Introduction

Neurodiverse children, including those with autism spectrum disorder (autism) and/or an intellectual disability (ID), are at heightened risk of experiencing interpersonal trauma such as abuse, neglect, and exposure to family violence (Fang et al., 2022b). It is estimated that these children are approximately two to three times more likely to encounter traumatic events of an interpersonal nature relative to their typically developing counterparts (Fang et al., 2022b). Ecological systems theories regarding such vulnerability indicate that child-related factors such as a high dependence on caregivers and/or communication challenges interact with environmental factors such as parent mental health challenges to confer these children’s risk of traumatic exposure (Ammerman et al., 1994; Fisher et al., 2008; Gore & Janssen, 2007; Hibbard & Desch, 2007; Howe, 2006; Sobsey, 2002; Verdugo et al., 1995).

Interpersonal trauma can profoundly impact the mental health of neurodiverse children, as characterised by the onset of depressive symptomatology, self-injurious behaviour, and post-traumatic sequalae including intrusive thoughts, distressing memories, and sleep disturbance (Bleil Walters et al., 2013; Brenner et al., 2017; Mandell et al., 2005; McDonnell et al., 2019; Mehtar & Mukaddes, 2011; Paquette et al, 2017). Kerns et al. (2015) hypothesised that such difficulties manifest due to the emergence of new symptomatology and/or an exacerbation of a child’s existing disability-related characteristics. Indeed, while this conceptual framework was initially developed to explain the intersection between autism and trauma, the high degree of overlap, co-occurrence, and shared interpersonal trauma vulnerability between children with autism and an ID suggests that this framework may further help to explain trauma-related sequalae in children with an ID (Matson & Shoemaker, 2009; McDonnell et al., 2019; Reiter et al., 2007).

Despite the significant research efforts directed toward understanding this group’s vulnerability, it is currently unclear how to best support their mental health. This has likely been due in part to a siloed approach to disability and trauma research (Prock & Fogler, 2018) and a historical tendency to misattribute trauma-related difficulties to a child’s primary neurodevelopmental diagnosis rather than to concurrent psychopathology (Mazefsky et al., 2012; Reiss et al., 1982). As a result, evidence-based therapeutic guidelines for this group of children are lacking, and best practice parameters have yet to be established.

The purpose of the current review was to examine the existing literature regarding therapeutic supports for children with autism and/or an ID who have experienced interpersonal trauma. We included therapeutic supports that emphasised psychological, social, emotional, and/or behavioural factors, and aimed to improve a child’s mental health and/or functioning. We were particularly interested in the types of clinical techniques, strategies, and/or modalities that have been reported in the literature and their associated methods of delivery. The specific aims of our review were: 1) to capture studies that have implemented a therapeutic support directly with this group of children and to understand whether their methodological characteristics were similar enough to allow for comparison across studies; 2) to synthesise literature that has included clinical recommendations for working therapeutically with these children - while such recommendations have been traditionally overlooked, they have the potential to reveal patterns of consensus that can be used to guide future clinical and research endeavours; and 3) to integrate therapeutic supports pertaining to both autism and/or an ID into a cohesive body of work and consolidate research that has used inconsistent terminology to capture the interpersonal trauma construct.

Method

Given the expected paucity of empirical literature in this clinical domain, a scoping review was selected as the preferred review method due to its exploratory approach (Munn et al, 2018). The Joanna Briggs Institute (JBI) scoping review methodology was employed and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) reporting guideline was followed (Peters et al., 2020; Tricco et al., 2018).

Eligibility Criteria

Eligibility criteria for an article differed based on whether the article had implemented a therapeutic support or made a clinical practice recommendation for working therapeutically with this group of children. Inclusion and exclusion criteria are presented in Table 1.

Table 1 Eligibility criteria
Full size table

Search Strategy

The JBI three-step search strategy was employed for this review (Peters et al., 2020).

Step one: A limited search of two online databases relevant to the topic (i.e. MEDLINE and PubMed) was conducted to identify initial keywords in the titles and abstracts. In consultation with a tertiary librarian, a comprehensive list of indexed terms and combinations of synonyms/interchangeable terms (i.e. syntax-combinations) was developed.

Step two: In October 2021, a targeted and comprehensive search using all relevant terms and database-specific age limits was applied across the following four databases: Ovid MEDLINE, Embase, Emcare, and PsychInfo. The search was not limited by date of publication or study design, and indexed terms were adapted to each of the databases. A full search strategy for MEDLINE is presented in Supplementary Table A.

Step three: A manual search of the reference lists of all eligible full-text sources was undertaken to identify additional articles that had not yet been included.

Selection of Literature

Three levels of screening were conducted. First, records which clearly met the exclusion criteria were reviewed and removed by one researcher (LK), after which the remaining records were screened by three researchers on the web-based software platform Covidence (LK, JB, and LM; Veritas Health Innovation 2015) at the title and abstract levels. Remaining records then underwent full-text review by two researchers (LK and JB), wherein any conflict was resolved through verbal consensus.

Data Extraction and Synthesis

An extraction form was developed and updated after trialling it on two eligible sources. Extraction differed according to whether the article had implemented a therapeutic support or made a clinical practice recommendation for working therapeutically with this group of children.

Implemented therapeutic supports: Participant and methodological characteristics, together with explicit details of the therapeutic support, were recorded. Therapeutic supports were grouped according to existing psychotherapeutic frameworks (i.e. cognitive-behavioural, attachment-theory).

Recommendations: Recommendations were synthesised using a deductive thematic analysis framework (Braun & Clarke, 2006; Kiger & Varpio, 2020). Recommendations were summarised in both a tabular and descriptive format, and the number of articles that addressed each theme was recorded. Some articles mapped onto multiple themes and were therefore cited more than once.

Results

The search retrieved a total of 3193 records, with a further eight sourced through manually searching the reference lists of the identified articles. After a process of duplicate removal and two stages of screening, 207 articles underwent full text review, with a total 52 articles being deemed eligible for inclusion. A PRISMA diagram presents the selection process (see Fig. 1). The 52 included articles are summarised in Table 2.

Fig. 1
figure 1

PRISMA flow diagram of the article selection process

Full size image
Table 2 List of included articles
Full size table

Overview of the Literature

Fifty-two articles described therapeutic supports for children with autism and/or an ID who have encountered interpersonal trauma (see Table 2). Of these, 15 were primary studies that implemented the support directly with this group of children (and/or their parents/caregivers), while the remaining 37 provided a clinical recommendation intended to guide future therapeutic efforts.

Primary Studies that Implemented a Therapeutic Support

Across the 15 studies that implemented a support, there was great diversity in therapeutic modality wherein seven clusters of supports were identified. Table 3 presents these support clusters, participant, and methodological characteristics, together with study findings.

Table 3 Characteristics of studies that implemented a therapeutic support
Full size table

Therapeutic Approaches

Four studies reported on eye movement desensitisation and reprocessing therapy (EMDR), one of which had adjunctive elements that included prolonged exposure and physical activity (Mevissen et al., 2020) and another incorporated a dedicated parenting skills component (Ooms-Evers et al., 2021). Three studies delivered cognitive-behavioural support, namely, manualised Trauma Focused Cognitive Behavioural Therapy (TF-CBT; Holstead and Dalton), the trauma narrative component of TF-CBT in conjunction with family therapy (Gerhardt & Smith, 2020), and nonspecific CBT alongside pharmacological treatment (Lin et al., 2020). A separate study employed an exclusively behavioural intervention that involved highly individualised strategies (i.e., reinforcement, arousal modulation, physical activity, and routine; Kildahl et al., 2021), whereas another implemented a blended behaviour-attachment focussed support that aimed to cultivate sensitive and responsive child-caregiver interactions and increase adaptive behaviour (Sterkenburg et al., 2008). A further two studies delivered established attachment interventions that included child parent psychotherapy (CPP; Harley et al., 2014) and Theraplay (Mohamed & Mkabile, 2015), and two others implemented non-directive therapeutic supports that drew upon art and symbolic play (Guest & Ohrt, 2018; Horovitz, 1981). Finally, the two studies remaining implemented “psychotherapy” without mention of a specific therapeutic modality or support target, both of which mentioned parental involvement (Cook et al., 1993; de Pilar Trelles Thorne et al., 2015).

Participant Characteristics

Across the 15 studies, six reported on participants that had a diagnosis of autism, three on children with an ID, and six on those who had both autism and an ID (see Table 3). Additional presenting difficulties included but were not limited to: post-traumatic stress disorder (PTSD) or subthreshold trauma-based symptomatology, self-injurious behaviour, anxiety, depression, attention deficit-hyperactivity disorder, disrupted attachment, and visual impairments. Interpersonal trauma exposure typically occurred within the family system in all but two studies, and nine studies involved parents/caregivers in the intervention process.

Methodological Characteristics

The 15 studies consisted of four case series, eight case reports, a repeated measures study, and a nonrandomised trial (see Table 3). Twelve of the studies described the therapeutic support in replicable detail by citing a treatment manual or by providing explicit details about the nature of the therapeutic support. The remaining three provided a reference to a support modality and/or alluded to an intervention component but did not elaborate. Eight of the studies specified their target outcomes from the outset (i.e., as part of their methodology), and similarly, eight studies utilised standardised assessment instruments to operationalise their target outcomes and collect data on their participants. Instruments varied in focus and assessed for trauma-based symptomatology, daily life impairment, broad mental health and behavioural outcomes, parental stress, and psychopathology. The seven studies remaining collected information on their participants via unstandardised parental report and/or clinical observation.

Only seven studies evaluated the efficacy of their therapeutic support by comparing standardised assessment data pre-to post-intervention. Three plotted their data graphically and/or in tabular form to visually inspect the data for trends, while four utilised formal statistical analysis techniques as their primary analysis method. These seven studies spanned five therapeutic support clusters (i.e., EMDR, TF-CBT, behavioural, behavioural-attachment, and attachment). The remaining eight studies described participants’ characteristics after implementation but, without baseline data, were unable to evaluate efficacy of the therapeutic support.

Findings of Studies that Evaluated Efficacy

Of the seven studies that assessed efficacy, reductions in post-traumatic symptomatology, general psychiatric challenges, behavioural difficulties and daily life impairment were reported. Two studies reported improved parent outcomes (Harley et al., 2014; Mevissen et al., 2020); however, in one study, scores on the majority of parental stress subscales remained elevated post intervention (Harley et al., 2014). Of note, the results of one study indicated that their TF-CBT intervention was largely ineffective in reducing internalising and externalising problems in children; however, aggressive behaviour and rule-breaking did decrease (Holstead & Dalton, 2013).

Method of Delivery

Eight of the studies that implemented a therapeutic support used nonverbal, sensory, and/or play-based therapeutic techniques (Guest & Ohrt, 2018; Harley et al., 2014; Horovitz, 1981; Kildahl et al., 2021; Mevissen et al., 2011, 2017, 2020; Mohamed & Mkabile, 2015), and an additional study made an explicit attempt to integrate their participant’s interests and preferred activities into the intervention process (Gerhardt & Smith, 2020). Efforts to modify language and match the communication abilities of participants were reported across five studies, which included the provision of clear instructions (Mevissen et al., 2011, 2017, 2020), simplified communication (Guest & Ohrt, 2018; Ooms-Evers et al., 2021), and the use of positive language (Harley et al., 2014). A further study emphasised the importance of reinforcing and validating verbal communication attempts (Kildahl et al., 2021). Finally, structural modifications to sessions were noted across a small number of studies and included ensuring predictability (Sterkenburg et al., 2008), slowing down the pace of sessions (Guest & Ohrt, 2018), prioritising one task at a time (Mevissen et al., 2020), and providing frequent breaks (Harley et al., 2014).

A Synthesis of Recommendations

Article Characteristics

The 37 articles that made a recommendation for working therapeutically with this group of children comprised two book chapters, 12 reviews, and 12 commentaries, with a small proportion (n = 11) being primary and secondary studies (see Table 2). As demonstrated in Table 2, the primary objective of a large portion (n = 23) of these articles was to provide recommendations, whereas the recommendations made by the remaining 14 were a subsidiary of a separate yet related research endeavour.

A visual inspection of these articles’ reference lists revealed that only nine of the 15 studies that implemented a therapeutic support were cited (Cook et al., 1993; Harley et al., 2014; Holstead & Dalton, 2013, Mevissen et al., 2011, 2017, 2020; Mohamed & Mkabile, 2015; Sterkenburg et al., 2008; De Pillar Trelles Thorne et al., 2015). The remaining citations were comprised of commentaries, opinion pieces and reviews, or primary studies that reported on therapeutic supports for adults with a history of childhood trauma, neurodiverse children who had not encountered trauma, and children who had experienced trauma but did not have an identified neurodevelopmental diagnosis.

Thematic analysis of the 37 articles revealed two distinct clusters of recommendations: child-focussed and parent/caregiver-focussed. Four different categories of therapeutic support were identified within each cluster, which are summarised below and detailed in Table 4.

Table 4 Recommendations classified by theme
Full size table

Child Focussed

Seventeen articles recommended that children be assisted with their emotional literacy and/or adaptive emotional regulation, wherein explicit strategies like mindfulness and progressive muscle relaxation were frequently mentioned. Eighteen articles indicated that behavioural targets should be, at least in part, the focus of therapeutic support efforts, with a large portion advocating for education about safety skills, intimacy, and interpersonal relationships. Fourteen articles made a recommendation that centred on modifying children’s thinking patterns, with many advising that clinicians should address children’s cognitive distortions, support children to construct an accurate trauma narrative, and provide psychoeducation about trauma. Finally, twelve suggested general psychotherapeutic approaches (e.g. CBT, EMDR, and counselling); however the majority of these did not elaborate on a specific therapeutic target.

Parent/Caregiver Focussed

Twenty-four articles recommended that parents/caregivers be part of the therapeutic work with this group of children. Eight of these articles did not specify what such involvement would entail (Byrne 2020; Dion et al., 2013; Mansell et al., 1996; Rose & Hardman, 1981; Schneider et al., 2019; Sullivan, 1993; Tharinger et al., 1990; Vervoort-Schel et al., 2021); however, fourteen emphasised that parents may require individual support with their own challenges. Ten articles advocated for parent psychoeducation, with the most common suggestion being that parents should be educated about trauma and its widespread effects, the unique needs of neurodiverse children, what is and is not within their child’s control, and that harsh discipline will. Additionally, thirteen articles recommended that parents be taught specific, practical skills to assist them in their parenting role (i.e. how to attune to their child’s needs and engage in emotion-focussed conversations about trauma), and separately, seven recommended that therapeutic support should focus on parent–child attachment with child-parent psychotherapy emerging as a recurrent suggestion (Lieberman et al., 2005).

Recommended Methods of Delivery

Twelve articles recommended that interactive, creative, play-based, and sensory tools be utilised when supporting children in a therapeutic capacity (e.g., role plays, art, drama, and social stories). The utilisation of visual and pictorial strategies was explicitly recommended by six articles and included the use of visual timers, visual activity schedules, story boards, and index cards. Nine articles recommended that clinicians should modify their language to meet the needs of this group of children, while, similarly, nine articles suggested that adaptations be made to session structure and content. Specific recommendations centred on ensuring predictability, shortening the length of sessions, providing ample breaks, and slowly introducing emotionally laden topics.

Discussion

To date, there has been a lack of clarity regarding what constitutes best therapeutic practice for children with autism and/or an ID who have a history of interpersonal trauma. To form a more coherent picture of this clinical domain, the current review identified studies that have implemented a therapeutic support with this group of children and synthesised clinical practice recommendations intended to guide future support endeavours.

Paradoxically, despite their high-risk status, only 15 studies were identified that implemented a therapeutic support with our population of interest. Across these studies, seven main types of therapeutic approaches emerged, which were marked by diverse participant characteristics and methodologies that varied in terms of their study design, target outcomes, the operationalisation of variables, and use of efficacy analyses. Unfortunately, the sheer paucity of primary studies in this body of literature and the significant participant and methodological heterogeneity observed limited even preliminary empirical assertions from being made in this review. As such, the extent to which one therapeutic support modality may have more “clinical promise” or “best practice potential” over another could not be reliably determined.

That said, a noteworthy discovery was that the types of therapeutic support implemented largely parallel first-line evidence-based treatments recommended by the National Institute for Health and Care Excellence (NICE) for children in the general population who have experienced interpersonal trauma. These include TF-CBT, attachment-based therapies, CPP, and EMDR (NICE, 2017, 2018). While further research is needed to develop an empirical approach to supporting the mental health of neurodiverse children with a history of interpersonal trauma, it is reassuring that several of the implementation studies drew upon existing therapies that are both scientifically grounded and widely endorsed within the general population.

Another interesting pattern to emerge across implementation studies was that few had an exclusively verbal emphasis and/or demanded a high cognitive load. This pattern further traversed the clinical practice recommendations wherein therapeutic support with a behavioural focus and/or an emotional target garnered the highest number of recommendations. While it is a common misconception that children with autism and/or an ID are unable to benefit from talk-based therapies (Gaus, 2007; Weston et al., 2016), future research may benefit from comparing the relative effectiveness of purist cognitive modalities, over other more behavioural and/or emotion-focussed therapeutic approaches when supporting this group of children.

Of note, recommendations that were behaviourally focussed centred on enhancing children’s communication skills and providing education on interpersonal relationships, safety, and abuse. This aligns with converging evidence demonstrating that a lack of skills and/or knowledge in these areas is significant risk factor for interpersonal trauma exposure in children with autism and/or an ID (Miller et al., 2017) and is core component of primary, secondary, and tertiary prevention programs for neurotypical children who have encountered similar traumatic experiences (Kim, 2010). With respect to the emotionally anchored recommendations, it was repeatedly suggested that children’s emotional vocabulary and regulation be the focus of therapeutic support efforts. Indeed, given that these emotional domains already prove challenging for neurodiverse children and are often amplified following interpersonal trauma exposure, it is unsurprising that recommendations of this nature repeatedly surfaced (Haruvi-Lamdan et al., 2018; Hill et al., 2004; Mazefsky et al., 2013; McClure et al., 2009; Rieffe et al., 2007).

Despite the consistency observed, the clinical practice recommendations did not appear to explicitly favour one particular type of therapeutic support over another. This could simply suggest that there is no “one size fits all” approach to working with this group of children. More likely however, is that in the absence of a substantial and robust evidence base wherein the majority of recommendations were not empirically supported, there exists considerable variation across clinical and research opinion. As such, the recommendations detailed in this review must be viewed as potential signposts for future clinical research but not as definitive evidence of what therapeutic supports should be implemented with this population.

Perhaps of greatest clinical utility was the discovery that the bulk of the literature presented in this review indicated that parents/caregivers be involved in the therapeutic work. While historically research tended to over-emphasise characteristics of the individual child in the attribution of interpersonal trauma vulnerability (Manders & Stoneman, 2009; Thomas-Skaf & Jenney, 2021), the findings of this review reinforce ecological risk models demonstrating the role of the systemic context in this complex clinical picture (Algood et al., 2011). Whether parents/caregivers are instigators of harm, inadvertent contributors, or are unclear about how to respond to their child’s traumatic encounter, a systems-approach to therapy is widely acknowledged as fundamental to harm reduction and therapeutic success in this group.

An ancillary discovery was that the methods of intervention delivery endorsed across the literature were consistent with those that have been well-defined in the broader neurodevelopmental literature (Spain & Happé, 2020; Surley & Dagnan, 2019). While this could suggest that therapeutic modifications for children with autism and/or an ID are homogenous irrespective of whether the child has experienced interpersonal trauma or not, this supposition has yet to be formally tested.

There were some limitations in the present study that should also be considered. First, despite employing a rigorous and systematic literature search with a diversity of terms, as well as a manual search of the reference list of all included articles, it is possible that not all studies of relevance were captured. Further, the frequency with which similar recommendations emerged is not necessarily a proxy of their relative significance. While it is hoped that a degree of intentionality underscored the provision of each recommendation, authors may have respectfully echoed recommendations of relevance in a somewhat perfunctory manner. It is, however, plausible that recommendations with high frequency counts do reflect areas of priority in the field and certainly provide possible avenues for exploration. Finally, as the literature on therapeutic supports for this group had yet to be sufficiently integrated, a systematic review with a comprehensive quality appraisal was deemed premature. That said, given that the primary objective of this review was to map and synthesise the “current state of affairs” in this clinical space, we were able to capture nuanced information that would have been lost had a systematic review been conducted instead.

In conclusion, this review highlights that the literature on therapeutic supports for children with autism and/or an ID and a history of interpersonal trauma is scant, diffuse, and largely lacking in methodological consistency. While there is clearly an absence of evidence to guide best practice, the emphasis on parent/caregiver involvement and the ever-so-slight shift away from verbal and/or cognitively demanding modalities make important progress toward understanding potential therapeutic support models for this group of children. Ultimately, it is hoped that this timely synthesis will incentivise researchers to work alongside families and clinicians to develop and systematically evaluate potential therapeutic supports for this vulnerable population, and that an empirical consensus will materialise in the not-too-distant future.