Abstract
The World Health Organization’s International Classification of Functioning, Disability and Health (ICF) aligns with the neurodiversity paradigm in viewing autistic people’s social communication holistically and in a strength-based manner. In this scoping review, we explored how social communication interventions for autistic children and youth map onto the domains of the International Classification of Functioning, Disability and Health in the field of speech-language pathology. OVID Medline, OVID Embase, OVID PsycINFO and Web of Science databases were searched to identify relevant articles. Population, intervention and study data were extracted, as well as data on each of the International Classification of Functioning, Disability and Health domains related to the social communication interventions. In total, 21 articles were included in our analysis. No studies explicitly mentioned the International Classification of Functioning, Disability and Health. All the studies focused on participation and environmental factors (e.g., people’s attitudes, physical environment), and some studies discussed other International Classification of Functioning, Disability and Health domains such as body structures and functions, personal factors and activities. The examples provided for each International Classification of Functioning, Disability and Health domain may be helpful for clinicians and researchers looking to understand how components of social communication interventions link to International Classification of Functioning, Disability and Health categories. Future work could analyse how social communication interventions in other fields (e.g., psychology, occupational therapy) map onto the International Classification of Functioning, Disability and Health.
Lay abstract
This review article investigates how the World Health Organization’s International Classification of Functioning, Disability and Health can be applied to better understand speech-language pathology social communication interventions. In recent years, academic articles have supported thinking differently about how autistic people communicate, including the many strengths autistic people have and how other people and the environment influence communication. The International Classification of Functioning, Disability and Health is a holistic, widely used framework that provides a neurodiversity-affirming perspective on social communication interventions for autistic children. We did not find any published literature applying the International Classification of Functioning, Disability and Health to speech-language pathology social communication interventions for autistic children and youth and therefore wanted to explore whether and how social communication interventions reflect International Classification of Functioning, Disability and Health concepts. To answer this question, we searched the academic literature using several databases using a methodology called a scoping review. We included articles that had autistic children and youth as participants and focused on social communication interventions in the field of speech-language pathology. We found 21 articles that met our inclusion criteria. No studies talked explicitly about the International Classification of Functioning, Disability and Health. All the studies focused on participation and environmental factors (e.g., people’s attitudes, physical environment), and some studies discussed other International Classification of Functioning, Disability and Health domains such as body structures and functions, personal factors and activities. It is important for speech-language pathology social communication interventions to consider how the physical environment, social environment and personal factors impact social communication. Clinicians and researchers may find our application of the International Classification of Functioning, Disability and Health to social communication interventions useful to shape how they think about interventions.
Keywords
Frameworks of disability and health provide a lens through which to view people and societies, shaping service provision and attitudes related to disability. In the past two decades, the World Health Organization’s (WHO) International Classification of Functioning, Disability and Health (ICF) has reframed how disability and health are conceptualized (WHO, 2001). The ICF provides a biopsychosocial framework to consider two major components of health – contextual elements (i.e., personal factors, environmental factors) and functioning and disability elements (i.e., body structures and functions, activities, participation), with bidirectional arrows connecting the domains to indicate their interconnectedness (WHO, 2001). Figure 1 shows a visual representation of the ICF. The five ICF domains provide a holistic perspective on a person’s health by examining their body structures and functions (i.e., physical body parts, physiological and mental functions of a body), activities (i.e., execution of tasks), participation (i.e., involvement in life situations and society), personal factors (i.e., personal experiences including race, gender, age, preferences, culture) and environmental factors (i.e., any factor not inherent to the individual, including five subcategories of (1) products and technology, (2) environment and human-made changes to environment, (3) support and relationships, (4) attitudes and (5) services, systems and policies) (WHO, 2001).
International Classification of Functioning, Disability and Health (ICF) visual.
The ICF and its five domains can be applied generally, or by using specific sub-categories and alphanumeric codes organized into ‘core sets’ (WHO, 2001). ICF core sets include specific selections of ICF codes that were identified as relating to certain diagnoses, including autism (Bölte et al., 2014, 2021; Schiariti et al., 2018). For example, the ICF core set for autism was developed in 2018 with the goal of standardizing assessment of functioning for autistic people in a variety of settings and includes 111 ICF categories (Bölte et al., 2018). The ICF has inspired many tools and resources that can be used to support autistic youth 1 . For example, the Autism Classification System of Functioning: Social Communication (ACSF: SC) is a strength-based descriptive system based on the ICF that asks questions about a child/youth’s social communication skills at their best and on a typical day (Di Rezze & Rosenbaum, 2016). Other examples include the TEA-CIFunciona tool from Argentina that uses a synthesized version of the ICF autism core set to describe autistic children’s functioning (Napoli et al., 2021), and the ‘your ideas about participation and environment’ (YIPE) self-report tool that can be used to set goals or frame conversations between children with disabilities and healthcare providers (Cheeseman et al., 2013).
For autistic people, the ICF can be a helpful framework with which to explore the values and components of interventions, including social communication interventions that are commonly recommended for autistic children and youth. ‘Social communication’ refers to how and why language is used to communicate with others in social environments and is culture- and context-specific (ASHA, n.d.-a). Social communication includes social interaction and understanding, social cognition, verbal and nonverbal social behaviours (e.g., body language, tone of voice) and language processing (ASHA, n.d.-a). Service provision primarily focuses on children and youth, although some social communication interventions exist for autistic adults (Cummins et al., 2020; Fuller & Kaiser, 2020).
Alignment between the ICF and the neurodiversity movement
Autistic people’s social communication differences have historically been pathologized by people using a normative-focused medical model of disability, with many interventions aiming to remediate ‘deficits’ in the autistic person’s social communication skills (American Psychiatric Association, 2013; Kapp, 2019, 2020), including in the field of speech-language pathology (SLP) (Hsieh et al., 2018; Westby & Washington, 2017). In recent years, the neurodiversity movement and frameworks such as Milton’s (2012) double empathy problem have reframed autistic contributions to social communication as different rather than ‘deficient’ (Heasman & Gillespie, 2019). Recent literature has emphasized the impact of personal and environmental factors on autistic people’s social communication and the importance of using strength-based approaches (Binns & Oram Cardy, 2019; Crompton et al., 2020). In addition to a shift in understanding, shifts in practice have been proposed to centre autistic people’s perspectives, consider contextual factors and to move away from deficit-focused assessments and interventions (Albin et al., 2024; Bottema-Beutel, 2024; Chen et al., 2022; Morrison et al., 2020).
The ICF and the neurodiversity movement developed separately, yet both incorporate personal and environmental factors into goal setting and focus on strengths-based service delivery (Bölte et al., 2021). The arrows connecting the ICF domains (Figure 1) align with recent literature outlining the interconnectedness of personal, sociocultural and sociopolitical, relational, interactional, and external factors in influencing a person’s social communication abilities (Black et al., 2022; Bottema-Beutel, 2024; Chen et al., 2022; Tola et al., 2021). For example, environmental factors align with the neurodiversity movement’s description of the ‘contextualization processes that devalue autistic ways of being and that render characteristics associated with autism disabling’ (Bottema-Beutel, 2024, p. 1599). Environmental factors also allow for the inclusion of different societal perspectives on autism and neurodiversity, as opposed to limiting our evidence base to only White, western scholarship (Nair et al., 2024).
The ICF’s personal and environmental factors domains also provide a framework to consider how culture influences perspectives on autism. In many Indigenous communities, perspectives on disability differ from White western deficits-based conceptualizations (AltogetherAutism, 2019; Bruno et al., 2025; Ineese-Nash, 2020; Lilley et al., 2019). In the Māori culture, for example, autism is referred to as ‘Takiwatanga’, translated to ‘in their own time and space’ (AltogetherAutism, 2019). Perspectives on autism in some Indigenous communities may align with a social model of disability, although not labelled as such. For example, Lilley et al. (2019) described Aboriginal and Torres Strait Islander women’s experiences of autism, and that although there were many challenges faced in daily life, most of the difficulties and stresses they spoke about were caused by their interactions with other people, services or institutions, not by their children. (Lilley et al., 2019, p. 44)
In addition to exploring the alignment between the ICF and the neurodiversity movement, it is important to note criticism that the ICF overemphasizes impairment, rather than environmental factors, as the primary cause of disability (Lundälv et al., 2015). There is also criticism that the ICF classifies deviations from societal norms rather than focusing on human needs (Hammell, 2015). Bölte et al. (2021) addresses the perceived ‘irreconcilable’ relationship between the ICF and neurodiversity and outlines the many ways in which they can align and complement each other.
The ICF in SLP literature
In addition to exploring how the ICF links to neurodiversity, it is important to explore whether and how existing SLP interventions already apply the ICF, both generally and for social communication services specifically. There is SLP literature discussing the ICF (Blake & McLeod, 2018; Ma et al., 2008; McLeod, 2009; Scholten et al., 2019; Threats, 2010; Westby & Washington, 2017), including how activities and participation (O’Halloran & Larkins, 2008) and contextual factors (Howe, 2008) can be incorporated in SLP interventions. The ICF has been applied in specific SLP practice areas such as speech (McLeod, 2009) and language disorders (Westby & Washington, 2017). The ICF is discussed explicitly by Speech-Language and Audiology Canada (SAC) in their position statement on outcome measurement (Speech-Language and Audiology Canada, 2010) and on the American Speech-Language-Hearing Association’s (ASHA) website, which includes an ICF section and ICF-focused case studies (ASHA, n.d.-b). The ICF is also incorporated in SLP training programmes, such as at McMaster University’s (n.d.) SLP programme, which incorporates the ICF into all coursework and problem-based learning.
For social communication particularly, there are social communication models in the SLP academic literature that incorporate the ICF (Jethava et al., 2022; Keegan et al., 2023) and literature acknowledging the impact of contextual factors on social communication (Black et al., 2022; Bottema-Beutel, 2024; Chen et al., 2022). Yet, to date, the ICF has not been applied to analyse social communication interventions for autistic children and youth. Exploring how social communication interventions map onto ICF domains aligns with a broader shift in the field of SLP to ensure that interventions are holistic and neurodiversity-affirming (Gaddy & Crow, 2023). Although speech-language pathologists may be aware of the ICF and its components, they may not feel confident incorporating the concepts into clinical practice (Scholten et al., 2019). Applying the ICF to social communication interventions and documenting the associated process may allow clinicians to expand their frameworks and implement a similar process to analyse the social communication interventions they use. A comprehensive review of SLP social communication interventions for autistic children and youth will also provide valuable information to clinicians and researchers on the common components of these interventions using the language of the ICF.
Objective and research questions
The primary objective of this scoping review is to describe the SLP social communication intervention literature for autistic children and youth through the lens of the ICF. The focal research question was:
In asking this question, we were interested in whether and how SLP social communication intervention concepts can be mapped onto the ICF concepts, and not simply whether ICF terminology was used.
Method
A scoping review (Arksey & O’Malley, 2005; Levac et al., 2010) was conducted to synthesize evidence, identify research gaps and disseminate knowledge regarding how social communication interventions in the field of SLP map onto the ICF. Scoping review principles, outlined by Arksey and O’Malley (2005) and refined by Levac and colleagues (2010), were followed to (a) identify research objectives; (b) search the literature for relevant studies; (c) select studies systematically; (d) chart the extracted data and (e) collate, summarize and report the results. This review is reported based on the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) extension for Scoping Reviews (PRISMA-ScR) by Tricco and colleagues (2018). Search strategy and inclusion criteria documents were uploaded to Figshare (2024), an open-access repository, in October 2024 prior to full-text screening and were made public in November 2024 (https://figshare.com/s/6e008854dd4277bf1267?file=49872516 & https://figshare.com/s/8e016c56bbcbe41065f7?file=49872489).
Development of research question and objectives
The first author, a speech-language pathologist and doctoral student (M.A.), met regularly with PR, a paediatrician and scientist, to develop the research question as stated above. This project began as part of an ICF-focused course at McMaster University. A scientist and speech-language pathologist (M.P.) also provided feedback on the proposed research question and manuscript. Preliminary database searches of OVID Medline and Embase, Cochrane reviews and Google Scholar were conducted to ensure that no current reviews existed that answer this research question.
Given the culturally dependent nature of social communication (ASHA, n.d.-a), the research questions and analysis are inherently influenced by the positionality of the authors. In our authorship team, all authors are Canadian, cisgender, neurotypical healthcare providers who have clinical experiences supporting autistic children and youth as speech-language pathologists (M.A. and M.P.) and as a paediatrician (P.R.) but are not autistic themselves. The research question was developed by M.A. in consultation with P.R., related to M.A.’s clinical experiences as a speech-language pathologist providing neurodiversity-affirming social communication assessment and intervention. McMaster University’s ICF-focused course allowed M.A. to explore the ICF’s utility as a clinical framework in the field of SLP and led to curiosity as to how the ICF could be applied to SLP social communication interventions.
Search strategy to identify relevant studies
Database search
Following discussion by M.A. and P.R., a McMaster University health sciences librarian provided guidance and feedback on the research question, databases chosen and the search terms. Electronic searches of OVID Medline, OVID Embase, OVID PsycINFO and Web of Science were conducted in October 2024. Search terms were categorized as follows: children/youth, diagnosed with autism, social communication (e.g., social skills, pragmatic language), and speech-language pathology (e.g., speech-language therapy, SLP). Terms were modified slightly according to each database’s properties. Table 1 shows an example of search terms used. Covidence was used to deduplicate, collate and review articles (‘Covidence systematic review software’, n.d.).
Search terms from OVID Medline database.
Terms in each category were connected with ‘OR’, and each group was connected with ‘AND’.
= truncation, / = subject heading.
Selection of relevant studies
Inclusion and exclusion criteria were developed by the study team and refined. Two reviewers, M.A. and SLP student M.K., piloted and further refined the inclusion/exclusion criteria on 15 articles and then completed minor revisions to inclusion criteria as needed before proceeding with title and abstract screening. Studies were included if: (a) they included participants who were children and youth between the ages of 2 and 21 years; (b) participants were diagnosed with autism or other pre-Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-V) terms (e.g., Aspergers, PDD-NOS); (c) social communication was listed as an aim and/or outcome of the intervention; and (d) there was SLP involvement in delivery, training or creation of the intervention. We focused on SLP literature to capture how speech-language pathologists specifically are approaching social communication intervention. Additional criteria included (e) peer-reviewed studies with full texts available in English and (f) published in or after 2001 – the year that the ICF was published (WHO, 2001).
Charting and extraction of data
A data-charting form was created using Microsoft Excel and was modified following discussions with authors M.A. and P.R. Pollock and colleagues’ (2023) recommendations for data charting and extraction, synthesis, and analysis for scoping reviews were followed. The data extracted included direct quotations with a corresponding page number or paraphrasing longer sections of text along with a corresponding page number. Data extraction was guided by the Population, Concept, Context (PCC) framework (Peters et al., 2020). For population data, we extracted the ages, diagnoses, and the number of children/youth in the studies. Concept data included details on the studies/interventions, such as the study design, if any of the social communication terms used in the article were defined, and who delivered the intervention (e.g., only speech-language pathologists, other professionals as well).
Context data explored how the intervention mapped onto the ICF. We extracted if the ICF was explicitly mentioned anywhere in the study, as well as which of the five ICF domains mapped onto the intervention purpose. We then analysed intervention content using the ICF by extracting and coding whether participation, activities (not otherwise captured by participation domain), body structures and functions (not otherwise captured by activities or participation), environmental factors, and/or personal factors were reported in intervention aims (i.e., what intervention strives to achieve, which can include purpose or goals) and/or outcomes (i.e., outcome measures used, outcomes authors report based on their intervention).
Any data relevant to each ICF domain were extracted. Coders read and re-read each article and assessed which ICF domains were captured by extracting quotes and assigning a ‘yes’ or ‘no’ rating to represent the presence or absence of each ICF domain. The ICF was applied broadly, using the definitions of each ICF domain from the ICF manual to guide data extraction and coding (WHO, 2001). Coders extracted data from each article by evaluating which components of the social communication intervention related to each ICF domain. The study did not need to use ICF terminology to fit a specific domain; rather, the ‘gestalt’ of the definitions was applied to capture which domains were present and examples thereof. For example, ‘activity’ is the execution of tasks or actions by an individual, while ‘participation’ involves activities completed in the context of a life situation (WHO, 2001). Intervention content was coded as ‘activity’ when there was no mention of participation in a life situation (e.g., listening to facts about turn-taking), but coded as ‘participation’ if the context of a life situation was present (e.g., applying turn-taking information to a conversation with peers). For environmental factors, data were organized using the five ICF categories of environmental factors (WHO, 2001): (1) products and technology, (2) natural environment and human-made changes to the environment, (3) service systems and policies, (4) support and relationships and (5) attitudes (WHO, 2001). The data extraction table also included a section for reflection and additional thoughts not captured elsewhere on the extraction form. The only area of data extraction where ICF terminology was required was the question of whether the ICF was explicitly mentioned in the study.
Data were extracted by two people. The first author (M.A.) extracted all the data on the Excel spreadsheet. An SLP student (M.E.) acted as a second coder and extracted data using the same document, marking any additions or disagreements in a different colour to track the percent agreement between the two reviewers. After both authors extracted data, they reviewed all data together and discussed any disagreements. Both authors agreed on over 98% of the data extracted, with only 10 of the 609 cells of data extracted having disagreements. All disagreements were resolved with discussion between the two reviewers. A third reviewer (M.P.) was available to adjudicate, but this was not required. Supplementary Materials, posted to Figshare (https://figshare.com/s/aa0d3bffe0d9dc8da010), show the raw extracted data spreadsheet.
Synthesis of extracted data
Following data extraction, the completed Excel sheet was re-read several times by M.A. Descriptive data were summarized in tables and charts, and written summaries were created in Word documents to synthesize the data further. The ICF-related data (e.g., examples from each domain, the purpose of each study and the classification according to the ICF) were synthesized, using summaries written by M.A. that maintained attribution for all study data (i.e., study authors, page numbers). Data were copied and pasted from the Excel sheet into Microsoft Word, and the highlight function was used to identify common thematic groupings. Memoing was also completed by M.A. and M.E. using comments and a separate word document to help synthesize the data. The ICF data summary documents and highlighted themes were reviewed with the authorship team to discuss how data should be interpreted and reported and to ensure the entire authorship team agreed with how the data were classified in ICF domains. M.A., M.P. and P.R. all reached consensus on how to report the results.
Results
Article selection
Once duplicates were removed, 1,354 titles and abstracts were screened. After screening 15 articles together to establish mutual understanding of inclusion and exclusion criteria, two reviewers (M.A. and S.K.) independently screened over 10% of all titles and abstracts to establish reliability for title and abstract screening, both screening 143 articles. M.A. and S.K. had high inter-rater agreement (96% agreement; Cohen’s κ = 0.65). Given the high inter-rater reliability and substantial agreement according to Cohen’s Kappa (Cohen, 1960), two reviewers proceeded with individually screening the remaining titles and abstracts. A third reviewer (M.P.) was available to adjudicate any disagreement throughout the screening process, which was not needed. Figure 2 shows a PRISMA flow chart. In total, 72 full texts were assessed for eligibility. Both reviewers screened four full texts together to ensure agreement on the inclusion and exclusion criteria at full text. Both reviewers then independently screened 10% of full texts (eight). They established 100% inter-rater agreement, and therefore, two reviewers proceeded with individual review of the remaining full texts. In total, 51 studies were excluded with reasons listed in the PRISMA flow chart (Figure 2), and a total of 21 studies were included.
PRISMA flow diagram.
Population, intervention and study characteristics
Table 2 shows a summary of population, intervention and study data for the included articles. Studies were published between 2001 and 2024. In total, 536 children (age range = 1–12 years) participated across studies. Despite including specific search terms to capture youth (e.g., adolescent, teen, youth), none of the studies had participants older than age 12. As per our inclusion criteria, within each study at least 50% of participants included met criteria for autism. Consistent with the evolving terminology related to autism since 2001, diagnostic labels referring to autism varied and included ASD, autism, autistic disorder, pervasive developmental disorder not otherwise specified (PDD-NOS). Co-occurring diagnoses included intellectual disability (ID), attention-deficit hyperactivity disorder (ADHD), anxiety disorder, Down syndrome, developmental coordinator disorder, epilepsy and/or seizure disorder. Studies were conducted in seven countries, most frequently the United States (N = 5) and Canada (N = 4). Studies were primarily quantitative (N = 15), with some qualitative (N = 1) and mixed methods (N = 5) studies. Study designs varied widely, including randomized control trials (Casenhiser et al., 2013; Shire et al., 2017; Williams et al., 2024), nonrandomized control trials (Pereira et al., 2022), multiple baseline or case studies (Franco et al., 2013; Girolametto et al., 2007; Hutchins & Prelock, 2013; Hutchins & Prelock, 2006; Katz & Girolametto, 2013; van der Meer et al., 2014), studies comparing pre- and post-intervention outcomes for two groups (Lerna et al., 2012, 2014; Mohammadzaheri et al., 2022), pilot studies (Lim et al., 2007; Müller et al., 2016), a programme evaluation (Miletic et al., 2024), a single-arm feasibility study (Adams et al., 2020), a quasi-experimental design (Sun et al., 2017), a programme report (Salt et al., 2001) and an exploratory intervention study (MacEvilly et al., 2024). The qualitative study was an implementation study (Godoy et al., 2024).
Population, intervention and study data.
Notes. NR = not reported; N = sample size; ASD = autism spectrum disorder; PDD-NOS = pervasive developmental disorder not otherwise specified; ID = intellectual disability; DLD = Developmental Language Disorder; DCD = Developmental Coordination Disorder; ADHD = attention-deficit hyperactivity disorder; SLP = speech-language pathologist; SLT = speech-language therapist; OT = occupational therapist; BCBA = Board Certified Behaviour Analyst; ECE = Early Childhood Educator; Qual = qualitative data; Quant = quantitative data; Mixed = mixed methods study (i.e., qualitative and quantiative methods).
Sample size listed includes all children, including both autistic and nonautistic participants. All studies needed to have at least 50% autistic participants. We only included participants that continued throughout the entire study (e.g., excluded participants who were recruited but dropped out).
We included interventions requiring SLP involvement either in carrying out the intervention, supervising those carrying out the intervention or developing the intervention. SLP involvement had to be listed in some capacity for an intervention to be included.
In addition to speech-language pathologists, most studies involved other disciplines (e.g., occupational therapists, music therapists, teachers) and/or SLP students in intervention delivery (N = 13). Eight studies had only speech-language pathologists delivering the intervention to children and families. Some interventions were reported in multiple studies, including More Than Words (N = 2) and the Picture Exchange Communication System (PECS) (N = 2). Ten studies reported that their interventions were manualized, and 11 studies did not report any details about whether their interventions were manualized.
Social communication terminology and definitions
All studies discussed social communication concepts related to their interventions (e.g., social skills, social behaviour, pragmatic language). Table 1 shows search terms used for the concept of social communication. The terminology used to refer to social communication varied widely across studies, including whether the terms were defined, as detailed in Table 2. The most common term was ‘social’ combined with other words, such as ‘behaviour’, ‘communication’, ‘skills’, ‘competence’, ‘cognition’, ‘initiation’, ‘interaction’, ‘awareness’ and ‘responses’. Other terms included the concept of play skills, functional communication or pragmatic skills and/or competence. Most studies used multiple social communication-related terms throughout their paper. Nine studies did not define any of the social communication terms that they used in their papers, including in the introduction, methods or other sections. Twelve studies did define at least one of the social communication terms they used. Table 2 shows further details.
ICF characteristics
Table 3 provides an overview of the ICF data for all studies. None of the 21 studies explicitly referenced the ICF (WHO, 2001). All data were extracted in reference to the child (e.g., personal factors of the child, participation of the child). For interventions that had multiple areas of focus, only data related to social communication was extracted. For example, Hutchins and Prelock (2013) used social stories to target social communication but also to target behaviours unrelated to communication with children (e.g., toileting, dressing).
ICF components present in the studies.
Note. A check mark indicates that the social communication intervention data mapped onto a particular ICF domain in each paper.
Subcategories of environmental factors include (1) products and technology; (2) natural environment and human-made changes to environment; (3) support and relationships; (4) attitudes and (5) services, systems and policies.
Intervention aims and ICF domains
We analysed the intervention aims using the ICF as a lens and extracted any relevant ICF domains for each intervention’s aims. The aims for each intervention were often categorized into multiple ICF domains. In total, 20 of the 21 interventions had a participation-related aim. Only Godoy and colleagues (2024) did not report a participation-focused aim but focused on the environmental factor of parent behaviour change. Interventions varied widely regarding their participation-related aims, with some listing specific intervention aims such as children selecting topics of mutual interest or asking follow-up questions (e.g., Müller et al., 2016), while other studies mentioned general aims such as children learning ‘how to understand and manage social interaction’ (MacEvilly et al., 2024, p. 58) or ‘enhance social and communicative skills’ (Lerna et al., 2014, p. 479).
Five studies included changes in environmental factors as aims, including equipping parents with the necessary skills to continue the intervention (Lim et al., 2007); increasing parental sensitivity and responsivity with their children (Girolametto et al., 2007; Godoy et al., 2024; Miletic et al., 2024) and promoting skill generalization in multiple environments (e.g., home vs. kindergarten, peers vs. teachers) (Pereira et al., 2022). Two studies (Hutchins & Prelock, 2006; MacEvilly et al., 2024) had body structures/functions aims, including ‘emotional regulation’ (MacEvilly et al., 2024, p. 58) and ‘remediating the social, behavioral, and communicative impairments characteristic of autism spectrum disorder’ (Hutchins & Prelock, 2006, p. 47). Two studies mentioned activities not in the context of participation, specifically improving abilities such as ‘[concentrating] on given tasks . . . following instructions’ (Lim et al., 2007, p. 35) and improving ‘theoretical knowledge of anxiety and anger management strategies’ (MacEvilly et al., 2024, p. 69).
Body structures and functions
Eight papers included body structure and function components in their interventions. For information to be listed under body structures and functions, it was not discussed in the context of activities or participation. Examples included psychological and cognitive changes such as deep breathing or self-talk to manage emotions (Casenhiser et al., 2013; Lim et al., 2007; MacEvilly et al., 2024; Salt et al., 2001); improving the theory-of-mind, described as the ‘root cause of the core deficits characteristic of ASD’ (Hutchins & Prelock, 2006, p. 49); requiring eye contact for a set number of seconds (Lerna et al., 2012); increasing vocabulary size (Girolametto et al., 2007); changing responses to sensory stimuli (Salt et al., 2001) or improving basic cognitive processes such as inhibitory control (Sun et al., 2017).
Activities
Six studies reported activities not in the context of participation, including children completing social skills homework (Lim et al., 2007), playing independently in a ‘typical’ way (Hutchins & Prelock, 2013), activities based on specific cognitive skills (e.g., memorizing sequences of coloured tokens for working memory or copying mosaic patterns for inhibitory control) (Sun et al., 2017) and increasing knowledge of social communication strategies or facts without applying them in social contexts (MacEvilly et al., 2024; Müller et al., 2016; van der Meer et al., 2014). For example, MacEvilly and colleagues (2024) measured children’s knowledge of social communication strategies, without capturing whether children’s knowledge of strategies changed their social behaviour. One study (Casenhiser et al., 2013) discussed how they modified an outcome measure to change the focus from activities to participation, explaining: our modification to the coding specified that the activity take place in the context of an interaction with the caregiver . . . without the modification, a child who repetitively lines up cars while ignoring the parent might ostensibly score high in attention to interactive activity or involvement (p. 227).
Participation
All studies had participation components in their interventions. Examples varied widely across studies, but there were several common aspects of participation. Most studies discussed intervention components to increase the child’s social initiation, including initiation of play with another person, asking questions, making comments, initiation of requests or using nonverbal communication such as physically approaching peers, picture exchange or speech-generating devices (Franco et al., 2013; Girolametto et al., 2007; Godoy et al., 2024; Hutchins & Prelock, 2013; Lerna et al., 2014; Lim et al., 2007; Miletic et al., 2024; Mohammadzaheri et al., 2022; Müller et al., 2016; Pereira et al., 2022; Salt et al., 2001; Sun et al., 2017; van der Meer et al., 2014). Many studies discussed children reciprocating in play or conversations, including concepts such as turn-taking, ‘back and forth’ conversation, participation in group games and engaging in back and forth songs (Adams et al., 2020; Casenhiser et al., 2013; Katz & Girolametto, 2013; Lerna et al., 2012, 2014; Lim et al., 2007; Miletic et al., 2024; Müller et al., 2016; Pereira et al., 2022; Shire et al., 2017; Sun et al., 2017; van der Meer et al., 2014; Williams et al., 2024).
Other participation concepts included children developing friendships (Adams et al., 2020; Katz & Girolametto, 2013; Lim et al., 2007), maintaining conversation topics (Hutchins & Prelock, 2013; MacEvilly et al., 2024; Müller et al., 2016), managing social scenarios requiring problem-solving and managing their emotions (e.g., winning and losing, repairing communication breakdowns, conflict resolution, coping with mistakes) (Casenhiser et al., 2013; Lim et al., 2007; MacEvilly et al., 2024; Müller et al., 2016; Shire et al., 2017) and reducing ‘negative’ social behaviours such as interrupting (Hutchins & Prelock, 2013) or ‘insisting that others continue an activity’ (Hutchins & Prelock, 2006, p. 55). Additional concepts included modulating nonverbal communication such as speech volume (Lim et al., 2007) or behaviours to show listening, including eye contact (Lerna et al., 2012; Miletic et al., 2024; Müller et al., 2016; Salt et al., 2001).
Environmental factors
All studies discussed environmental factors. Most mentioned support and relationships, such as parent involvement to support intervention strategies (Casenhiser et al., 2013; Girolametto et al., 2007; Hutchins & Prelock, 2006; Lim et al., 2007; MacEvilly et al., 2024; Salt et al., 2001; Williams et al., 2024). Miletic and colleagues (2024) discussed support and relationships from people outside of parents, including support from other family members in implementing child-led strategies or parents benefitting from support from other parents in the intervention to reflect on their own use of strategies. Some studies discussed peers rather than parents providing support to carry out interventions (Katz & Girolametto, 2013; Shire et al., 2017). Products and technology were commonly mentioned, including intervention tools such as books or social stories (Hutchins & Prelock, 2006; Hutchins & Prelock, 2013; Katz & Girolametto, 2013), video games (MacEvilly et al., 2024), visual materials (Müller et al., 2016), augmentative and alternative communication tools (Lerna et al., 2012, 2014; van der Meer et al., 2014) or parents watching video recordings of their own interactions with their child (Godoy et al., 2024; Miletic et al., 2024). Some products discussed were related to positive or negative reinforcement for children (e.g., providing toys, token system, star chart for ‘good behaviour’) (Lim et al., 2007; Mohammadzaheri et al., 2022; Müller et al., 2016). Some interventions required technology for their online service delivery (Miletic et al., 2024) or used phone apps for parents to track their home practice (Williams et al., 2024).
Examples of natural and human-made changes to the environment included actions to generalize gains to new environments (e.g., new people, new physical spaces) (Lim et al., 2007; MacEvilly et al., 2024; Pereira et al., 2022; Salt et al., 2001) or discussed as a limitation when generalization did not occur (Mohammadzaheri et al., 2022; van der Meer et al., 2014; Shire et al., 2017). Many studies discussed their intervention occurring in children’s natural environments such as at home or school (Adams et al., 2020; Franco et al., 2013; Katz & Girolametto, 2013; Miletic et al., 2024; Pereira et al., 2022; Williams et al., 2024). Environmental changes were also discussed in relation to service delivery changes due to COVID-19 (Godoy et al., 2024). One study also discussed changes to the environment to facilitate a child’s regulation, including lighting changes or providing sensory equipment such as swings (Casenhiser et al., 2013). Attitudes were discussed in relation to parents changing their attitudes about their child’s social communication skills and what intervention activities and service delivery models can best support skill development (Girolametto et al., 2007; Godoy et al., 2024; Lim et al., 2007; Miletic et al., 2024; Williams et al., 2024). Attitude changes also included changing their knowledge and attitudes towards autism (Salt et al., 2001).
Services, systems and policies were difficult to categorize, as only three studies discussed these factors impacting intervention components. Examples included challenges applying the intervention due to differing cultural beliefs about parent involvement in intervention (Godoy et al., 2024, p. 131); the COVID-19 pandemic preventing parents from applying intervention activities outside of their immediate family (Miletic et al., 2024) and challenges delivering interventions at home compared to at school (Adams et al., 2020). Many studies provided valuable background information on how services, systems and policies in their contexts (whether country, city or health system) impacted how the intervention came to be. Details included describing the health system or organization to which an intervention was tied and how it operates (Girolametto et al., 2007; Lerna et al., 2012, 2014; Lim et al., 2007; MacEvilly et al., 2024; Salt et al., 2001); how the laws of a specific country impact the services for which children with disabilities are eligible (van der Meer et al., 2014) or what other health and rehabilitation services children received in addition to the described intervention (Casenhiser et al., 2013).
Personal factors
Nineteen studies discussed personal factors. Examples included considering how a child’s school activities overlap with intervention activities (Lim et al., 2007) or identifying how each child’s strengths, challenges and interests inform intervention goals and/or activities (Hutchins & Prelock, 2006; Hutchins & Prelock, 2013; Lerna et al., 2014; Miletic et al., 2024; Mohammadzaheri et al., 2022; Müller et al., 2016; Pereira et al., 2022; Shire et al., 2017; van der Meer et al., 2014; Williams et al., 2024). Many studies asked parents about personal factors as part of demographic information, including children’s living arrangements, childcare status (e.g., daycare or in-home, months attending childcare), siblings or parent characteristics such as their employment, mental health or marital status (Girolametto et al., 2007; Godoy et al., 2024; Katz & Girolametto, 2013; Salt et al., 2001). No studies had children report their own interests or goals, even when studies had children fill out other information, such as homework (Lim et al., 2007). Some studies discussed children’s individual sensory, physical or emotional profiles to ensure that interventions were tailored to their physical needs (Casenhiser et al., 2013; Franco et al., 2013).
Connections between ICF domains
There were connections identified between components of the ICF. For example, participation was sometimes related to personal factors (e.g., personal experiences, characteristics or preferences), such as selecting personally interesting topics of conversation (Müller et al., 2016) or considering personal factors to create more engaging interventions that increase the child’s participation (Hutchins & Prelock, 2013; Mohammadzaheri et al., 2022). Body structures and functions were combined with participation to describe making eye contact with a communication partner (Müller et al., 2016) or using emotional regulation strategies in real-world situations or teaching parents to support co-regulation (Casenhiser et al., 2013).
Some interventions that had changes in environmental factors as intervention aims (e.g., parent behaviour change) discussed child participation changes as outcomes. For example, Godoy and colleagues (2024) did not mention participation as an objective but did discuss intervention outcomes such as the child approaching other children and initiating communication as outcomes of their parent-focused intervention. More Than Words, a social communication intervention for autistic children focused on parent training, had some participation aims but was mostly environment-focused, given that it targets parental behaviour change to create supportive social communication environments for children. For studies discussing More Than Words (Girolametto et al., 2007; Miletic et al., 2024), the focal intervention aims are related to parents’ behaviour change, yet the changes in child participation were mentioned as key outcomes. Another example of participation relating to environmental factors was the role of the COVID-19 pandemic impacting a child’s ability to participate in social interactions (Godoy et al., 2024; Miletic et al., 2024).
Discussion
By applying components of the ICF, 21 studies were analysed to explore SLP social communication interventions for autistic children and youth. No studies explicitly cited the ICF. All studies discussed participation and environmental factors as part of the intervention, and some discussed body structures and functions, personal factors and activities. Although we included search terms related to youth, none of the 21 studies focused on children or youth over the age of 12. We intended to capture interventions including children between the ages of 2 and 21 years, but some interventions included children as young as age 1 in addition to older children (e.g., Lerna et al., 2014 included ages 1–6 years). The examples extracted in our study should therefore be contextualized as relating to social communication interventions for younger children between the ages of 1 and 12 years. The examples detailed within each ICF domain may be helpful to SLPs looking to understand how different components of social communication interventions link to ICF categories. Our study results also illustrate the current landscape of social communication interventions and provide detailed examples of intervention components, organized according to the ICF. Below, we discuss key takeaways related to applying the ICF to social communication interventions and describe future directions associated with this work.
Defining social communication terminology
We extracted whether each paper defined at least one of the social communication terms they used in the study; nine did not do so. Of the 11 studies that did provide at least one definition, terms were variable (e.g., social + interaction, behaviour, cognition, communication abilities, responses, initiations, cognitive performance, engagement). It is important for clinicians and researchers to define clearly the social communication-related domains they are targeting (e.g., social cognition, social skills, social relationships). A 2021 scoping review identified 293 sources that defined social communication in autism and found a lack of consensus in their definitions and an overall lack of clarity in the skills measured (Tajik-Parvinchi et al., 2021). Providing specific operational definitions is essential for transparency and rigour and helps readers make judgements about ecological validity. For example, there is evidence that social cognition, social skill and social motivation measures minimally predict social interaction outcomes for autistic adults (Morrison et al., 2020) and that contextual factors such as the social abilities of neurotypical communication partners and the built environment can be more significant predictors of social outcomes (Black et al., 2022; Bottema-Beutel, 2024; Chen et al., 2022; Morrison et al., 2020). SLPs should thoughtfully consider how this evidence impacts the social communication components they choose to target in interventions and report on what specific social communication components they are targeting.
How do SLP social communication interventions reflect the ICF domains?
Although ICF components were present in all studies extracted, none of the 21 studies explicitly discussed the ICF. This is consistent with underutilization of the ICF in other rehabilitation fields (Barradell & Scholten, 2024) and in the field of SLP (Scholten et al., 2019). For example, Scholten and colleagues (2019) studied ICF uptake by SLP students and found that there were challenges understanding the ICF beyond a rudimentary level and limited opportunity to bridge academic learning about the ICF to practice (Scholten et al., 2019). Our study’s findings similarly reflect an underutilization of the ICF in SLP social communication interventions. Future research on the specific barriers and facilitators to ICF implementation for social communication interventions would be valuable to guide decision-making.
Regarding how intervention content mapped onto ICF domains, participation was unsurprisingly the most common focus, as social communication requires another person with whom to engage and is inextricably linked to participation. Environmental factors, specifically the support and relationships subcategory, are also likely to be part of any social communication intervention, given the relational nature of social communication. Some papers discussed activities not related to participation, such as increasing knowledge of social communication concepts, rather than in the context of applying those activities in social contexts (MacEvilly et al., 2024; Müller et al., 2016; van der Meer et al., 2014). Our results must be contextualized within our definitions of ICF domains, namely, to consider activity components in the absence of participation to differentiate the categories. However, all studies that discussed participation would include activity components as part of participation.
What can be learned from the process of applying ICF concepts to SLP social communication interventions?
By assessing the presence or absence of ICF domains and reporting examples of how each ICF domain is reflected in SLP social communication interventions for autistic children and youth, we have provided a high-level overview and exploration of the literature. Our analysis demonstrates that the ICF can be used as a lens to analyse and report social communication interventions and provide a common language to discuss what the intervention aims to change, and the outcomes reported by authors, as has been done in other areas of SLP and rehabilitation (McLeod, 2009; Westby & Washington, 2017). However, describing the presence or absence of specific ICF domains may not capture whether the intervention incorporates child and family perspectives or focuses on a child’s strengths (Bölte et al., 2021; Karhula et al., 2021; WHO, 2001). In addition to exploring each domain, we must describe the interconnectedness among the ICF domains, including between contextual factors (i.e., personal factors, environmental factors) and functioning and disability factors (i.e., body structures and functions, activities, participation), as well as various permutations of combinations (e.g., connection between personal factors and environmental factors) to achieve holistic and generalizable interventions. There is some literature focusing on the bidirectional relationships between ICF domains and their real-world implications (Rouquette et al., 2015), including in the field of SLP (Dempsey & Skarakis-Doyle, 2010; Westby & Washington, 2017).
When mapping the interventions onto ICF domains, we found that similar intervention components could be categorized differently using the ICF depending on how authors reported their intervention. For example, the observed ‘behaviour’ of eye contact was coded as body structures and function when described as ‘child looked into therapist’s eyes for 2 seconds’ (Lerna et al., 2012, p. 612), without any activity or participation context provided, but was coded as ‘participation’ when described as the purpose of making eye contact to gain the attention or show engagement with a communication partner (Müller et al., 2016). This example illustrates the importance of authors clearly describing their intervention components and the context in which they occur. In other words, the ‘meaning’ of an observed ‘behaviour’ requires context. It is possible that Lerna and colleagues (2012) also targeted eye contact within a participation context when carrying out their intervention, but this is not explicitly mentioned in their description. Personal and environmental factors should be reported to contextualize how attitudes, culture, support and relationships, and services, systems and policies inform functioning domains (i.e., body structures and functions, activities, participation) (WHO, 2001). Building upon the example of eye contact, personal and/or environmental factors related to eye contact were not discussed in any studies. Eye contact is described in the autism literature as highly dependent on who the autistic person is speaking to (environment) and can feel aversive for some autistic people (personal factors) (Hadjikhani et al., 2017; Trevisan et al., 2017). Interventions risk lacking vital context and causing harm if they fail to contextualize their interventions by describing relevant personal and environmental factors.
Regarding participation, extracting data by summarizing examples of participation for all studies (e.g., making friends, having conversations with peers, playing a back-and-forth activity) provides a valuable overview of the literature, yet can miss the nuances of how participation interacts with personal and/or environmental factors. For example, the participation-focused goal of having conversations with peers was discussed by several studies (Casenhiser et al., 2013; Lim et al., 2007; MacEvilly et al., 2024; Müller et al., 2016; Pereira et al., 2022). Conversation preferences and topics of interest (personal factors) often differ between autistic and neurotypical people (Crompton et al., 2020; Heasman & Gillespie, 2019; Morrison et al., 2020) and within the heterogeneous group of autistic people (Crompton et al., 2020). Incorporating personal factors would allow clinicians to consider whether having the autistic child apply neurotypical conversation norms (e.g., taking turns, small talk, nonverbal communication) aligns with that child’s preferences for socializing.
Limitations and future directions
The authorship team did not include autistic researchers or people with lived and living experience; their inclusion might have yielded different research questions or perspectives in data analysis. Our analysis was also limited by the detail provided by the original authors on their interventions, which varied across studies. We acknowledge that how authors reported on their interventions may differ from how they were carried out. We did not complete quality appraisal or risk of bias assessment for the studies extracted as this is an optional component of scoping reviews, and our primary objective was to explore applications of the ICF to SLP social communication literature without a specific focus on the quality of the interventions analysed. However, appraisal of the studies and assessment of bias could be a valuable future direction for researchers interested in this information on the SLP social communication interventions.
In addition, none of the studies included participants older than 12 years of age. Our results therefore reflect SLP social communication interventions for younger children, rather than youth. There are several reasons why youth may not be represented. First, many popular social communication interventions for autistic youth are not specifically SLP social communication interventions (e.g., PEERS; Laugeson et al., 2012) and therefore would not have fit our inclusion/exclusion criteria of SLP involvement in the intervention. Second, autism research primarily focuses on younger children and ‘early intervention’, and youth may be studied less often (Jang et al., 2014; Maksimović et al., 2023). Social communication interventions from other disciplines were outside of the scope of this review, which focused on the SLP literature. However, future research may wish to examine social communication interventions for autistic children and youth broadly, as evidence from other fields may be of interest to speech-language pathologists, particularly for youth. Another future direction could be to use the ICF as a lens to study social communication interventions in specific fields such as occupational therapy, behaviour therapy, psychology, or education, to reflect on how various professions conceptualize social communication interventions using the ICF as a lens.
A future direction for our authorship team is to apply the findings from this article to analyse SLP social communication interventions with additional detail, including whether interventions are strengths- or deficit-based, and whether there is child or family involvement in intervention goals or outcome measurement. These explorations would provide valuable information on SLP social communication interventions but are outside of the scope of this article, which focuses on how the ICF can be used as a lens to analyse social communication interventions. Although the ICF is considered a strength-based and holistic framework, the ICF (WHO, 2001) does not include a formal way to describe whether the information provided in the ICF domains is strength-based or deficit-based. To capture the strength-based element of the ICF for children and youth, clinicians and researchers may consider using language from the F-Words for Childhood Development (Rosenbaum & Gorter, 2012), a paediatric ‘animation’ of the ICF that uses strength-based language such as functioning, friends and fun, among other terms, and is written from the child’s perspective.
Putting it all together: illustrating how ICF concepts could be considered in an intervention
In our discussion, we have described how SLP social communication interventions can be analysed through the lens of the ICF, and have outlined various considerations for applying the ICF in a manner that considers contextual factors. In the following, we have created a fictional example of how an SLP social communication intervention can apply the ICF:
A fictional SLP social communication intervention, ‘Connections’, aims to increase the amount of time spent socializing on the playground, and does so by measuring the length of time that 7- to 9-year-old autistic children spend in social interactions with peers at a playground. The intervention uses the outcome measures of parents reporting how long their child spends interacting with peers at the playground, children reporting on their own playground participation and a therapist watching videos of the children at the playground and coding them for pre- and post-intervention observations.
The ICF could serve as a valuable framework for researchers and clinicians to consider the intervention structure, aims and outcomes. The researchers decide to use the ICF in conjunction with another intervention planning framework, the Rehabilitation Treatment Specification System (RTSS) (Van Stan et al., 2019), to outline clearly their intervention aims and key ingredients (Hamilton et al., 2024). For example, the researchers identify that their intervention primarily targets child participation, but after looking at the ICF, they also wonder if their intervention targets environmental factor changes, specifically support from parents to facilitate interactions at the playground, or attitude changes in parents to understand the different ways that autistic children may enjoy playing (O’Keeffe & McNally, 2025). The researchers ask themselves how their intervention will consider personal factors and environmental factors: How will they obtain information about the children’s baseline communication preferences and skills? What aspects of the parents’ or children’s attitudes on playing with peers do they plan to change? How will they capture the culture of each family and how their values may influence intervention uptake? How will they describe if intervention strategies were individualized to each child and family’s strengths, preferences and goals? These questions are merely a few examples to illustrate how applying the ICF to guide our thinking can lead researchers and clinicians alike to develop social communication interventions thoughtfully. Figure 3 shows an illustration of these reflections, which can be used as a visual to prompt SLP researchers and clinicians to consider how contextual factors and functioning and disability factors interact.
ICF functioning and disability factors and contextual factors.
Conclusions and clinical implications
SLPs delivering social communication interventions may find our application of the ICF helpful to evaluate and reflect upon their intervention practices, including how the ICF framework may be a valuable lens through which to view their therapy practices. Figure 3 and the associated case study provide an example of how the ICF can be applied to a fictional SLP social communication intervention. This article provides a high-level overview of the ICF domains that are commonly present or absent in SLP social communication interventions and examples of each domain. This review adds to the body of literature applying the ICF to SLP practice, working towards holistic, strengths-based interventions that consider the impact of personal and environmental factors on functioning, disability and health. When applying the ICF, clinicians and researchers must be clear about what aspects of social communication their intervention targets and ensure that relevant terminology is defined. The ICF’s personal and environmental factors domains can create a space for clinicians and researchers to reflect upon the alignment between intervention goals and the contextual factors (e.g., cultural norms, communication preferences) that are relevant to the autistic child and their family.
Footnotes
Acknowledgements
We gratefully acknowledge Dr Vanessa Tomas, who provided guidance on this manuscript as the course instructor for McMaster University’s ICF Theory and Use course (Nguyen et al., 2016). We also gratefully acknowledge Laura Banfield, McMaster University Librarian, for her guidance on our search terms and databases, Minseo (Sunny) Kim (M.K.) for her contributions to screening articles for this review, and Mona Elmikaty (M.E.) for her support as a second coder for our data extraction.
Author contributions
Funding
The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: M.A. is funded by a Social Sciences and Humanities Research Council (SSHRC) Canada Graduate Scholarship Doctoral Award (CGS-D).
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Data availability statement
Sample selection and demographic characteristics
This is a scoping review in which we analysed the published literature, and therefore, no data were collected directly from individuals or families. We therefore have no recruitment or sampling information to report. We have included some demographic characteristics (e.g., age) but did not include gender, race/ethnicity and socioeconomic status, given they were not part of our research question.