Barriers and facilitators to the assessment of cognitive-communication disorders in children and adolescents after traumatic brain injury: a survey of Australian clinical practice

CCDs in paediatric populations and ‘silent’ sub-populations

CCDs are defined by the American Speech-Language Association (ASHA) as being, ‘disorders encompass(ing) difficulty with any aspect of communication that is affected by disruption of cognition’ (ASHA 2005, p. 2). CCDs can impact any mode of a child or adolescents’ reading, writing, listening or speaking (Chapman et al. 2016). Onset may be delayed (Ylvisaker 1993; Chapman et al. 2016), and outcomes can be measured months or even years after initial injury, regardless of whether a child sustains a moderate to severe (e.g. Mateer and Williams 1991; Eslinger et al. 1992; Ylvisaker 1993) or mild (concussive) injury (e.g. McKinlay et al. 2002; Bernard et al. 2017; Stockbridge and Newman 2019; Tuerk et al. 2020). Importantly, not all Australian children experience the same risk of brain injury. Emerging evidence highlights the existence of ‘silent’ paediatric TBI sub-populations (Maas et al. 2017), which experience increased rates of brain injury and/or poorer long-term outcomes compared to the general population, including Indigenous children/adolescents (Esterman et al. 2018; Peterson et al. 2019), those with pre-existing developmental disabilities/disorders (McKinlay et al. 2012; Kang et al. 2013; Martin et al. 2022), individuals with mental health disorders (Fann et al. 2002), juvenile offenders (Linden et al. 2020), those who have experienced abuse (Cusimano et al. 2021), non-white minority youth (Haider et al. 2007; Jimenez et al. 2013) and youth athletes (Kimbler et al. 2011). Problematically, within Australia such sub-populations may also experience barriers in accessing healthcare (Gilbert and Partridge 2012; Elkington et al. 2019; Department of Social Services 2021; Green et al. 2021; Nolan-Isles et al. 2021). Thus, a substantial proportion of children and adolescents that live with undiagnosed brain injuries (and secondary communicative, behavioural and psychosocial impairments) may be from these sub-populations. Whether a child or adolescent has received an official diagnosis of TBI or not, assessment of their cognitive-communication abilities is important for diagnosis, prognostication, treatment delivery and coordination of services.

Assessing CCDs in clinical practice

Assessment happens in clinical practice for a variety of purposes, for example to diagnose, set goals, evaluate treatment outcomes and prognosticate (Laver-Fawcett and Cox 2021). Assessment refers to the process of collecting data that can be interpreted and used to inform clinical decision making (Intercollegiate Stroke Working Party 2012). Despite assessment being foundational to health service delivery, it is considered to be a complex clinical skill due to the requirement of ‘creativity, experience, knowledge, and original thought’ (Laver-Fawcett and Cox 2021, p. 30). However, assessment may also be complex in part due to the careful consideration that clinicians must give to their measurement practices (i.e. the selection of constructs to measure, the timing of measurement post-injury, and determining how measurement will occur through the selection of measurement instruments and potentially interdisciplinary practice). Resources such as those produced by the Allied Health Professions (AHP) Outcome Measures UK Working Group (2019) outline key considerations for clinicians to ensure the validity and reliability of their measurement practices. Clinical guidelines exist to support the assessment and treatment of speech, language and swallowing outcomes for children and adolescents with TBI during their first year of recovery (Mei et al. 2018). These guidelines, although comprehensive in scope, do not provide in-depth guidance with regards to clinical assessment practices for CCDs. This is not surprising given that significant knowledge gaps still exist for paediatric CCDs, for example in understanding the most important constructs to measure for diagnosis or to determine treatment success, and the optimal timing of measurement (particularly in the chronic stage of recovery). Indeed, the need for ongoing research in this area, particularly for guidelines that support cognitive impairments underlying communication disorders, is recognised by Mei et al. (2018). Decisions about what to measure, when and using which measurement instruments are critical, having implications for accurate and timely identification of paediatric CCDs, appropriate intervention, and optimal health and life outcomes. Given the individual and societal imperative for high quality assessment following TBI, particularly for children and adolescents in ‘silent’ populations, there is a need to understand how measurement currently occurs in clinical practice and the factors that encourage or impede optimal practice.

Theoretical Domains Framework

The Theoretical Domains Framework (TDF) is a framework that may be used to understand the factors influencing health professional behaviours in clinical practice (Atkins et al. 2017). In its refined form, the TDF integrates 14 domains (Michie et al. (2005): (1) knowledge; (2) skills; (3) social/professional role and identity; (4) beliefs about capabilities; (5) optimism; (6) beliefs about consequences; (7) reinforcement; (8) intentions; (9) goals; (10) memory, attention and decision processes; (11) environmental context and resources; (12) social influences, (13) emotions; and (14) behavioural regulation (Cane et al. 2012). Huijg et al. (2014) developed a reliable and valid questionnaire to evaluate the determinants of health care professionals’ behaviour based upon the TDF domains. The TDF has precedented use in communication research (Arnold et al. 2020; Wallace et al. 2021; Kwok et al. 2022) and the implementation of evidence-based guidelines in TBI management (Tavender et al. 2015; Silverberg et al. 2021). An additional benefit of the TDF is that it not only enables the identification of factors influencing implementation, but it can be used to inform future behavioural change interventions (Cane et al. 2012). In this study, the TDF has been used to identify the positive and negative factors influencing Australian SLPs’ measurement practices when assessing CCDs in children and adolescents with TBI.

Therefore, this study aimed to explore Australian SLPs’ measurement practices when assessing CCDs in children and adolescents with TBI. The specific research questions were,

Study design

This study employed a cross-sectional, online survey design and collected qualitative and quantitative data. This design was undertaken not only as a time- and cost-effective approach (Wang and Cheng 2020) but to also enable the convenient participation of a large number of geographically diverse respondents (Evans and Mathur 2005). The web-based survey platform Qualtrics (www.qualtrics.com) was used to develop the survey in accordance with the Checklist for Reporting Results of Internet E-Surveys reporting guidelines (Eysenbach 2004; refer to Supplementary File S1 for an overview of the survey design and reporting).

The survey included 34 questions with Likert rating scales, multiple-choice, open-ended response and yes/no response formats in four sections: (1) participant demographics (six questions) (see Table 1), (2) case history practices (12 questions) (see Figs 1 and 2), (3) measuring CCD outcomes (20 questions) (see Tables 2 and 3), and (4) facilitators and barriers to the measurement of CCD outcomes in clinical practice (32 questions) (see Table 4, Fig. 3). All SLPs were presented with questions in survey sections 1, 2 and 4. However, only SLPs who reported having children and adolescents with TBI on their current caseload were presented with survey section 3. The number of respondents to each question is available upon request from the corresponding author. Question skip logic was employed to create custom paths for each participant.

Fig. 1.

Case history questions asked by Australian SLPs to identify children and adolescents with TBI in general clinical practice.

Fig. 2.

Case history questions asked by Australian SLPs when supporting children and adolescents with documented TBI.

Fig. 3.

Percentage of Australian SLPs providing agreement responses (agree and strongly agree) and less than agreement responses (strongly disagree, disagree, and neutral) to the TDF domains.

Topic areas included in the survey were based on important considerations of measurement outlined by Williamson et al. (2012, 2017); the role of SLPs in the identification, diagnosis and treatment of CCDs described by ASHA (2005); and existing guidelines, models, literature and recommendations describing paediatric CCD measurement (McCauley et al. 2012; Turkstra et al. 2015; MacDonald 2017; Mei et al. 2018). Questions concerning measurement instrument use were also informed by a previous scoping review undertaken by the research team (Crumlish et al. 2022).

Questions in section 4 were developed in alignment with the Determinants of Implementation Behaviour Questionnaire (Huijg et al. 2014). Based upon the TDF (Michie et al. 2005), this questionnaire been used to assess factors that influence SLP behaviours in research (e.g. Wallace et al. 2021) and clinical practice (e.g. Chang et al. 2018; Shrubsole et al. 2019; Arnold et al. 2020). Participants rated their agreement (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, 5 = strongly agree) with 28 TDF positively and negativelyphrased statements. Such statements reflected the 14 validated domains outlined by Cane et al. (2012).

A mixture of positive and negative phrasing was used to control for the possibility of acquiescence response bias (Lavrakas 2008). Positively phrased statements with higher response ratings were considered to represent facilitators, whereas positively phrased statements with lower ratings were considered to reflect barriers. Conversely, higher response ratings to negatively phrased statements were considered to reflect barriers, whereas lower ratings reflected facilitators (Young et al. 2018). Pilot testing of the survey occurred with five experienced SLPs and led to amendments to the wording of some questions to improve overall clarity.

Participants

Australian SLPs were eligible to participate if they were supporting children or adolescents in clinical practice at the time of the survey’s completion. SLPs did not need to be supporting children or adolescents with documented TBI to participate, as (1) most brain injuries are mild and unreported (Leo and McCrea 2016) and (2) the experiences of both specialist (i.e. SLPs with clinical experience) and generalist SLPs (i.e. those without/little TBI experience) were sought in this study. There were no other inclusion or exclusion criteria.

A total of 134 SLPs commenced the survey. However, 23 surveys were excluded, as the participant did not complete past survey section 1 (participant demographics) (n = 23). Therefore, 111 responses were included in the final analysis. The majority of SLPs were located within Queensland (n = 45; 40.5%) and worked within a private practice setting (n = 40; 36%). Most SLPs had either 4–6 years (n = 34; 30.6%) or more than 10 years (n = 34; 30.6%) of clinical experience. Forty-three (38.7%) SLPs had no experience working with children or adolescents with TBI (n = 43; 38.7%). Just over half of the SLPs within this study reported having children and adolescents with TBI in their current caseload (n = 57; 53.3%) (See Table 1). Results are reported as a proportion of the respondents completing each question, as not all participants completed all questions.

Procedures

Convenience and snowball sampling (Gill 2020) were used to recruit Australian SLPs. During a 1-month period (August–September 2020), a link to the online survey was disseminated through Speech Pathology Australia’s National eNews newsletter, social media networks (e.g. Twitter, Facebook, LinkedIn), Australian Special Interest Brain Injury groups and professional networks. SLPs were required to provide informed consent via Qualtrics prior to being presented with survey questions. All data were collected in a de-identified manner.

Data analysis

The survey results were exported into Apple Pages, and close-ended questions were analysed descriptively using frequency distribution and percentages. Qualitative data were analysed using content analysis, with broad concepts and categories being inductively generated (Graneheim and Lundman 2004). Analysis involved (1) reading and clarifying participants’ responses to open-ended questions, (2) dividing responses into meaning units (words/phrases that were related based upon their meaning and context), (3) refining meaning units while still ensuring their original meaning and context, (4) grouping meaning units with others that were most closely related and (5) grouping content codes into subcategories and higher-level categories. Agreement was achieved through author discussion when discrepancies in coding arose.

In survey section 4, facilitators and barriers to assessment approaches were analysed by calculating an average score (between 1 and 5) for each TDF statement and domain based upon published precedence (Young et al. 2018). Based upon work by Arnold et al. (2020) and Young et al. (2018), the scoring system was reversed for negative statements to ensure consistency in the scoring approach. A score of less than three (i.e. <3) reflected a barrier, whereas a score of more than or equal to three (i.e. ≥3) reflected a facilitator. Mean scores for TDF statements were summed and then divided by the number of statements in the domain to calculate the domain score. Facilitators were domains with the highest average score, whereas barriers were domains with the lowest average score.

As outlined by Koch (2006), an audit trail was also maintained by the lead author (LC) throughout the data analysis process (i.e. when extracting raw data, prioritising the most relevant questions for the study’s aims, and completing quantitative and qualitative analysis) to further increase rigour.

Which case history questions do SLPs ask to (i) identify children and adolescents with TBI when injury has not been reported and (ii) inform their measurement practices when supporting children and adolescents with documented TBI?

Just over half of the SLPs (54.0%, n = 54/100) reported that they ‘always’ ask if their clients experienced a traumatic birth when completing a case history. However, more than half of the SLPs reported ‘never’ including questions in their case history about other possible causes of TBI, including car accidents, sporting accidents, family violence, concussions or falls (see Fig. 1).

When undertaking case history analysis with children and adolescents with documented TBI, the majority of Australian SLPs reported ‘always’ asking about a child's or adolescent's age of injury and pre-injury communication abilities respectively (n = 65/71; 91.5%; n = 59/71; 83.1%). Maternal communication abilities (n = 18/71; 25.3%) and socio-economic status (n = 11/71; 15.5%) were ‘always’ asked by a smaller proportion of respondents (see Fig. 2).

When SLPs assess CCDs in children and adolescents with documented TBI, (i) what are the clinical purposes of measurement, (ii) which constructs are measured, (ii) when are constructs measured and (iii) how are constructs measured?

Australian SLPs reported undertaking assessments with children and adolescents with TBI primarily for the purposes of informing goal setting (n = 52; 100%) and guiding intervention (n = 50; 96.2%). Other purposes included screening (n = 35; 67.3%), diagnosing (n = 35; 67.3%) and measuring therapy outcomes (n = 35; 67.3%).

Of the clinicians who reported working with children and adolescents with TBI, 73% (n = 38) reported assessing the cognitive-communication abilities of children and adolescents with TBI. Clinicians who were not assessing cognitive-communication abilities in children and adolescents with a documented history of TBI (n = 14; 27.0%), primarily reported assessing some combination of language, pragmatics, speech and/or swallowing outcomes (n = 5; 9.6%) instead (see Table 2).

Australian SLPs who reported assessing the cognitive-communication abilities of children and adolescents with TBI specifically, reported measuring a variety of constructs through their measurement instrument selection. The five most frequently reported measurement instruments (each reported by more than 50% of participants) measured language (see Table 3). Twenty percent of Australian SLPs reported measuring quality of life (QoL) or psychosocial outcomes of children and adolescents with TBI (n = 10/50; 20%) (see Table 2).

Almost 70% (n = 36) of Australian SLPs who reported undertaking assessments with children and adolescents with TBI reported doing so more than 12 months following a child's or adolescent's injury. More than 90% (n = 46) of Australian SLPs reported recommending ongoing monitoring and reviews, primarily when parental or child-based concerns arose (n = 29; 58%), or when clients reached new developmental, transitionary time points (n = 28; 56%) (see Table 2).

Australian SLPs reported mostly working with other health professional in a multi-disciplinary capacity to assess CCDs (n = 18/37; 48.6%). Most Australian SLPs reported working with clinical psychologists (‘sometimes’: n = 21/32; 65.6%), occupational therapists (‘sometimes’: n = 22/37; 59.5%) and neuropsychologists (‘sometimes’: n = 20/37; 54.1%) (see Table 2).

Australian SLPs reported using a total of 52 measurement instruments to assess the cognitive-communication abilities, QoL and psychosocial outcomes of children and adolescents with TBI. The most frequently reported measurement instrument was the Clinical Evaluation of Language Fundamentals – Fifth Edition (Semel et al. 2014) (n = 32; 86.5%). The Paediatric Quality of Life Inventory (Varni et al. 1999) was the most commonly reported measurement instrument for measuring QoL and/or psychosocial impacts (n = 4/10; 40%) (see Table 3).

What are the facilitators and barriers to Australian SLPs’ assessment of CCDs in clinical practice?

A total of 90 Australian SLPs rated all TDF statements, and their responses were included in the final analysis (see Table 4). The overall agreement and disagreement percentages for all Australian SLPs (n = 90) for each TDF domain are shown in Fig. 3. Having ‘optimism’ was identified as a primary/key facilitator for completing cognitive-communication assessments in clinical practice, with more than 90% of SLPs agreeing that undertaking cognitive-communication assessments with children and adolescents with a history of TBI positively impacts upon therapy outcomes (see TDF statement 5.1) and improves families’ understanding of their child’s impairments (see TDF statement 5.2). The domains of (1) ‘social professional role and identity’ (94%), (2) ’beliefs about consequences’ (90.0%) and (3) ’goals’ (74%) were also identified to be positive facilitators. That is, SLPs believed that assessment was (1) part of their role, (2) important for a child or adolescent’s intervention and outcomes and (3) something that they aimed to undertake and to improve their knowledge and skills within.

Eight TDF domains had disagreement levels greater than 50% and so were considered to be barriers to Australian SLPs’ abilities to undertake cognitive-communication assessments in clinical practice (i.e. ‘behavioural regulation’ (80%); ‘emotion’ (78%); ‘reinforcement’ (74%); ‘beliefs about capabilities’ (73%); ‘skills’ (69%); ‘memory, attention and decision processes’ (65%); ‘environmental context and resources’ (58%); and ‘social influences’ (55%)). ‘Behavioural regulation’ statements relating to SLPs having plans for identifying all children and adolescents with TBI (14.1) and for accessing research to remain aware of current recommendations (14.2) had the highest disagreement percentage (80%) and so were considered to the primary barriers to Australian SLPs’ abilities to undertake cognitive-communication assessments in clinical practice (see Fig. 3 and Table 4).

Figures contained in Supplementary Files S3 and S4 show the agreement and disagreement percentages for SLPs working within specific sectors (workplace) and with varying years of clinical experience respectively (see Supplementary Files S3 and S4). Closer examination of agreement and disagreement percentages within specific workplace sectors showed variance in terms of the TDF domains that were considered to be primary barriers (see Supplementary File S2). ‘Optimism’ and ‘social professional role and identity’ were primary facilitators across sectors (see Supplementary File S2), and ‘behavioural regulation’ remained the primary barrier to Australian SLPs in private practice (89.1%), the Department of Education (93.3%) and non-governmental organisations (95%). In comparison, Australian SLPs working in the Child Development Service (n = 3) reported ‘emotion’ to be the primary barrier (100%) to their clinical practice, reflecting perceived difficulty in undertaking assessments and feelings of nervousness. SLPs in (1) health (n = 20), (2) Juvenile Justice Service (n = 2), and (3) ‘other’ (n = 7) sectors reported ‘(i) reinforcement’ from clients (statement 7.1) or from their workplace (statement 7.2); (ii) having sufficient time, support (statement 11.1) and opportunities for professional development (statement 11.2); and (iii) having confidence (statement 4.1) and/or training in assessing CCDs to be the primary barriers to their assessments respectively (67.5%; 100%; 78.6%) (see Supplementary File S2).

‘Social professional role and identity’ continued to be a facilitator to SLPs’ abilities to assess the cognitive-communication abilities of children and adolescents with TBI, regardless of clinicians’ years of clinical experience (see Supplementary File S3). For clinicians with no experience/less than 1 year of clinical experience supporting children and adolescents with TBI (n = 50) and 4–6 years of professional experience (n = 7), ‘behavioural regulation’ continued to be the primary barrier in clinical practice (see Supplementary File S3). For clinicians with 1–3 years of experience supporting children and adolescents with TBI (n = 14), ‘emotion’ (82.1%) was reported to be the primary barrier in clinical practice, whereas for clinicians with 7–10 years of experience (n = 8) or more than 10 years’ experience (n = 11), the primary barriers were reported to be ‘beliefs about capabilities’ (56.3%) and ‘reinforcement’ (72.7%) respectively (see Supplementary File S3).

Clinical challenge 1: identifying all children and adolescents with CCDs arising from TBI in case history interviews

This study aimed to understand the case history practices of SLPs when assessing children and adolescents both with documented TBI and without. More than half of the respondents reported ‘never’ enquiring about the incidence of car accidents, family violence, falls, sporting accidents and/or concussions when undertaking a case history with children and adolescents in general clinical practice (i.e. when a history of TBI had not been reported by families) (see Fig. 1). This finding is of concern, given (1) the significant number of children and adolescents living with undiagnosed brain injury (Cantor et al. 2004; Leo and McCrea 2016) and (2) families may fail to associate later-presenting CCDs with a previous TBI even if formal diagnosis has occurred (Turkstra et al. 2015), potentially resulting in important information not being shared with clinicians. Clinicians who do not routinely enquire about a range of TBI causes during case history interviews may fail to identify the underlying cause of a child's or adolescent's communication difficulties. Hence, it may be important for the profession moving forward to incorporate case history questions that explore a range of TBI causes into their general case history interviews. TBI populations show heterogeneity in their injury causes and general presentations (Covington and Duff 2021), and without first identifying the risk of TBIs, clinicians may struggle to work backwards and consider the likelihood of a secondarily occurring CCD.

The ‘one size fits all’ nature of TBI research (Maas et al. 2017) may cause additional challenges for clinicians attempting to develop their own case history practices in the absence of comprehensive guidance. Although results showed that clinicians intend to identify all children and adolescents with CCDs from TBIs (see Table 4, TDF statement 8.1), many SLPs reported experiencing barriers in having clinical plans that allowed them to actually do so (see Table 4, TDF statement 14.1). The generic nature of TBI research may mean that research fails to translate and be useful in clinical settings where clinical heterogeneity exists. Emerging research demonstrates that ‘silent’ populations have increased risk of TBI compared to the general population (e.g. Esterman et al. 2018; Linden et al. 2020; Cusimano et al. 2021); for example, Aboriginal peoples experience a two-fold increase in risk of brain injury compared to the general Australian population (Fitts et al. 2019); an increased risk for juvenile offenders, children with pre-existing diagnoses and victims of abuse has also been established (e.g. McKinlay et al. 2012; Fitts et al. 2019; Cusimano et al. 2021; Martin et al. 2022). Without guidance, the onus falls on clinicians to be aware of the heightened risks of brain injury for such silent populations and to account for differences in family awareness, injury causes, personal factors, CCD outcomes and environmental factors in their case history practices. Therefore, it is understandable that Australian SLPs across sectors have reported that generating clinical plans is a primary barrier to their case history practices (see Supplementary File S2), a finding that only improved when SLPs had at least 7–10 years of experience (see Supplementary File S3). Ultimately, greater guidance on the far-ranging causes of TBIs and the clinical heterogeneity of paediatric CCD populations is needed so that all children and adolescents with brain injury can be readily identified through case history practices. Greater representation of injury causes, ethnicities, races, genders and psychosocial variables within future guidance would not only enable better clinical identification but would also ensure that research can actually generalise and be implemented in clinical practice where heterogeneity exists, thereby reducing research wastage.

This study also examined the case history practices of SLPs across a range of sectors (see Table 1). Given the barriers that can exist for silent populations accessing healthcare (Gilbert and Partridge 2012; Elkington et al. 2019; Department of Social Services 2021; Green et al. 2021; Nolan-Isles et al. 2021), it may stand to reason that many children and adolescents with TBI access or present to SLP services outside of dedicated brain injury sectors (e.g. youth justice sector, Department of Education). Thus, there is a need for all SLPs, across sectors, to be able to confidently undertake case histories and CCD assessments with children and adolescents with TBI. Yet results from TDF analysis revealed that feelings of difficulty and nervousness (see Table 4, statements 13.1–13.2) and beliefs about capabilities (see Table 4, statements 4.1–4.2) were consistent barriers to SLPs’ assessment practices across sectors (See Supplementary File S2). Such results establish, for the first time, the need for all Australian workplaces to receive guidance on case history and assessment practices for paediatric populations with CCDs from TBI. There may also be need for a range of CCD case history practices to be developed, such that SLPs working in distinct sectors (e.g. Juvenile Justice) can aptly probe not only for possible causes of TBI but environmental and family factors that markedly influence their clients’ recovery and ongoing development after injury (e.g. Yeates et al. 2012; Ryan et al. 2016, 2017). However, for some workplaces, this may be additionally challenging, given the reported barriers in time, resources and reinforcement that were also reported (see Supplementary File S2). The perceived barriers in undertaking and developing plans to support case history interviews that were reported in this study may mean that some CCDs post-TBI are missed in Australian clinical practice. This is of concern, as clinicians must be able to identify the possibility of a child or adolescent presenting with a CCD if they are to make deliberate choices in their measurement practices to achieve accurate and precise assessments.

Clinical challenge 2: measurement inconsistency

Analysis of findings highlighted that SLPs are not using a common approach to CCD assessments, with 52 distinct measurement instruments (see Table 3), a range of constructs (see Table 3) and differences in interdisciplinary practices being reported (see Table 2). Application of the TDF within this study provides the first insight into possible reasons for the lack of consistency, with participants reporting not receiving tertiary training to support their selection of appropriate measurement instruments to use when assessing CCDs (see Table 4, TDF 2.1; mean = 2.60). Although not specifically queried within this study, barriers with tertiary education may also have included limited clinical education and exposure to paediatric TBI populations during undergraduate and/or post-graduate training. Participants also reported perceived barriers in the TDF domain of ‘environmental context and resources’ (i.e. time, support and opportunities for professional development) (see Table 4, TDF statements 11.1–11.2; mean = 2.83, 2.27 respectively) across sectors, paralleling findings by Kwok et al. (2022).

The reduced QoL of children and adolescents with TBI and the impacts of a range of psychosocial factors on recovery have been reported within the research literature (e.g. health-related QoL (Gabbe et al. 2010), friendship quality (Heverly-Fitt et al. 2014) and incarceration (Seagly et al. 2018)). Yet results indicated that the majority of SLPs reported not measuring QoL or psychosocial constructs (n = 40; 80%; see Table 3). SLPs’ tendency to not measure QoL may simply reflect differences that were also reported in how SLPs complete CCD assessments (i.e. in different interdisciplinary arrangements) and in working with different health professionals (see Table 2). Yet if clinicians are prioritising other constructs ahead of QoL, results may reinforce recent movements within the speech pathology profession towards measuring ‘meaningful’ health outcomes (Morris et al. 2015; Wallace et al. 2017; Janik Blaskova and Gibson 2021), particularly due to the finding that clinicians and clients with TBI can have different perspectives on which constructs are most important to measure in clinical settings (Tate 2014).

Ultimately, inconsistencies in measurement practices have implications for individuals, health services and research advancement. Incorrect/missed diagnoses, treatment selection and prognostications from measurement inconsistencies may have real life consequences for individuals and cohorts of individuals (e.g. juvenile offenders). For clinicians, it also may make referring and managing children and adolescents with TBI across sectors or service deliveries challenging, particularly if health services also have different terminologies for CCDs and CCD outcomes (Ciccia et al. 2021). Inconsistencies in the selection of measurement instruments and constructs also results in incomparable client data, which can impact upon continuity of care. Inconsistencies in measurement practices further limit the potential for meta-analysis of clinic-based data to further progress research in this area. Without consistent measurement, clinical data is not comparable, and robust data sets cannot be established to inform upon future clinically relevant research (Maas et al. 2017).

Clinical challenge 3: knowing what to measure

Results indicated that a range of constructs were measured through measurement instrument selection, including language, symbolic play and disability (see Table 3). Inconsistencies in the selection of constructs may reflect SLPs’ difficulties in knowing which constructs are the most important to measure when assessing CCDs. Emerging recommendations and guidance on construct selection have been published (McCauley et al. 2012; Maas et al. 2017; MacDonald 2017; Mei et al. 2018), but by design, they relate to the selection of constructs for research purposes (McCauley et al. 2012) or to research that has not integrated the perspectives of children and adolescents with CCDs (MacDonald 2017; Mei et al. 2018) or is not specific to CCDs (Maas et al. 2017; Mei et al. 2018) (for comprehensive discussion see Crumlish et al. 2022). In the absence of a core outcome set or explicit recommendations, best-practice is unknown, and determining ‘important’ constructs to measure likely reflects clinicians’ understandings of the range of CCD outcomes, childhood and adolescent development, and recovery trajectories, as well as an understanding of the health outcomes that are most meaningful for/to children and adolescent clients and their families.

Interestingly, the top five measurement instruments (each reported by >55% of respondents) all measured the construct of language (see Table 3). Although SLPs’ knowledge of the difference between CCDs and developmental language disorders (DLDs) were reported to be a facilitator to their assessments (see Table 4, TDF statement 1.1), over-reliance on single sentence, oral language measurement instruments may reflect difficulties with delineating DLDs from CCDs in clinical practice and the need to continue bringing attention to the role of discourse and communication assessments for children and adolescents with TBI. Alternatively, it may reflect SLPs measuring language in lieu of communication due to the need to measure ‘something’ in light of a shortage of appropriate/purposefully developed paediatric CCD/TBI measurement instruments (Mei et al. 2018). Regardless of why, the measurement of single sentence language abilities may miss CCDs and lead to missed diagnoses, poor treatment outcomes and incorrect prognostications, ultimately causing children and adolescents with subtle CCDs to be ‘falling through the cracks’ (Ciccia et al. 2021, p. 856). Although it may be important to measure language in the early stages of recovery to determine return to typical baseline (Mei et al. 2018), other outcomes, such as ‘communication’ (Chapman et al. 2016), ‘quality of life’ and ‘discourse’, may be of equal or more importance as children and adolescents enter into the chronic stages of recovery and ongoing development. As highlighted by Mei et al. (2018) and others (Hill et al. 2018), more research is particularly needed to support discourse language measurements for children and adolescents with brain injuries. Language sampling in the form of narrative retell (83.8%; n = 31) and play-based interactions (75.7%; n = 28) were also reported by SLPs (see Table 3), although conversational, expository and persuasive discourses were not. Such discourses, in both spoken and written formats, would have significance for children's and adolescents’ social interactions, as well as their academic success, particularly as they progress through higher schooling years (Heilmann et al. 2020; Lundine 2020).

SLPs must also be cautious of relying upon adult cognitive-communication recommendations when considering which constructs to measure. Although adult cognitive-communication guidelines are comparatively well-established (e.g. Institut national d’excellence en santé et en services sociaux–Ontario Neurotrauma Foundation 2017; Ontario Neurotrauma Foundation 2018; e.g. Togher et al. 2014), marked differences between adult and paediatric brain injuries necessitate different approaches to assessment (Byom et al. 2014). Qualitative analysis of open-ended statements revealed, ‘CCD assessments are more developed in the adult world. I am in an adult Special Interest Group as well and they talk about CCD in a different way from how we perceive it in paediatrics. It requires a lot of thought, experiment even and problem solving to translate this knowledge to children due to the age at which injury occurs and interrupts with where they are on their developmental trajectory…. We have a long way to go in the paediatric area with getting CCD recognised as an entity of TBI’.

Clinical challenge 4: Australian SLPs (regardless of sector and experience) seek training and professional development opportunities

The key perceived barriers to CCD assessments were behavioural regulation, emotion and reinforcement (see Fig. 3 and Table 4). This suggests that, as a profession, SLPs’ feelings of nervousness and difficulty, the development of clinical plans and receiving recognition are the key barriers to their abilities to undertake CCD assessments. However, specific analysis according to SLPs’ years of experience or sector did show variance. Interestingly, even clinicians with 7–10 years or more than 10 years of experience reported the TDF domain of ‘beliefs about capabilities’ to be a key perceived barrier in their CCD assessments (see Supplementary File S3). As TDF statements concerning tertiary education and training were key barriers for the SLP profession (see Table 4, TDF statements 2.1 and 4.2), such results may indicate the need for additional training and clinical experience for students, as well as professional development opportunities for clinicians of all experience levels. The benefits of additional TBI training have also been supported by Duff and Stuck (2015), who predicted that increased training in TBI for American SLPs would improve service delivery to school-aged children with brain injuries.

Strengths, limitations and future directions

To the authors’ knowledge, this is the first study to explore the measurement practices of SLPs undertaking CCD assessments with children and adolescents with TBI and the barriers and facilitators to assessments through the application of a behaviour change framework. This study also took a unique perspective, recruiting SLPs working within dedicated and non-dedicated brain injury services in an attempt to understand the measurement practices and clinical barriers that exist profession-wide. Positively, this study successfully recruited SLPs from every state and territory within Australia, across a range of sectors and with varying years of professional experience. This was considered to be an important recruitment strategy by the authorship team, given that most brain injuries are mild and unreported (Leo and McCrea 2016) and as both specialist and generalist SLPs are likely to be involved within the management of children and adolescents with brain injury (Turkstra et al. 2015; Ciccia et al. 2021), particularly those within silent populations.

A key limitation of this study was its specific focus on Australian clinical practice. However, results may still be of interest to an international readership, given Australia’s multiculturalism, Indigenous populations, rates of sport among children and adolescents, and dedicated roles of SLPs within Juvenile Justice services, all factors that have been the focus of emerging paediatric TBI research internationally (e.g. Lequerica and Krch 2014; Lakhani et al. 2017; Snow 2019; Johnson et al. 2021). However, future studies may aim to understand if the patterns of CCD measurement practices identified within Australia exist more globally. Additionally, as some questions within the survey had lower response rates, data for such items would not have reflected the views and experiences of all participants. Finally, a greater number of participants may also have engaged within this study, had it not been undertaken during the COVID-19 pandemic.

Measurement inconsistency and barriers to clinical practice in this study support the need for more comprehensive guidance for clinicians. However, more primary research is needed to support the development of such guidance. Although many gaps exist, a prudent next step may be to understand the most important outcomes to measure for children and adolescents with CCDs post-TBI. Such information could then be integrated with previous research to continue contributing towards the important strides that were achieved by Mei et al. (2018) and to give specific focus to CCDs. This understanding could also subsequently inform measurement instrument development/validation, timing of measurement and the development of much needed treatment research, allowing for the Lancet Commissions’ recommendations of both measurement standardisation and treatment development (Maas et al. 2017) to be achieved. This would also allow for the advancement of paediatric CCD research to continue evolving in a systematic and efficient manner.