sound system		Test: Relational Analysis. Look for stable PCC and stable percentage of occurrence of processes over time (6–12 weeks should do).
2. Variability without gradual improvement		Test: Inconsistency Assessment: What kind of inconsistency? DEAP Inconsistency Assessment (Dodd et al., 2002) uses single words, also observe CS.
3. Persisting phonological processes		Test: Relational Analysis (study the norms). Look for phonological processes/patterns that ‘should have been’ eliminated.
4. Chronological mismatch		Test: Study inventories for later sounds and absent earlier sounds.
5. Idiosyncratic rules		Test: Relational Analysis. Look for atypical patterns.
6. Restricted use of contrast		Test: Contrastive Assessment; Phoneme Collapses.
DIAGNOSTIC SIGNS	7. Puzzle phenomenon	Test: Look at phonetic mastery relative to phonemic organization.
	8. Unusual errors	Test: Observations of Conversational Speech (CS) if possible. Look for unusual error-types: systematic sound preference – ‘favourite sound’/‘favourite place’/‘favourite manner’ of articulation.
	9. Marking	Test: Look for marking with nasality and marking with vowel length.
	10. Stimulability	Test: Stimulability testing to 2 syllable positions.

Current thinking in academia, and in the clinical field, tends towards a focus on the overlap of symptoms, and the overlap of treatment methodologies, for children with CAS and children with moderate through to severe phonological disorder. A common-sense (to some) symptomatic approach to treatment has emerged. Velleman (2005) wrote: ‘CAS is different from “regular” phonological disorders, but there are still patterns to be found and treated. There is a great deal of overlap in the symptoms of CAS and the symptoms of other phonological disorders, so it is often difficult to decide whether a diagnosis of CAS is appropriate. But, in a sense, that does not matter. Treat the symptoms, not the label. But then, in an area that enjoys its spirited controversies, this idea does not appeal to all!

Six characteristics CAS and phonological disorder may have in common

The characteristics that can be evident in either disorder, or that may be present when CAS and phonological disorder co-occur in the same child, are

consonant (C), vowel (V) and phonotactic inventory constraints (i.e., consonants, vowels and syllable-word shapes are missing from the respective inventories);
omissions of segments and structures: that is, omissions of consonants, vowels and syllable shapes that are already present in the child’s repertoire;
vowel errors including vowel replacements and vowel distortions;
altered suprasegmentals: that is, atypical prosody;
increased errors with utterance length and/or complexity; and,
the use of simple, but not complex, syllable and word shapes.

In Box 6.1, the characteristics and signs of phonological disorder are listed in the left column and suggestions for how to test and what to look for are listed in the right column. Similarly, Box 6.2 shows the characteristics of CAS, how to test, and observations to be made.

Box 6.2 Testing for and observations of Childhood Apraxia of Speech

Characteristics	Testing for childhood apraxia of speech Use DEMSS very young children with low volubility and for children with severe CAS/sCAS.
Receptive–expressive gap	Administer any language test procedure(s) the child can manage. Formal testing may not be possible for some children (too difficult). Assessment may range from a list of the child’s words and/or word approximations, to a Structural Analysis, right up to demanding standardised language batteries.
Delayed syllable/word structure development	Use a process analysis (e.g., DEAP (Dodd et al., 2002), HAPP-3 (Hodson, 2004), The Quick Screener (QS) (Bowen, 1996)) and/or a Conversation Sample. Look for syllable structure processes, phonotactic constraints.
Deviant syllable structures and word structures	Use DEAP, HAPP-3, QS, SPACS (Williams, 2006) and/or a Process Analysis of a Speech Sample. Look for deviant syllable structure errors (especially Initial Consonant Deletion and schwa insertion/addition, and replacing a consonant with a diphthong).
Sequencing difficulties	Use motor speech examination; DEAP, NDP-3 (Williams & Stephens, 2004); or make informal observations of ‘long words’ of >2 syllables (e.g., Debbie’s clinically useful words in Figure 6.1). Families frequently recount ‘big words’ and the word combinations their child often has difficulty sequencing. Look for metathesis (pasketti), word reversals (mat door) and unusual sequencing: muesli pronounced as yoomsli/yoombleese; Open up it.
Word stress errors; syllable stress errors	Look for excessive and equal stress and Weak Syllable Deletion. Make a syllable stress patterns inventory. Consider an informal long words task.
Vowel constraints; vowel deviations	Look for vowel replacements that do not match adult targets. List vowel inventory/constraints and calculate Percentage of Vowels Correct (PVC) if applicable. Do vowels ‘wander’? Use the Quick Vowel Screener (Bowen, 2010) and Speech Characteristics Rating (Table A43.1).
Prosodic differences	Use the Speech Characteristics Rating. Does prosody affect intelligibility (Flipsen, 2006; Klopfenstein, 2009)? Does it affect the child’s presentation/‘image’.
Phonological awareness difficulties – school-aged children	Use PA Tests, e.g., SPAT-R (Neilson, 2003). Note rhyme awareness, syllable awareness, blending. https://shop.acer.edu.au/acer-shop/group/SPT. Look for literacy issues generally (Zaretsky, Velleman & Curro, 2010).

Intervention goals that are common to phonological disorder and CAS

The six characteristics are listed in the left column of Table 6.1. Reading across Row 1, we see that the first characteristic is consonant and vowel inventory constraints, or in other words, missing vowels and consonants. In phonological disorder, this manifests as systemic simplifications (substitution processes), such as stopping, gliding and fronting where one sound replaces another (see Table 2.4 for consumer-friendly descriptions and examples these error-types and more).

Table 6.1 Phonological disorder and childhood apraxia of speech: Characteristics and goals in common

Phonological disorder (PD) and childhood apraxia of speech (CAS)	Phonological disorder	Childhood apraxia of speech	Phonological disorder and CAS
Characteristics in common	Typical errors	Typical errors	Typical goals
(1) Consonant inventory constraints; vowel inventory constraints; and phonotactic inventory constraints.	Simplifications in the form of systemic or substitution processes, e.g., stopping, gliding; simplifications of syllable structures, e.g., FCD, CR, WSD	Simplifications AND increased segmental complexity: e.g., affricates replacing stops; clusters replacing singletons; diphthongs replacing vowels	Consonant inventory expansion. Vowel inventory expansion. Phonotactic inventory expansion.
(2) Omissions of consonants, vowels and syllable shapes that are already in the inventory.	Simplifications in the form of syllable structure processes and phonotactic errors: e.g., ICD, FCD, CR, WSD	Simplification AND increased structural complexity, e.g., epenthesis (schwa insertion) /səked/ for ‘scared’.	Syllable shape inventory expansion. Word shape inventory expansion. Increased accuracy of production of target structures.
(3) Vowel errors.	Vowel errors are less common in children who don’t have CAS	Vowel errors are more common, and more persistent in CAS	More complete vowel repertoire. More accurate vowel production.
(4) Altered suprasegmentals.	Weak syllable deletion	Excessive and equal stress	Production of strong and weak syllables. Differentiation of strong and weak syllables.
(5) More errors with longer and/or more complex utterances, including the so-called ‘SODA’ errors of substitution, omission, distortion and addition.	SODA, process errors, and segmental complexity errors increase as contexts become more difficult, reducing intelligibility.	SODA, process errors, and segmental complexity errors increase as contexts become more difficult. This is even more obvious in children with CAS.	Generalisation of new consonants and vowels, syllable structures, and word structures, to more challenging contexts.
(6) Use of simple, but not complex, syllable shapes and word shapes.	Syllable structure processes: ICD, FCD, CR, WSD and reduplication	Syllable structure processes are more prevalent and persistent, even when phonetic repertoire is apparently adequate.	More complete phonotactic repertoire. More varied use of phonotactic range within syllables and words. Improved accuracy.

In CAS, consonant and vowel inventory constraints manifest as the same sorts of simplification errors that are found in phonological disorder, but in addition to these, errors that involve increased phonetic complexity are found. For example, my client Costa, 5;8, with CAS referred to ‘Henry the green engine’ which has 10 consonants, as [ˈdʒϵndˌɹi vɜ ˈdʒɹʷind ˈϵnˌdʒɹənd] which has 15. Other examples of his ‘complexification’ were:

jump	dʒɹʌːmpə	blackout	ˈvəɹakˌʒout	real	bəɹiʊʷdə
twin	tʃəɹɪnᵈ	kilometer	ˈklɪləˌməməˈlitlə	fishing	ˈtʃɪˌtʃɪŋ
fastest	tʃəˈɹatˌsɪst	Mrs. Oates	ˈmɪtʃˌʌz ˌoutˈʃəz	creepy-crawley	ˈkəlibˌɹiˈɡlɔˌɹid

Costa’s complexity errors included schwa addition and insertion (both occurred in real), affricates replacing stops, for example, top pronounced as [dʒɒbə] or [dzɒptə]; clusters replacing singletons, for example, rabbit pronounced as [bɹæbɪt] or [dɹæbɪt]; and diphthongs replacing vowels, for example, bed pronounced as [baɪd] or [biːəd].

In the right column of Row 1, we see that the typical therapy goals in common are consonant and vowel inventory expansion to give the child ‘more to work with’. The same format is used in the following five rows of Table 6.1. The clinician then has to determine how best to address these goals given the child’s overall presenting picture. This question of ‘how’ is addressed towards the end of this chapter, and the reader who cannot stand the suspense is referred to Table 6.4!

A difficulty associated with this commonality of characteristics is that, when families who do not have a background in SLP/SLT seek out information without professional guidance, suspecting or even convinced that their child has CAS, they will often recognise enough ‘features’ of CAS to be certain that they ‘know’ what their child’s speech problem is. Clinical experience suggests that they often do this without realising that those same symptoms are far more likely to signal phonological disorder, given the low incidence of CAS.

Compounding this problem, when lay people turn to the Internet for elucidation, they find a disproportionately high number of websites dealing with CAS, the less frequent speech sound disorder, and these may contain inaccurate information. Some of the misleading sites present unsubstantiated opinion and supposition as fact to sell products (e.g., dietary supplements and complementary treatments) and services (e.g., online treatment). By contrast, few sites deal explicitly and accurately with phonological disorder, the more common SSD. It may not be obvious to lay people that SSD affects about 7 children in every 100, with approximately 86% of these children having articulation and phonological difficulties (if we combine SD-GEN and SD-OME), or 98% if we combine SD-GEN, SD-OME and SD-DPI, whereas the CAS-affected children (SD-AOS) represent a miniscule <1% of the SSD population (as discussed in Chapter 2, and see Shriberg et al. (2010)).

An exemplary Internet source of trustworthy CAS information for consumers and clinicians is the Apraxia-KIDS website (Gretz, A7). A positive effect of the growth of Apraxia-KIDS and CASANA has been increased accuracy of information about speech development and disorders circulating on the Web, and with it enhanced communication concerning CAS between consumers and professionals. For some clinicians, this has enabled a ‘more equal’ relationship with clients and a sharper appreciation of the effects of communication disorders on affected individuals and their families, perhaps with the added benefit of improving their skills both as counsellors, and of knowing when to refer to a professional counsellor (Bitter, A15; Overby & Bernthal, A16; Stoeckel, A40).

Among the issues and concerns that arise in parent and family counselling, and in counselling ‘older’ children and youth with persisting SSD, are the potential long-term consequences of these conditions. Gierut (1998) provides a concise summary of likely repercussions, indicating that some individuals can expect lifetime challenges in terms of their retrieval, manipulation and comprehension of linguistic information; their expressive language capabilities; and their education and work choices. Discussing these ramifications, Gierut, p. S87, takes the view that, ‘research calls for both retrospective and prospective studies of the etiology of phonological disorders and the identification of integrated causal relationships and their outcome on a speaker’s daily life’. Such implications impact the assessment and intervention process, right from the opening moments of the initial consultation or case history interview.

Case history interview

Table 6.2 provides ‘assessment prompt’ notes, developed over many years, that clinicians can use, and potentially modify, for case history-taking when they suspect that a child has CAS. Procedures and observations will vary with the child’s age and stage; there is overlap between the 10 sections, and not every section will be needed for every child. If they feel it may facilitate communication, information sharing and insight building, the clinician may wish to provide parents with the prompt, without the footnotes, prior to, or at the beginning of the case history and assessment process. This is not to suggest that it be used as a questionnaire, but rather as a general indication to parents of the territory to be covered in the process of conducting the initial assessment and working towards a provisional diagnosis or diagnosis.

**Table 6.2** A 10-point CAS Assessment Prompt

Motor speech examination worksheet

In working through the motor speech examination worksheet, displayed in Table 6.3, the tasks chosen and the order or presentation depend on the severity of the particular child’s difficulties and any predictions the clinician makes regarding his or her probable performance. The worksheet is intended to help the investigator confirm or reject CAS as a diagnosis, bearing in mind that any determination of oral apraxia would have been made during the structural–functional examination (Skinder-Meredith, A43).

Table 6.3 Motor Speech Examination Worksheet

This worksheet was designed by Edythe Strand, and is used by permission (Bowen, 2009, pp. 212–213). The stimuli used as examples in this version are from the DEMSS (Strand, A45).
(A) Observations during connected speech Example for a young child or one with very Severe impairment
	Vowels	Consonants	Typical maximum word length	Syllable shapes C, CV, VC	MLU
Conversation
Picture description
Narrative
(B) Observations of elicited utterances Example for a young child or one with very Severe impairment Examine, dynamically, the child’s ability to sequence movement for phonetic sequences in various contexts: (1) Vowels (2) CV VC CVC (3) Monosyllabic, bisyllabic, polysyllabic words (4) Phrases (5) Sentences of increasing length looking at the child’s: Movement accuracy; Vowel production; Consistency; and Prosody, and the level of support required. You don’t have to use the DEMSS stimuli (below); use stimuli that ‘suit’ the child.
	Immediate repetition	Repetition after delay – no cues	Simultaneous production needed	Gestural/tactile cues needed
Isolated vowels
CV me hi
VC up eat
Reduplicated syllables mama booboo
CVC1 mom peep pop
CVC2 mad bed hop
Vowel errors Note the different coarticulatory contexts
*Utterances of increasing length (Note the use of simple* words)**
Bi-syllabic 1 baby puppy
Bi-syllabic 2 bunny happy today canoe
Multi-syllabic banana video
Phrases Make up stimuli to suit the child being tested Me too; Big boy
Sentences Make up stimuli of increasing length to suit the child. Dad. Hi dad. Hi daddy.

As with the assessment prompt (Table 6.2), the procedures in Table 6.3 overlap, and not all will be done with every client. There is no particular order of presentation of these tasks, other than the logical hierarchy that the clinician deems appropriate for the particular individual.

Multi-Syllabic words

James (A50) identified the following 10 words as being the most ‘clinically useful’ or most revealing diagnostically: ambulance, hippopotamus, computer, spaghetti, vegetables, helicopter, animals, caravan, caterpillar and butterfly, in her study of polysyllabic words and words containing consonant clusters. These words are displayed in Figure 6.1.

images — **Figure 6.1** Debbie’s 10 long clinically useful words

Reading and reviewing the literature

In order to adequately inform and support families, and to fulfil the requirements of evidence-based and ethical practice (Powell, A39), we must be au fait with the relevant literature, whether by reading it diligently as a component of a personal learning plan or within a professional learning network (PLN), or by absorbing it in ‘distilled’ form at CPD/CEU events, where it is often presented by individuals pursuing, or holding, doctoral degrees and who have truly immersed themselves in a topic.

Among the first steps in the PhD process (Mewburn, 2013; Petre & Rugg, 2010) are choosing both a topic area and a research question or questions. This requires the identification of a ‘do-able’ (by one person) piece of original research, and then the development of a proposal: which is probably when the real work of the dissertation begins. Completing a focused literature review around the research topic and questions is an important aspect of this endeavour, with the review itself eventually becoming the essence of the introduction to the doctoral dissertation. In response to what is found in the literature, and in response to what emerges from discussion with advisors or supervisors (Deem & Brehony, 2000) and mentors, the potential doctoral candidate may reach a point of wanting to reformulate the topic and/or questions, abandoning some questions and honing and sharpening others. The aim of the review is to frame and shape the research, demonstrate that it can fill a crucial gap, and link it to the larger body of knowledge (Mullins & Kiley, 2002). Ultimately, according to Sternberg (1981), the literature review tells readers that the candidate has grasped the subject, has connected the topic to larger historic and current themes, can demonstrate that the proposed contribution is unique, and that the candidate can produce and critically evaluate an astutely refined and focused bibliography.

Dr. Chantelle Highman is an Australian SLP working as a clinician and researcher in Perth, Western Australia. She completed her doctoral degree in School of Psychology and Speech Pathology at Curtin University, investigating potential early speech motor and language precursors in infants at risk of CAS, and looking for evidence for a motor-specific core deficit for the disorder (Maassen, 2002). Her methodology across three studies has involved a combination of retrospective reports by parents of children with CAS, case study analyses of retrospective CAS data and a prospective longitudinal study of siblings of children with CAS, providing an insight into the potential earliest features of disordered speech motor control (Highman, Hennessey, Leitão & Piek, 2013; Highman, Hennessey, Sherwood & Leitão, 2008; Highman, Leitão, Hennessy & Piek, 2012). One driver of this research is a question often asked by parents of children with CAS regarding the ‘early warning features’ they should be watchful for in their younger children, an obvious one that most families probably think we would be able to answer quite easily.

By some quirky synchronicity, Dr. Highman completed her literature review just as the final draft of the ASHA Draft Technical Report on CAS was circulated for comment. In her response to Q41, she discusses the role and process of reading the literature as it relates to practice, and the therapy and research implications of the ratified report (ASHA, 2007a) and position statement (ASHA, 2007b).

Q41. Chantelle Highman: Staying current with published research

The ASHA (2007a) Technical Report on CAS is wide ranging and comprehensive, prompting more questions than it answers, and some of those questions will likely be answered by SLPs/SLTs undertaking higher degrees. The report also makes for absorbing reading for those with the interest and inclination. A recurring theme in discussion with clinical SLPs/SLTs is that, although they would like to, they rarely read the literature for a range of reasons that include lack of time and limited access to journals and other publications (like professional association reports and position statements). On the other hand, for SLPs/SLTs with a theoretical or research bent, reading the literature is routine, time is set aside, and new articles are eagerly awaited and consumed. Can you suggest strategies that clinicians might employ to make the task of keeping abreast with the published evidence base less challenging and more enjoyable; suggest what readers of the Technical Report might gain from it in a practical, clinical sense; and indicate the key areas for further CAS and other child speech research that it points to?

A41. Chantelle Highman: Keeping up to date with the literature

The literature corroborates the many practical issues faced by clinicians when they attempt to implement Evidence-based practice (EBP). These include lack of time, lack of ready access to published literature and lack of skill in evaluating the evidence (e.g., Bernstein Ratner, 2006; Gosling & Westbrook, 2004; Jansen, Rasekaba, Presnell & Holland, 2012; Johnson, 2006; Roddam & Skeat, 2010; Rose & Baldac, 2004; Vallino-Napoli & Reilly, 2004). Daunting though these barriers may appear to be, clinicians know from their professional associations’ Codes of Ethics that keeping up to date with relevant literature is integral to SLPs’/SLTs’ responsibilities. Indeed, when surveyed, most clinicians report placing a high value on the importance of research, and desire to keep up to date with the evidence-base underpinning practice (Stephens & Upton, 2012; Vallino-Napoli & Reilly, 2004). These are worthy aspirations, because in combining our clinical knowledge with new evidence and theories, we may be better positioned to provide more effective, efficient and appropriate services (Reilly, Douglas & Oates, 2004).

Keeping up with the literature

Drawing on the suggestions provided by Johnson (2006) and Reilly et al. (2004), and from my own experience as a clinician and researcher, practical strategies to make keeping up with the literature less challenging and more enjoyable for the clinician include:

Identifying a particular area of interest or a current focus, such as diagnosis of CAS, or treatment approaches for phonological disorders. Allowing your focus to be directed by clinical cases and questions can be particularly motivating.

Linking up with other clinicians and researchers. Tackling one focussed area as a ‘team’ will ensure practical implications can be maximised. This may include joining, or forming a relevant interest group or journal club, or extend to partnering with students and universities for joint benefits.

Making use of the information available on reputable websites (e.g., Bowen, 1998; Childhood Apraxia of Speech Association of North America, CASANA site (www.apraxia-kids.org), see Gretz, A7). Joining the associated discussion groups to receive posts, which often include relevant abstracts, is a time-efficient way of staying connected with clinical and research issues. Making use of the archives and files areas allows clinicians to search for clinically relevant topics, articles and associated discussions with ease.
The advent of online social networking, in particular Twitter and selected blogs, can make ‘following’ a line of research (or researcher) easier.
Investigating your access to online journals and setting up automatic alerts for particular topics and journals’ table of contents will allow you to stay informed without having to repeatedly search for new information.
Checking association websites (e.g., the Mutual Recognition of Professional Association Credentials signatories: ASHA, IASLT, NZSTA, RCSLT, SAC-AOC and Speech Pathology Australia) for information regarding EBP and continuing professional development, as well as for access to technical reports, position statements and clinical guidelines. For example, the ASHA (2007a) CAS Technical Report and the Royal College of Speech and Language Therapists’ policy statement on Developmental Verbal Dyspraxia (DVD) (RCSLT, 2011) provide syntheses of the relevant literature relating to CAS/DVD. SLPs/SLTs who reside outside of the United States can join ASHA as international affiliates and access many resources including all editions of the ASHA journals, and enquiries should go to joinasha@asha.org.
Utilising available published conference proceedings on topics of interest (e.g., Shriberg & Campbell, 2002). Some of these contain not only the presentations, but also transcripts of frank discussions between the top researchers in the field, affording a more in-depth understanding of topic areas. Videos of topic and panel presentations from the 2013 CAS Research Symposium (www.apraxia-kids.org/symposium-videos) are readily available; ensuring that keeping up to date is only a mouse-click away.
Making use of sources that have already synthesised information for you, such as review articles, book chapters, meta-analyses and databases of interventions (e.g., speechBITE, www.speechBITE.com). Johnson (2006) provided a list of speech and language-relevant sources, as well as practical examples of how to apply such information to clinical problems.
Attending relevant CPD/CEU events, especially those that present a synthesis of the available literature on a given topic.

Practical, clinical applications of the technical report and position statement

The ASHA CAS Technical Report (2007a) and position statement (2007b), as well as the RCSLT (2011) policy statement on DVD are examples of readily available resources that synthesise the literature. The documents provide clinicians with a review of CAS-related literature, referring often to typical development and other speech sound disorders. As such, they provide the busy clinician with a current perspective on this controversial area. As indicated earlier, many questions remain unanswered. However, a number of directly relevant and clinically applicable recommendations were presented in the reports. These are summarised below in terms of the four questions referred to previously:

Is it (CAS) a recognised clinical disorder?

The ASHA Technical Report and accompanying Position Statement clearly support the existence of CAS as a recognised paediatric speech sound disorder. The literature suggests that CAS is a complex disorder, characterised by particular speech production and prosodic features, warranting clinical services and further research. However, the report also acknowledges the overlap of features with other speech sound disorders, and the present lack of a set of validated, differentially diagnostic criteria for CAS. The committee suggested that the term CAS be applied to the idiopathic form as well as those occurring in the context of neurological and neurobehavioural disorders, and suggests using the qualifier suspected (i.e., suspected Childhood Apraxia of Speech, sCAS) in cases where the diagnosis is provisional. The RCSLT policy statement on DVD (2011) also supported the existence of the symptom complex of CAS (using the terminology DVD; see the policy statement for a discussion). In contrast to the ASHA CAS Technical Report, the RCSLT policy statement recommended the term DVD should only apply to idiopathic cases, where there is not an associated known neurological or neurobehavioural disorder.

What are its core characteristics?

In the committee’s proposed definition of CAS, ‘a core impairment in planning and/or programming [the] spatiotemporal parameters of movement sequences’ is described (ASHA, 2007b, p. 1), highlighting the importance of skilled motor movements underlying speech production. Such impairment results in speech production and prosodic errors. Despite recognising the lack of a validated set of differentially diagnostic criteria that reliably distinguish CAS from other speech sound disorders, the committee did report on features likely to be diagnostic. Based on the reviewed research and consensus opinion, the following three segmental and suprasegmental core characteristics were proposed (ASHA, 2007b, p. 2) and also included in the 2011 RCSLT position statement

inconsistent errors on consonants and vowels in repeated productions of syllables or words,
lengthened and disrupted coarticulatory transitions between sounds and syllables and
inappropriate prosody, especially in the realisation of lexical or phrasal stress.

The ad-hoc CAS committee acknowledged that an affected child’s speech characteristics may change over time, and that the three features ‘are not proposed to be the necessary and sufficient signs of CAS’ (ASHA, 2007b, p. 2). That is as there is no set of validated criteria, one cannot diagnose (or rule out) CAS on the presence (or absence) of these features alone. A number of additional features often observed in CAS are discussed in detail within the report. These include those features that often comprise clinical diagnostic checklists for CAS (e.g., vowel errors, increased errors as syllable and word complexity increase) and those also observed in children with other speech sound disorders (Forrest, 2003; McCabe, Rosenthal & McLeod, 1998).

How should it be assessed?

The CAS Technical Report does not incorporate definitive assessment guidelines or a single diagnostic test for CAS, due to the lack of evidence supporting such specific recommendations (Caspari, 2007). However, guidance for assessment, based on theoretical and research findings, was presented. A key recommendation from the report was that SLPs/SLTs (and not other professionals) are responsible for diagnosing CAS. Assessment should involve sampling speech and language skills across a range of task difficulties, and considering detailed case history information.

Of the available research reviewed, maximal performance tasks of multi-syllable production (e.g., diadochokinetic tasks, non-word repetition, and multi-syllabic word production) and observation of prosody (especially lexical stress) were reported to be the most informative. However, until validated protocols are identified, assessing CAS remains a cautious exercise, where the clinician must assess broadly and in detail, and usually over time. This is particularly the case for younger or less verbal children who are not yet able to take part in detailed speech assessments that sample speech production abilities across contexts (Davis & Velleman, 2000, but see Strand, McCauley, Weigand, Stoeckel & Baas, 2013). All areas of speech and language, including receptive language, expressive language, speech sequencing, phonological development (including inventory, independent and relational analyses), oral-motor skills, communicative intent, compensatory strategies, facilitative techniques and pre-literacy skills need to be considered in identifying the strengths and weaknesses of the individual child.

The report documented assessment considerations from researchers with substantial clinical experience (ASHA, 2007a, p. 54), which included the importance of differentiating performance on

functional/automatic versus volitional actions,
single [speech] postures versus sequences of postures,
simple contexts versus more complex or novel contexts,
repetitions of the same stimuli versus repetitions of varying stimuli (e.g., sequential motion rates vs. alternating motion rates), and
tasks for which auditory versus visual versus tactile versus a combination of cues are provided.

As is often the case clinically, a ‘working’ diagnosis may need to be considered, with diagnostic therapy (Davis & Velleman, 2000) further informing us of the nature of the deficit and relative contribution of speech motor planning and phonological abilities to the child’s communication impairment. A recently developed assessment tool, the dynamic evaluation of motor speech skill (DEMSS, Strand et al., 2013; Strand, A45) addresses the challenge of assessing younger or less verbal children by using dynamic assessment at varying utterance levels. The DEMSS profiles overall accuracy, vowel accuracy, prosodic accuracy and consistency across utterance types from simple consonant–vowel to multi-syllabic words and longer utterances and has been shown to have sound psychometric properties (Strand, A45).

How should it be treated?

There were few treatment efficacy studies relating to CAS available to the ASHA CAS committee (reflecting the state of the literature at the time), and none that met the highest levels of scientific rigour (Reilly et al., 2004). However, the report provided guidance on treatment approaches and supporting efficacy data, where available, including discussions on intensity of intervention and funding issues. Descriptive and single subject experimental designs provided preliminary support for the use of augmentative and alternative communication (AAC) devices to improve overall communicative competence in children with severely compromised speech production abilities (e.g., Bornman, Alant & Meiring, 2001, cited in ASHA, 2007a). In addition, research describing intervention approaches that specifically target speech production has provided preliminary efficacy information and direction for practice. In particular, the integral stimulation (Strand & Debertine, 2000), and Dynamic Temporal and Tactile Cueing (DTTC, Strand, Stoeckel & Baas, 2006) approaches that incorporate principles of motor learning (McCabe & Ballard, A47; Strand, A45).

Since 2007, exciting treatment research has begun to emerge, exploring intervention approaches for CAS (McCabe & Ballard, A47; Strand, A45; Williams & Stephens, A46). The RCSLT policy statement on DVD (2011) summarised and grouped some of these approaches, with emerging support for interventions that target

expansion of phonetic inventory
tactile cueing to facilitate sound, syllable and word production
consistency of word production
frequent and repetitive practice
incorporation of phonological awareness (PA)
prosodic aspects of speech production

Subsequently, specific treatment appro- aches have been investigated using progressively stronger levels of rigour. For example, Rapid Syllable Transition Treatment (ReST), which directly targets dysprosody in CAS by using intensive practice of accurate stress patterns in multi-syllable non-words, has shown positive effects in small treatment studies (Ballard, Robin, McCabe & McDonald, 2010; McCabe, Macdonald-DaSilva, van Rees, Arciuli & Ballard, 2010) and is now being investigated via a randomised control study (e.g., Murray, McCabe & Ballard, 2012a,b; and see McCabe & Ballard, A47).

Key areas for further CAS and child speech research

Research on CAS is still in its infancy, with the lack of consensus about core diagnostic features presenting a significant barrier to progress (ASHA, 2007a, 2007b). Despite this, there has been a flurry of research addressing some important questions about CAS (e.g., the core features of CAS via longitudinal case studies, Highman, Hennessey, Leitão & Piek (2013), perceptual aspects of CAS, Froud and Khamis-Dakwar (2012; A42), and treatment protocols, Murray, McCabe and Ballard, (2012; A42)). As with speech sound disorders in general, there is still much scope (and an urgent need) for research addressing areas such as:

Diagnostic criteria
Cross-cultural features and presentations
Genetic and neurological factors
Epidemiological factors (e.g., prevalence)
Treatment efficacy
Long-term outcomes
Potential early markers

Perhaps after implementing strategies to keep up to date with the literature, you will be motivated to make a positive contribution to addressing some of these areas, and in doing so, improve the outlook for our clients.

Neurophysiological investigations

The reader has been introduced to Dr. Karen Froud in the preamble to her essay on non-linearity in the previous chapter (Froud, A38). Her co-author in A42 is Dr. Reem Khamis-Dakwar.

Dr. Khamis-Dakwar holds a PhD in speech-language pathology from Teachers College, Columbia University, and is currently working as assistant professor in the Department of Communication Sciences and Disorders at Adelphi University in Long Island, New York. She is the director of the Neurophysiology in Speech Language Pathology Lab, where she conducts research into the neural correlates of linguistic processing and representation in specific sociolinguistic situations, such as Arabic diglossia, and functional changes related to SLP treatment and language learning. She is an expert in speech-language service provision for culturally and linguistically diverse populations, especially Arabic-speaking communities.

A42. Karen Froud and Reem Khamis-Dakwar: Neural correlates of speech sound perception in childhood apraxia of speech

Our attempt to examine the neural underpinnings of CAS using neurophysiological methods was undertaken against a backdrop of conflicting research showing lack of agreement with regards to the aetiology, symptoms, and most effective treatment of CAS (e.g., Davis et al., 1998; Forrest, 2003; Shriberg, Aram & Kwiatkowski, 1997b). The prevailing view in the United States is that CAS is primarily a motor planning impairment affecting voluntary movement sequences for speech (ASHA, 2007a). This is despite extensive research (documented in detail elsewhere in this volume) providing evidence of CAS as involving impairment or delay of aspects of language-specific phonological representation (PR), from a number of different domains: auditory speech sound perception (Bridgeman & Snowling, 1998; Maassen, Groenen & Crul, 2003); language and literacy development in CAS (Adel Aziz, Shohdi, Osman & Habib, 2010; Crary, 1984; Davis & Velleman, 2000; Lewis, Freebairn, Hansen, Iyengar & Taylor, 2004; Marion, Sussman & Marquardt, 1993; Marquardt, Sussman, Snow & Jacks, 2002); phonetic and perceptual analysis of speech of individuals with CAS (Munson, Bjorum & Windsor, 2003).

Against this background, we found it difficult to ignore the notion that CAS might in fact involve a representational component. The major advantage of working on communication impairments from a neurophysiological perspective is that we can often examine complex processes that underpin surface behaviours, capitalising on the millisecond timing precision of electroencephalography (EEG) to gain information about brain responses as they unfold over time. In considering whether CAS might in fact involve a representational component, we had recourse to literature from several distinct (though related) disciplines: foundational understanding of typical phonological development and representation; the neuroscientific literature on phonological processing, specifically the mismatch negativity (MMN) component; and previous studies that had examined acoustic parameters in acquired apraxia of speech (AOS) in adults. By bringing together insights from these various fields, we were able to develop and test a hypothesis concerning a specific neural signature of phonological processing in CAS. This process, our preliminary findings (Froud & Khamis-Dakwar, 2012), and some implications for our understanding of CAS and other speech sound disorders, are described here.

Speech sound representations in typical language development and in CAS

Typically developing infants are able to respond to native and non-native speech sounds, words and grammatical structures (Kuhl, Conboy, Padden, Nelson & Pruitt, 2005). Through the course of language acquisition, there is a gradual process of ‘tuning’ to the native language that results in an eventual insensitivity to non-native speech sounds, indicating a ‘neural commitment to the acoustic and statistical properties of native language phonetic units’ (Kuhl et al., 2005, p. 238). One perspective on this tuning process is that, as native phonology develops, fine-tuning of the native-language sound system results in under-specification of particular features in the phonological system of the language (Bauman-Waengler, 2004; Bernhardt & Stoel-Gammon, 1994). For example, an articulatory feature that is non-contrastive in a specific language (e.g., aspiration in English stops), gradually ceases to be specified, whereas the same feature value is retained in a linguistic system that treats it as contrastive (e.g., aspiration in Hindi stops). These changes to the available feature values of speech-sound representations are necessary for rapid, automatic and efficient processing (Dinnsen, 1996).

Some studies of acquired apraxia of speech in adults showed that acoustic parameters changed following brain damage. Specifically, coarticulation effects were shown to be disturbed in the speech production of adults with a diagnosis of AOS (Mayer, 1995; Dogil & Mayer, 1998), with the typical ratio of speech transitions (i.e., the time spent transitioning from one speech sound to another) being greatly reduced. Dogil and Mayer suggested that these changes could be related to ‘over-specification’ at the phonological level, leading to difficulties in transferring phonological representations to phonetic (articulatory) gestures. Specifically, coarticulation and feature spreading – which are interactions at phonetic boundaries – become compromised or impossible if individual segments are fully specified at the level of phonological representation.

Using this theoretically motivated approach as a basis, we hypothesised that atypical speech behaviours in CAS could stem from phonemic representations that have not been subjected to the typical developmental process of under-specification. This would result in a representational deficit involving the availability of too many options for articulation and processing. One way to evaluate this hypothesis is to examine the neural mechanisms associated with speech sound perception and processing in CAS.

EEG, ERPs and CAS

Recent years have seen the incorporation of neurophysiological methods into investigations of underlying functional impairments in speech and language disorders. In particular, EEG studies that examine fluctuations in the electrical fields generated by brain activity have proved useful in advancing understanding of the very rapid processes involved in speech and language. EEG is a non-invasive method for measuring electrical potentials generated in the brain, by measuring voltage fluctuations at the scalp via electrodes that are connected to highly sensitive amplification and digital recording equipment. Event-related potentials (ERPs) are derived from continuous recordings of EEG by repeatedly presenting stimuli and averaging together sections of the recorded EEG that are concurrent with stimulus presentation. ERPs therefore represent the electrical potentials that are time-locked to particular cognitive processes. Neurophysiological methods with high temporal precision, such as EEG, can capitalise on the very fine temporal distinctions between different domains of language processing (Osterhout, 2000), and the ERP method has been successfully used to isolate specific neural responses implicated in distinct linguistic processes, including syntax, semantics and phonology (Rugg & Coles, 1995; see Friederici, 2000 for an overview).

There is only one study so far that reports EEG data recorded from children with CAS (Rosenbek & Wertz, 1972), which the authors, using the terminology of the day, called developmental apraxia of speech (DAS). This was a retrospective study that examined continuous EEG recordings, finding focal or diffuse abnormalities in 15 out of 26 children with DAS. The examination of specific neurological features associated with CAS is different from ERP approaches, and cannot answer questions related to specific levels of linguistic processing and representation. However, other ERP studies have revealed the existence of a specific signature of phonological memory, referred to as the MMN component (Näätänen, Kujala & Winkler, 2011). The MMN is a fronto-central ERP component that can be elicited by the presence of a ‘deviant’ sound in a sequence of repetitive auditory stimuli (referred to as an ‘oddball paradigm’– see Dehaene-Lambertz & Gliga, 2004, for a review). Several studies revealed that MMN responses to speech sounds that constitute language-specific phonological contrasts are greater in native speakers of that language than in control participants who do not speak that particular language (Cheour et al., 1998; Dehaene-Lambertz, 1997; Dehaene-Lambertz, Dupoux & Gout, 2000; Näätänenet al., 1997; Pulvermüller et al. 2001). Such evidence suggests that MMN can be used as an index of the activation of language-specific phonological memory traces. The advantage of this method is that children (with or without speech sound disorders) do not need to produce any speech, and indeed do not even need to pay attention to the speech sounds as they are being processed; the MMN happens very fast (around 150 milliseconds post stimulus presentation) and is considered an automatic, pre-attentional brain response that indexes unconscious mapping between an incoming stimulus and a pre-existing phonological representation.

For our experiment, we selected pairs of speech sounds that are phonemically and allophonically distinct in English phonology (/pa/ vs. /ba/, and /pa/ vs. /pʰa/). We recorded continuous EEG from children with and without CAS while they listened to sequences of these speech sounds, and interrogated the recordings offline for evidence of MMN responses. Our prediction was that children without CAS would show MMN responses to the phonemically distinct pairs of speech sounds, but not to the allophonic contrast; whereas children with CAS, in the event that CAS is associated with a representational deficit like that proposed by Dogil and Mayer (1998), should show abnormalities in MMN. Our participants were groups of right-handed children who were monolingual English speakers, five each in two groups: typically developing (TD) children, and children with a CAS diagnosis. There were three males in the comparison group, four males in the CAS group; and the children were between 5;1 and 8;9 years of age, with normal hearing.

Our findings revealed the expected dissociation between allophonic and phonemic contrasts in the TD group, with the phonemic speech sound contrasts being associated with a significant MMN response, and no MMN observed in response to the allophonic contrast – as predicted for a typically underspecified phonological system. However, the children with CAS responded quite differently, showing a typical MMN in response to the allophonic contrast only. In response to the phonemic condition, children with CAS showed a positive ERP response, rather than the negative-going electrical potential that was expected. Interestingly, a positive-going MisMatch component has been reported previously in infants, and has been taken to indicate an immature or underdeveloped response to phonemic speech sound contrasts (Dehaene-Lambertz & Dehaene, 1994; Friederici, 2002; Leppänen, Pihko, Eklund & Lyytinen, 1999; Pihko et al., 1999;Rivera-Gaxiola, Silva-Pereyra & Kuhl, 2005; Weber, Hahne, Friedrich & Friederici, 2004).

Neural underpinnings of speech sound disorders

Although inconclusive about the nature of the underlying impairment at this stage, our preliminary investigation of CAS using neurophysiological methods strongly suggests that there is some representational involvement that is impacting the speech sound production disorder. The direction of causality remains opaque and it may not be possible to fully determine the causal relationships between the different components of a symptom complex like CAS. Nevertheless, we have provided preliminary evidence that both a motor planning component and a phonological component appear to be crucially involved in speech apraxia, and this in turn suggests research directions for the elucidation of causal mechanisms underpinning observable changes in speech sound production in CAS, as well as in other speech sound disorders.

One possible approach to the interpretation of the very limited available evidence on the neural underpinnings of speech sound disorders could be that CAS constitutes part of a spectrum. We propose, pending further investigation, that speech sound disorders as a broad category involve differential degrees of impairment to a complex network that regulates interactions between phonological representations, their phonetic instantiations, and the motor realisations of speech sounds. Research in the next decade will capitalise on the greater accessibility of neuroimaging methodologies and the increasing application of such approaches to speech and language disorders. It is our hope that rigorous, theoretically motivated investigations that exploit the precise temporal and spatial capabilities of brain-imaging techniques will shed light on the underlying mechanisms that characterize and distinguish the disorders from one another. Such work should be considered a prerequisite to the development of finely tuned and efficacious therapeutic protocols that can be used to attenuate the impact of speech sound disorders on social participation and educational attainment in affected children.

Characteristics and general observations of CAS

Adopting the suggestions of Shriberg, Campbell, Karlsson, McSweeney and Nadler (2003), the segmental and suprasegmental characteristics the clinician will look for with CAS as a suspected, provisional or working diagnosis are listed below. Then follows a guide to the general observations an SLP/SLT might make during differential diagnosis. These are arranged under the overlapping section headings of: general characteristics, phonetic characteristics/phonetic error-types, sound sequencing difficulties, timing disturbances, disturbed temporal–spatial relationships of the articulators, contextual changes in articulatory proficiency, phonological awareness, receptive language and expressive language.

Segmental characteristics of CAS

Articulatory struggle and silent posturing
Transpositional substitution errors
Marked inconsistency (especially token-to-token variability)
Sound and syllable deletions
Vowel and/or diphthong errors

Suprasegmental (prosodic) characteristics of CAS

Inconsistent realisation of stress
Inconsistent realisation of temporal constraints on both speech and pause events
Inconsistent oral–nasal gestures underlying the percept of nasopharyngeal resonance

Procedures and observations vary with the child’s age and stage. Note that there is overlap between the 10 sections.
1. Hearing	Audiology report	Hearing history
2. Development	Perinatal history Milestones Motor development	Cognition see point 4 below Psychometric/Paediatric report. Be mindful that that speakers with CAS have speech processing deficits in encoding, memory and transcoding (Shriberg, Lohmeier, Strand & Jakielski, 2012).	Social development Wants to communicate; solitary; aloof; ‘separation issues’; clingy, why? Play: who with? Nature of play? Persistent personality? Reactive personality? Ask about teasing/bullying (Hennessy & Hennessy, 2013, pp. 119–122).
	Feeding Latching issues, fighting the breast, sucking, lactation consultant, other intervention re: feeding (and/or sleep pattern), drinking from cup, chewing, gag reflex, vomiting, reflux, failure to thrive, holding food in the mouth, breast or bottle, diet, mouth as a sensor, mouth stuffing, variety of foods.	Health/Wellbeing Illnesses, ‘always at the doctor’s office’, accidents, injuries, ear infections, seizures, ‘separation’, hospital, operations, tires easily, sleep pattern; always ‘on the go’.	Intelligibility Ask parents: can you put a percentage on it? Who understands? Does another child ‘interpret’? Does the other child make mistakes in interpreting? Does the child’s intelligibility vary? Worse when tired? Worse/better in certain situations? Worse with longer utterances? Uses signs or actions to help get the message across. Aware?
	Babble Quiet baby, no babbling, late babbling; lots of babbling, lots of vocal play (gurgling/ raspberries); undifferentiated babbling (all sounded the same); few or no consonants in babbled utterances; ‘babble’ mainly squeals and grunts (i.e., not true babbling). Imitation Tries hard to imitate all the time; Little attempt to imitate sounds; disinterested in imitating words; refusal to imitate words; upset if asked to imitate words. Does/does not imitate play.	Sounds/Words Can say words that are not used in every day speech. How many intelligible words; how many approximations; first words (when?), low vocabulary for age (parents’ judgement), comparison with other children in the family and/or age-peers; one word for many meanings (‘big’ for all machines/vehicles). Only says words at home. Won’t attempt certain words. Can’t say own name. Plays remarkably quietly. Low volubility at home (parents’ judgement).	Gesture/Grunts Uses gesture instead of words; uses vowels and grunts instead of words. Creative use of gesture. Frustration Frustrated when not understood (or passive, or unhappy), or ‘resigned’ or ‘adjusted’ to not being a talker?
	Lost sounds/words Says a word and it is never heard again, keeps a word for a while then ‘loses’ it; words come and go. Sounds come and go. Groping/Struggle Mouthing words? Silent posturing?	Comprehension^a Ask parents for example of how well the child comprehends spoken language. Is there a receptive–expressive gap (comprehension higher than output would suggest)?	Theories? History? Ask parents what they think the problem might be (or might be ‘called’). Have you wondered about a ‘label’, searched the web and joined a discussion group, received advice or ‘suggestions’ from family, friends and others? Thought about family history? What brought you here?
3. Language Ask parents; formal tests
4. Cognition Ask parents; formal tests
5. PA/Literacy Ask parents; formal tests. Story time? Consider offering ONE piece of advice.^b
6. Neuromuscular examination	Posture (sitting/W-sitting?) Muscle Tone Coordination Reflexes	Gait	Sensory function Involuntary movements (if yes, query dysarthria) Physiotherapist or OT reports?
7. Motor speech examination (Table 6.3)
8. Speech and non-speech characteristics – note the general agreement between ASHA (2007a,b) and Davis, Jakielski and Marquardt (1998) Non-speech characteristics speech (Davis et al., 1998; see also Teverovsky, Bickel & Feldman, 2009 for parents’ perception of the functional characteristics of children with CAS within the ICF-CY framework)
(1) Impaired volitional oral movements (2) Reduced expressive compared to receptive language skills (receptive–expressive gap) (3) Reduced diadochokinetic rates Speech characteristics (Davis et al., 1998)
Limited consonant repertoire Limited vowel repertoire Frequent omissions High incidence of vowel errors Inconsistent articulation errors (Token-to-Token) Altered suprasegmentals (prosody) Increased errors with output length/complexity Difficulty in imitation (groping or refusal) Use of simple syllable shapes
9. Speech assessment Standardised Articulation and Phonology Test (e.g., DEAP, HAPP-3) Independent and Relational Analysis Inconsistency Assessment Compare Single Word and Conversational Speech PCC and PVC Intelligibility Ratings Use CS sample for MLUm and Structural Analysis if formal language testing is not possible Look for silent posturing/groping Is the prosodic contour of utterances/sentences intact on imitation? Contrastive stress (I WANT one/I want ONE/I want one) Rule a dysarthria component out or in if possible (Skinder-Meredith, A43) Rule a phonological component out or in if possible
10. Speech characteristics rating (Table A43.1)

Segmental characteristics of CAS

Obgyn Key

Fastest Obstetric, Gynecology and Pediatric Insight Engine

Phonological disorder and CAS: Characteristics, goals and treatment

Treat the symptoms, not the label

Six characteristics CAS and phonological disorder may have in common

Intervention goals that are common to phonological disorder and CAS

Case history interview

Motor speech examination worksheet

Multi-Syllabic words

Reading and reviewing the literature

Neurophysiological investigations

Characteristics and general observations of CAS

Suprasegmental (prosodic) characteristics of CAS

General observations of CAS

Related

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree

Fastest Obstetric, Gynecology and Pediatric Insight Engine

Phonological disorder and CAS: Characteristics, goals and treatment

Share this:

Related

Related posts:

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree