Considerations for the identification of autism spectrum disorder in children with vision or hearing impairment: A critical review of the literature and recommendations for practice

Abstract

Objective:

There is higher risk for autism spectrum disorder (ASD) across many pediatric neurological conditions characterized by vision impairment or hearing loss. Early and accurate identification of ASD is imperative in promoting access to appropriate and early evidenced-based intervention; however, differential diagnosis can be particularly challenging in children with sensory impairment given the heterogeneity of ASD combined with the impact of vision impairment or hearing loss/deafness on development and behavior. A neuropsychologist’s unique expertise and appreciation of the interplay between sensory and behavioral manifestations can be valuable for making an early and accurate ASD diagnosis in children who are blind/visually impaired or deaf/hard-of-hearing. This article highlights clinical considerations when identifying ASD within the context of vision impairment or hearing loss/deafness.

Method:

We discuss clinical considerations for the early identification of ASD in children who are blind/visually impaired and deaf/hard-of-hearing. Information presented in the article is based on a critical review of the literature and the expertise of the author group.

Conclusion:

Ongoing development of clinical expertise and evidence-based assessment methods are important when informing the early differential diagnosis of ASD in individuals with sensory impairment. Accurate identification is also vital for the development of targeted interventions across the lifespan.

Introduction

Autism Spectrum Disorder (ASD) is a behaviorally defined disorder characterized by impairments in social communication and the presence of restricted and repetitive patterns of behavior (RRBs). Recent estimates of autism prevalence are 1 in 54, or about 1.8% (Maenner, Shaw & Baio, 2020); however, there is an increased risk for ASD across many pediatric neurological conditions associated with sensory impairment (e.g., Down syndrome, epilepsy, Fragile X, neurofibromatosis type 1, Noonan Syndrome, preterm birth, tuberous sclerosis complex; Agrawal et al., 2018; Strasser, Downes, Kung, Cross, & De Haan, 2018; Richards, Groves & Oliver, 2015). Early and accurate differential diagnosis of ASD can be particularly difficult in children who are blind/visually impaired (BVI) or deaf/hard-of hearing (DHH) given the complex impact of these sensory differences on social development and behavior. As such, consideration of symptom etiology is crucial to making differential diagnosis of ASD in children with vision impairment or hearing loss/deafness.

Diagnostic overshadowing (VandenBos, 2007) occurs when symptoms that can be explained by multiple factors are erroneously attributed to one particular diagnosis, thus leading to inaccurate diagnostic classification. When considering an ASD diagnosis within the context of sensory impairment, diagnostic overshadowing can take two forms: 1) misdiagnosis of ASD when symptoms are attributed to ASD, but symptoms are truly better explained by the impact of the sensory impairment, and 2) a missed diagnosis of ASD when symptoms are attributed to the sensory impairment, but are best explained by a diagnosis of ASD. For example, an 18-month-old child with a history of hypoxic-ischemic encephalopathy (HIE) resulting in injury to the visual cortex and cortical visual impairment (CVI), may not look at the face of the examiner during an evaluation. Whereas this could be misattributed to social avoidance commonly observed in ASD, children with CVI may avoid looking at the face because they are unable to process the visual complexity of the face. Without appreciating the injury pattern secondary to HIE and symptoms associated with CVI, a clinician may misattribute facial avoidance as a symptom of ASD. Accurate diagnosis is vital because the treatment approach for ASD (i.e., supporting social skills) is very different from the treatment for CVI (i.e., supporting visual processing). In contrast, an 18-month-old toddler with a history of Down syndrome and hearing loss may present with reduced responsiveness to social stimuli consistent with a true diagnosis of ASD, but this diagnosis is deferred because the clinician misattributed reduced responsiveness to the child’s hearing loss. Without understanding the manner in which social responsiveness is expressed in children with DHH, the result is a missed opportunity for social-based intervention during a critical period of brain development.

Given the lack of ASD screening and diagnostic tools validated for use in individuals with vision impairment or hearing loss, there is greater reliance on clinical judgement when making an ASD diagnosis than is typical when evaluating individuals for ASD in the absence of vision or hearing impairment. Neuropsychologists with training in early neurodevelopment may be particularly well suited to differentiate ASD in the context of sensory impairments given their knowledge about typical early development, differences in the acquisition of developmental milestones secondary to sensory impairments, and appreciation of any interactions with complex medical and/or neurological factors that underlie the sensory impairments. Expertise integrating all of these key components is inherent to an accurate neuropsychological case conceptualization.

The goal of this article is to discuss clinical considerations for neuropsychologists when assessing risk for and/or diagnosing ASD within the context of BVI and DHH, given the dearth of clinical guidelines for the best practices when evaluating for ASD in these populations, especially those who have underlying medical and/or neurological conditions of BVI and DHH. Neuropsychologists who do not routinely work with children with vision impairment or hearing loss should be aware of their unique and typical social and behavioral presentations across development to inform differential diagnostic considerations. Therefore, this article will focus on the unique aspects of social and behavioral presentation among children with BVI and DHH, including specific differences among those children with and without diagnoses of ASD.

The clinical guidance to follow is informed by the expertise of the author group and is supported by research when available. We intentionally utilize both person first and identity first language throughout this article in recognition of differing preferences within the disability community. Additionally, any use of the phrase “ASD diagnosis” in the sections below refers to a clinical diagnosis of ASD.

Blindness/Visual Impairment (BVI)

World Health Organization criteria defines visual impairments as visual acuity less than 0.3 (typical visual acuity is 1.0) or a visual field of fewer than 30 degrees (typical visual field is 180 degrees) (Kiani et al., 2019; Van den Broek, Janssen, van Ramshorst & Deen, 2006); however, there is great variability in the severity and type of visual disturbance based on etiology. Individuals with BVI may have disruptions to the functioning of the eye and optic nerve anterior of the optic chiasm, called ocular vision impairment, whereas others may experience vision deficits due to disruption of posterior visual pathways, called cortical visual impairment (CVI; Chokron, Kovarski, Zalla, & Dutton, 2020). CVI is characterized by difficulty with processing of visual information in general, including visual perception and/or visual attention. It is associated with unique visual behaviors (e.g., light gazing, color preference, need for movement) that present on a continuum of severity and are used to facilitate identification (Roman-Lantzy, 2018). CVI is the leading cause of congenital blindness in the United States (Swaminathan, 2011) and is most commonly associated with neurologic injury secondary to cerebral anoxia/hypoxia, preterm birth, and hydrocephalus (Khetpal & Donahue, 2007; Luek et al., 2019). It also has been reported that 10.5% of children with neurodevelopmental disabilities have CVI (Swaminathan, 2011). Although there are a number of causes of BVI and many factors that may alter the degree/type of visual impairment, we will refer to this population in general as BVI, unless otherwise indicated by research findings.

The history, duration, and quality of visual exposure are important factors to consider in understanding the degree of impact of BVI itself on brain re-organization and neurodevelopment. Congenital forms of BVI impact individuals during the neonatal period (i.e., vision is disrupted very early with limited to no period of typical vision), whereas acquired forms of BVI occur after the neonatal period (i.e., after a period of typical vision). Examples of congenital forms of BVI include congenital structural injury or malformation to the ocular and/or visual system, which may be due to a genetic condition (e.g., ocular malformations; optic nerve hypoplasia), preterm birth (e.g., retinopathy of prematurity), neonatal infection (e.g., corneal perforation due to ophthalmia neonatorum), or neonatal stroke or asphyxia (e.g., optic nerve atrophy, hypoxic-ischemic encephalopathy, cortical atrophy in the occipital lobe). Acquired forms of BVI may be associated with emergence of a brain tumor (e.g., optic glioma), traumatic brain injury/stroke (e.g., occipital hemorrhage), or a genetic condition (e.g., CLN3/Batten disease).

The etiology of BVI is also important to consider because associated neurological injury or malformation that may accompany BVI contributes to the degree of neurodevelopmental risk beyond visual impairment alone. For example, young children with BVI are most likely to have congenital forms associated with broader neurological conditions that impact very early neurodevelopment. In a study characterizing prevalence of BVI in 2,155 children from birth through age 3 years entering early intervention services, 40% of children with BVI were considered legally blind, and 68% presented with multiple disabilities (Hatton, Schwietz, Boyer, & Rychwalski, 2007). The most common visual conditions in this group included CVI, retinopathy of prematurity, and optic nerve hypoplasia (Hatton et al, 2007).

Prevalence of Co-Morbid BVI and ASD & Etiological Considerations

The prevalence of ASD in the BVI population is higher than in sighted children. A systematic review and meta-analysis examining 10 articles published from 1994 to 2016 (Do et al., 2017) found an overall ASD prevalence of 19% in children with BVI, with a high degree of variability in prevalence based on specific etiologies of BVI and associated clinical factors. Conditions causing congenital forms of blindness, including total blindness (i.e., individuals who have no light perception) are the most strongly associated with autism (Hobson & Bishop, 2003; Jure et al., 2016). Jure and colleagues (2016) found that 65% of children with retinopathy of prematurity in their study also met diagnostic criteria for ASD, and Mukaddes and colleagues (2007) found that a diagnosis of ASD was associated with greater severity of visual impairment and neurological involvement (i.e., intellectual level and presence of cerebral palsy). It is important to consider that current prevalence estimates are limited by the lack of appropriately modified and well-validated ASD diagnostic tools in this population that can be used to inform diagnostic decisions.

Although BVI children are more likely to be diagnosed with ASD and developmental delays in general (Dale & Sonksen, 2002), it is not clear whether ASD and BVI are shared outcomes of a common etiology (e.g., genetic condition, maternal infection/exposure, neonatal brain injury) and/or whether visual deficits contribute to the presentation of ASD symptoms directly. Nonetheless, BVI can most certainly impact how behavioral symptoms manifest, which must be considered within the context of differential diagnosis.

Considerations for Differential Diagnosis of ASD in BVI Children

Early visual experiences play a large role in brain development and lack of visual input can disrupt the course of typical social development and interfere with processes necessary for communication and sensory development (Chokron et al., 2020). Typically developing individuals with BVI demonstrate differences in social communication and often exhibit RRBs specifically associated with visual impairment, which are called blindisms (e.g., rocking, head bobbing, repetitive rubbing the eyes). The presence of social communication differences and RRBs among typically developing individuals with BVI elevates the risk of diagnostic overshadowing when evaluating for ASD. In order to avoid this, an individual’s social impairments and repetitive behaviors should exceed that which is expected among typically developing individuals with BVI. Furthermore, for those individuals who acquire BVI later in development, it is important to consider the age BVI presented in order to understand how language and social communication development might have been impacted and the contribution to the overall behavioral presentation. This often requires consultation with those who have expertise working with individuals with BVI, as current research comparing typically developing individuals with BVI to those with BVI and ASD is extremely limited.

Social communication development among typically developing BVI individuals is different because they have limited or no access to visual forms of nonverbal communication, such as facial expressions, gestures, and imitation, which impacts social development. In general, infants with BVI are less likely to respond with visual attention, pointing, or smiling, and may be more likely to avert eye gaze with caregivers. Infants with BVI initiate fewer interactions with caregivers altogether (Molinaro et al., 2020), but many do demonstrate some use of facial orientation when they engage (Rattray & Zeedyk, 2005). Specifically, most often infants with BVI initiate interactions through tactile and vocal modes of engagement (Rattray & Zeedyk, 2005). As typically developing young children with BVI mature into the toddler and preschool years, social communication differences (i.e., joint attention) continue to become more apparent, such as far less sharing of interests or experiences with caregivers (i.e., less than one-third relative to 80-90% of the sighted toddlers; Dale & Salt, 2008).

Joint attention includes triadic interactions that lead to shared focus of attention on an object or event (Bigelow, 2003), and is typically evaluated through observation of visual behaviors within the context of ASD assessment; however, this may need to be assessed differently for individuals with BVI. For example individuals with BVI may demonstrate more auditory or tactile means of joint attention (e.g., hand manipulation, body orientation, or performing an action with an object). Auditory and tactile forms of joint attention tend to emerge later in development for typically developing children with BVI than do the visual forms of joint attention in sighted typically developing children, which should be considered when evaluating for ASD in children with BVI (Bigelow, 2003). Furthermore, clinicians should be cautious not to falsely attribute hand manipulation for the purpose of joint attention as “use of hand as tool,” a behavior commonly observed in sighted children with ASD.

Among those with low vision and CVI, the frequency of visual aspects of joint attention is often dependent on variables such as contrast sensitivity (i.e., the ability to perceive differences between an object and its background; Alfaro, Morash, Lei, & Orel-Bixler, 2018), and the visual the density and complexity of visual information in the surrounding environment. Notably, typically developing children with BVI tend to interact more with familiar than novel objects compared to their typically development sighted peers (Lueck, Chen & Kekelis, 2010). As such, when evaluating children with BVI, it can be helpful to incorporate some of the child’s own toys in addition to examining exploration with some novel toys. This may provide a more reliable assessment of joint attention and overall play behavior.

Social connectedness in children with BVI is highly reliant on language development (Conti-Ramsden & Perez-Pereira, 1999); however, typically developing children with BVI may demonstrate delayed language skills and atypical language patterns. This is because impaired vision impacts the visual aspects of joint attention that are so critical for building vocabulary and understanding the nuances of language that are often discerned through a combination of hearing language used while simultaneously observing non-verbal behaviors (e.g., sarcasm, humor, etc.). As such, typically developing children with BVI are likely to acquire use of personal possessive pronouns later than their typically developing sighted peers (average age of 36.5 months versus 19 months; Sonksen & Dale, 2002), and may also use echolalia as a means of social connection (i.e., using specific learned phrases for certain contexts and activities). The use of repeated questioning is common among typically developing children with BVI, in order to compensate for reduced or lack of visual cues that convey additional information about what a speaker means or intends (Molinaro et al., 2020). It is also common for children with BVI to shift conversations toward self-directed topics and use extraneous language to describe objects or their interests (Vinter, Fernandes, Orlandi & Morgan, 2013). Similarly, typically developing individuals with BVI may talk disproportionately longer than a conversation partner, which can contribute to a perception of imbalance in interactions and present as circumscribed interests (Behl, Akers, Boyce, & Taylor, 1996).

Although self-stimulatory behaviors are common among sighted children with autism, typically developing children with BVI may engage in specific kinds of behaviors that are less common among children with ASD (i.e., blindisms). Specifically, children with BVI are more likely to engage in eye-poking, pressing, and rubbing, light gazing or staring as a large portion of demonstrated stereotyped behaviors and these behaviors may decrease with age (e.g., Molloy & Rowe, 2011). Motor stereotypies are also common in individuals with BVI including head/body rocking, jumping, and repetitive hand movements. Repetitive play is more frequently observed, such as repetitive movements with objects (e.g., Hobson, Lee, & Hobson, 2009). Therefore, clinically it may be more helpful to focus on deficits in social communication in children with BVI than the presence of RRBs when evaluating for ASD concerns in order to reduce diagnostic overshadowing.

Table 1 provides a summary of clinical guidance for evaluating the presence or absence of ASD symptoms in BVI children informed by the clinical expertise of the author group and research when available.

Table 1.

Considerations in determining the presence of autism spectrum disorder symptoms in BVI populations

Symptoms of ASD	Typically Developing BVI Child	Presentation of Symptoms of ASD in BVI Children	Special considerations
Deficits in social-emotional reciprocity	• Orients toward a speaker by turning body, leaning in, touching • Echolalia can be common, but used to foster social interactions • May experience delays in development of theory of mind and joint attention, but these skills tend to develop • Use of personal pronouns can be delayed (36 months versus 19 months among sighted children; Sonksen & Dale, 2002)	• Consider if child’s social emotional reciprocity and delays in language acquisition and initiation are beyond what would be expected for child with vision loss (consider intervention history and amount of visual input) • Difficulty with reciprocal conversations and interactions • Atypical social approach • Reduced sharing of affect/interests/enjoyment in social interactions • Does not respond to name or culturally appropriate attention-getting measures (e.g. tapping, touching) • Difficulty understanding other’s needs or processing verbal or behavioral emotional cues	• Consider that social reciprocity and back and forth conversation may be impacted by the language skill of the BVI child • Consider that several causes of congenital blindness can be associated with other neurological comorbidities that can contribute to delayed development • Consider the familiarity of the environment (e.g., a BVI child may spend additional time exploring a new environment, which can contribute to reduced reciprocity, especially during an initial meeting)
Deficits in nonverbal communicative behaviors used for social interactions	• Joint attention may develop later among typically-developing BVI infants (Baron-Cohen, 1997) • Infants may be less likely to respond with visual attention, pointing, or smiling, and may be more likely to turn away during interactions with caregivers (Bigelow, 2003; Molinaro et al., 2020)	• Lack of using behavioral strategies to supplement interactions (e.g. orienting self toward a conversation partner, placing a hand on a partner)	• Behaviors involving vision are not appropriate to consider, but BVI children may demonstrate compensatory strategies to supplement interactions (e.g. orienting self toward a conversation partner, placing a hand on a partner)
Deficits in developing, maintaining and understanding relationships	• Typical social relationships can develop, maintain, and understand relationships appropriate to developmental level • May show social interest through asking who is present in a room • Indicate excitement or pleasure in response to social interactions	• May be a failure to initiate or respond to peers • Limited social curiosity (e.g., will not ask about people present or absent in a room) • Peer interactions may be inappropriate • May show interest in cause and effect toys and delayed acquisition of symbolic play skills inconsistent with nonverbal IQ	• These difficulties are usually not demonstrated in typically developing children who are BVI with intact nonverbal IQ
Stereotyped or repetitive motor movements, use of objects, or speech	• Echolalia can be common as a means to participate in social exchanges • Repetitive play can be observed, such as moving objects repeatedly, but can be redirected • Rocking and eye pressing/poking can be common, but able to be redirected	• May engage in repetitive play that is not within the visual domain, especially in the auditory domain • Repetitive play that does not include a social partner • Rocking, flapping, spinning can be common and difficult to redirect	• Repetitive speech and echolalia may be can be reflective of receptive language deficits rather than symptoms of ASD
Insistence on sameness, inflexible adherence in routines, or ritualized behavior	• Show varied play • Show flexibility • Can transition without major difficulty	• May be inflexible with respect to nonfunctional routines • Resistance to change/transition	• Changes to their physical environment can contribute to increased orientation/mobility challenges (e.g. rearranging furniture)
Highly restricted, fixated interests that are abnormal in intensity or focus	• May veer conversation topics to self-directed interests, particularly if conversation topics are related to visual input • Able to shift from preferred interests without significant difficulty	• Show highly specific interests that are atypical in intensity or topic • Have difficulty shifting from preferred interests	• May ask questions about upcoming events or changes to routine to orient selves to these changes
Hyper or hyporeactivity to sensory input or unusual interest in sensory aspects of the environment	• Light gazing can be common, but often decreases over time • Eye poking, pressing, and rubbing can be common	• Tend to demonstrate sensory seeking behaviors in the visual domain which do not decrease with age • Sensory processing differences may be present in domains not related to vision (e.g., rubbing textures)	• Monitoring and evaluation by an ophthalmologist is important to monitor vision input over time to ascertain if sensory seeking behaviors are related to changes in visual input, rather than a sign of ASD

Note. Autism Spectrum Disorder (ASD); Blind/visual impairment = BVI; Intellectual Quotient = IQ

Screening and Diagnostic Tools

A lack of appropriate tools and overlap between behaviors associated with BVI and ASD contribute to delays in the initial diagnosis of ASD among those with BVI (i.e., 79 months versus 56 months for children without visual impairment; Kancherla, Braun & Yeargin-Allsop, 2013). Tools designed for the assessment of young children with ASD rely heavily on visual input (e.g., pictures and toys) and nonverbal communication (e.g., eye contact), which contributes to difficulty in assessing children with BVI. A review of standardized tools used for screening and assessment of ASD concluded that none of the diagnostic tools included norms for children with sensory impairments and only one screening tool included such norms (de Vaan, Vervloed, Boom, Antonissen, Knoors, & Verhoeven, L., 2016), the Autism Behavior Checklist (ABC; Krug, Arick, & Almond, 1978). However, these norms were published over 40 years ago, with some estimates of variable reliability (e.g., high interrater reliability, but low split-half reliability). Additionally, ASD diagnostic criteria have changed significantly since the time the measure was developed, which further limits current relevance (de Vann et al., 2016).

Some tools have been developed or modified to aid in assessment of ASD among individuals with BVI, but many of these modifications apply to large age ranges and negate developmental considerations. The Observation of Autism in People with Sensory and Intellectual Disabilities (OASID) was validated on 60 individuals ages 6 to 55 years (17 of which were under 18 years of age; de Vaan, Vervloed, Peters-Scheffer, van Gent, Knoors, & Verhoeven, 2018), which limits the utility of the tool for early childhood assessment. Some researchers adjust existing tools to increase applicability for children with BVI. Fazzi and colleagues (2007) used a modified version of the Childhood Autism Rating Scale (CARS), omitting item VII about visual responsiveness to screen ASD symptoms among children aged 2 to 11 years with Leber’s congenital amaurosis. After diagnostic evaluation, the authors did not find that anyone in the sample demonstrated symptoms consistent with ASD. The authors concluded that the CARS may identify some behaviors that are associated with blindness (e.g., eye rubbing), but overall ratings were not consistently clinically elevated, particularly with respect to social interactions. As such, the CARS could be a useful tool to aid in screening of autism among children with this particular visual condition (Fazzi et al., 2007).

Other studies have systematically modified the Autism Diagnostic Observation Schedule, Second Edition (ADOS-2; Lord et al., 2012) and the Autism Diagnostic Interview, Revised (Rutter, Le Couteur, Lord, 2003) for children (ages 5 to 9 years) with BVI (Williams et al., 2014). Williams and colleagues made specific modifications to both tools, including to materials used and questions asked. Strong agreement between clinical diagnosis and diagnostic classifications on both the modified ADOS and ADI-R and lower agreement with the ADI-R standard algorithm that considers behaviors that occurred when a child was aged 4-5 years were observed (Williams et al., 2014). Some behaviors that were strongly associated with clinical diagnosis included response to name, shared enjoyment in an interaction, frequency and quality of social overtures, whereas behaviors such as socially-directed facial expressions, pointing, and stereotyped phrases were poorly associated with clinical diagnosis (Williams et al., 2014).

Deafness & Hearing Loss (DHH)

Hearing loss or deafness affects more than 48.1 million individuals in the United States, with nearly 1 in 5 Americans displaying hearing loss as defined by the World Health Organization criteria (Lin, Niparko & Ferrucci, 2011), and 14.9% of US children are affected by hearing loss or deafness (Niskar et al., 1998). There are different forms of hearing loss that include conductive hearing loss, sensorineural hearing loss and mixed hearing loss. Sensorineural hearing loss happens after damage to the inner ear or auditory nerve. Conductive hearing loss is associated with concerns with the middle ear structure that includes a smaller ear canal, punctured ear drums, or blockage due to chronic ear infections/excessive cerumen. Mixed hearing loss is associated with damage to both the outer and inner ear. Hearing loss or deafness can occur from a variety of etiologies and unknown causes. Approximately 50 to 60% of cases of childhood hearing loss are caused by genetics, including non-syndromal genetic causes (Connexin 26 and Connexin 30). Some cases of hearing loss are accounted for by maternal infections during pregnancy (cytomegalovirus), child exposure to ototoxic drugs (certain antibiotics and chemotherapeutic agents), otitis media with effusion, and congenital anatomical malformations (atresia, ossicular fusion). Syndromal genetic causes, such as Waardenburg syndrome and CHARGE syndrome (Morton & Nance, 2006), account for approximately 20% of genetic causes of childhood hearing loss.

The different etiologies and varying associated risks of additional disability require a sophisticated working knowledge related to the science of hearing loss or deafness for the provider working with this population. A population-based study of children with hearing loss or deafness reported additional neurodevelopmental problems were present in 30% (Van Naarden, Decoufle & Caldwell, 1999). The early environment can complicate the developmental picture, as children with hearing loss within the context of an absence of early auditory stimulation and delays in the acquisition of language have been found to have poorer neurocognitive outcomes on measures of auditory and visual working memory, attention, and inhibition (Pisoni et al., 2008).

Prevalence of Co-Morbid ASD and DHH & Etiological Considerations

A recent systematic review and meta-analysis of 7 studies conducted from 1994 to 2016 indicated an overall ASD prevalence rate of 9% in DHH populations (Do et al., 2017); however, children with profound hearing loss are more likely to have a co-morbid diagnosis of ASD compared to children with milder hearing loss (Szymanski et al., 2012). Prevalence of ASD is also associated with hearing loss/deafness etiology. For example, children with Fragile X syndrome are likely to have a higher rate of ASD when compared to the general population (Miller et al., 2010). Animal models of Fragile X syndrome have implicated that hyperexcitability and an imbalance of inhibition and excitation at the level of the brainstem are associated with the difficulties in sound localization often seen in individuals with Fragile X syndrome (McCullagh, Salcedo, Huntsman & Klug, 2017). CHARGE syndrome, which includes hearing loss as part of the diagnostic picture, is also associated with increased rates of ASD (Harthshorne, et al., 2005).

Considerations for Differential Diagnosis of ASD in DHH Children

When evaluating children who are DHH for ASD, differences in social communication must be considered to reduce diagnostic overshadowing. It is important to gather information about a child’s history of sound access (e.g., profound hearing loss since birth versus recent onset of conductive hearing loss secondary to a middle-ear infection) and language access and intervention (e.g. whether through spoken language, signed language, or augmentative and alternative communication), as such factors may contribute to any deficits in social-emotional reciprocity, nonverbal communication skills, and difficulties initiating, maintaining, and understanding social relationships. Notably, over 90% of children who are DHH are born to hearing parents, which often leads to obstacles in early communication with caregivers and sometimes results in significant delays in language and social development secondary to adequate access to language and communication. Within the DHH population, is important to consider if the delays in language acquisition are above and beyond the delays that may be expected of a DHH child based on their hearing loss, access to communication, and intervention history. For example, a child who has not had consistent access to language through aided hearing (e.g., hearing aids, cochlear implants) or American Sign Language (ASL) may be expected to have more profound language deficits than a child who has had early access to communication, such as a DHH child who has been exposed to ASL since birth.

There are some hallmark features of ASD and symptoms that can be helpful when differentiating ASD in DHH children. Atypical language features can emerge in DHH children with ASD who use sign language, such as echolalia in sign language (Shield, Cooley, and Meier, 2017) and signed palm reversals, where the palm of the hand faces toward the child rather than their communication partner, suggestive of early perspective-taking difficulties (Shield and Meier, 2012). Nonverbal communication is an important factor in language access and production in typically developing young DHH children, who often reference their communication partner to support their understanding through signed or spoken language. Since typically developing DHH children often show intact eye contact, joint attention, and pointing, it is important to screen for atypical development in these skills within this population (Szarkowski, Wiley, Yoshinaga-Itano & Mood, 2015). Denmark, Atkinson, Campbell & Swettenham (2019) found that deaf children with ASD produced fewer facial expressions and reduced quality of facial expressions compared to those produced by typically developing deaf children. When evaluating DHH children who use ASL, direct communication and assessment by an individual fluent in ASL should be considered best practice; when using an interpreter, the child may reference the interpreter rather than the examiner, showing reduced eye contact or facial referencing with the examiner.

Whereas more research is needed in the area of restricted and repetitive patterns of behavior in DHH children with ASD, clinical experience suggests that these symptoms may serve as a hallmark feature of ASD that is not regularly seen in typically developing DHH children. Again, it is important to be thoughtful about how a child’s language access and development may impact repetitive patterns of speech. DHH children with aided hearing (e.g., hearing aids, cochlear implants) should also be evaluated and monitored by an audiologist to ensure that their devices are working properly and not contributing to increased sensitivities to auditory information that may overlap with sensory sensitivities seen in ASD. In clinical experience, DHH children with ASD have been observed to sometimes have difficulty tolerating hearing aids and cochlear implants due to their ASD-related sensory sensitivities.

Table 2 provides a summary of clinical guidance for evaluating the presence or absence of ASD symptoms in DHH children informed by the clinical expertise of the author group and research when available.

Table 2.

Considerations in determining the presence of autism spectrum disorder symptoms in DHH populations

Symptoms of ASD	Typically Developing DHH Child	Presentation of Symptoms of ASD in DHH Children	Special Considerations
Deficits in social-emotional reciprocity	• Appropriate social smile • Appropriate eye contact • Engages with others with integrated eye contact, gestures, vocalizations • Can imitate behaviors, vocalizations, and/or signs • Display appropriate joint attention	• Consider if child’s social emotional reciprocity and delays in language acquisition and initiation are beyond what would be expected for child with hearing loss (consider intervention history) • Difficulty with reciprocal conversations (in primary language modality) • Atypical social approach • Reduced sharing of affect/interests/enjoyment in social interactions • Does not respond to name or culturally appropriate attention-getting measures (e.g. waving, flashing lights) • Difficulty understanding other’s needs or processing signed emotional cues • Pronoun reversal (Shield, 2012, Shield 2017)	• Consider that social reciprocity and back and forth conversation may be impacted by the language skill of the DHH child • Consider if the language modality of the child matches his/her communication partner (e.g. reduced reciprocity is expected if a child who uses ASL is an environment with no communication partners that use ASL).
Deficits in nonverbal communicative behaviors used for social interactions	• Typically have good eye contact, reference faces for communication purposes, show intact joint attention	• Poor eye contact, lack of pointing, poor joint attention	• Nonverbal communication is typically a strength for individuals who are DHH, as eye contact and facial expressions are important for communication • When using an interpreter, child may make more eye contact with the interpreter
Deficits in developing, maintaining and understanding relationships	• When provided an accessible means of communication (e.g. ASL, hearing aids), typically developing DHH children can show ability to develop, maintain, and understand relationships	• May be a failure to initiate or respond to peers when communication is taken into account (e.g. primary communication modality, communication partner’s modality) • May see a failure to recognize Deaf cultural norms • May show reduced shared enjoyment • Difficulty making and sustaining friendships even when communication is accessible • Delayed acquisition of symbolic play skills inconsistent with nonverbal IQ	• These difficulties are usually not demonstrated in typically developing children who are DHH with well-established communication and intact nonverbal IQ
Stereotyped or repetitive motor movements, use of objects, or speech	• May use more gestures and classifiers if exposed to and educated in ASL	• Children who are DHH and sign may engage in echolalia through sign, palm rotation errors (facing palm toward themselves rather than others; Shield, 2012; Shield, 2017) • Idiosyncratic gestures and use of made up gestures despite formal sign being taught (different from home sign) • Rocking, twirling, flapping, spinning	• Repetitive speech may be impacted by a child’s language development and can be reflective of receptive language deficits rather than symptoms of ASD
Insistence on sameness, inflexible adherence in routines, or ritualized behavior	• Show varied play • Show flexibility • Can transition without major difficulty	• Highly repetitive play • Resistance to change/transition	• Children who are DHH may present with some difficulty with change if they have language deficits that impact their understanding of what is happening/going to happen
Highly restricted, fixated interests that are abnormal in intensity or focus	• Show varied interests • Can shift from preferred interests without significant difficulty	• Show highly specific interests that are atypical in intensity or topic Have difficulty shifting from preferred interests
Hyper or hyporeactivity to sensory input or unusual interest in sensory aspects of the environment		• May show sensitivity to sounds or resistance to wearing hearing aids/cochlear implants	• Monitoring and evaluation by an audiologist is important to make sure hearing aids or cochlear implants are working properly and not creating sensitivity that is not indicative of ASD

Note. American Sign Language = ASL; Autism Spectrum Disorder (ASD); Deaf and hard-of-hearing = DHH; Intellectual Quotient = IQ

Screening and Diagnostic Tools

Currently, there are no specifically designed instruments to detect/diagnose ASD within the DHH population (Thompson & Yoshinaga-Itano, 2014). With regard to screening, the Social Communication Questionnaire (SCQ; Rutter M., Bailey A., Lord C., et al.) is often used when working with DHH children at Kennedy Krieger Institute; however, as Szarkowski and colleagues (2014) emphasized, it is important to use this measure with caution, especially with families that use ASL as their primary mode of communication. As such, even those who do not show risk for ASD on this tool are still evaluated critically for symptoms of ASD. With regard to diagnostic tools, Mood and Shield (2014) found that the ADOS-2 was more accurate in classifying DHH children with ASD who communicate using ASL when standardized rules for module selection, administration, and scoring were modified (see Mood & Shield, 2014 for specific modifications and justifications); however, research examining the utility of these modifications is limited. Considering the dearth of ASD screening and diagnostic tools appropriate for use in children who are DHH, Szarkowski and colleagues (2014) emphasize the importance of consultation and collaboration across specialists with expertise in ASD and DHH, as well as with individuals who serve the child, in making diagnostic decisions related to ASD in a child who is DHH.

Discussion

It is important that neuropsychologists are aware of behaviors commonly observed in children with sensory impairment in order to reduce the potential for diagnostic overshadowing. In the absence of established clinical guidelines, we present clinical considerations for neuropsychologists assessing risk/making a differential diagnosis of ASD in children with vision impairment or hearing loss/deafness. From a social perspective, both BVI and DHH children demonstrate unique patterns of social communication that must be appreciated when making diagnostic decisions. Furthermore, while RRBs do not generally present in typically developing individuals with DHH and the presence should be considered a red flag, certain types of RRBs are common in children with BVI. As such, careful consideration of the etiology of RRBs in children with BVI is important and presence of RRBs may be less helpful than social communication symptoms when distinguishing the presence of ASD in this population. Here we offer additional general considerations for practice that can and should be applied when considering a diagnosis of ASD in children with BVI or DHH.

General Practice Recommendations

Consider Symptoms in Relation to the Impact of the Sensory Impairment and Co-Morbid Neurological Conditions on Development and Behavior to Avoid Diagnostic Overshadowing

Neuropsychologists with expertise in early development are trained to consider the typical developmental trajectory and co-morbid behaviors that characterize many neurological conditions (Karba, Lemay & McLeod, 2020). Given that BVI and DHH are conditions can be congenital or acquired and are often associated with neurological conditions that impact brain development, a neuropsychological understanding may be particularly helpful when considering the presence or absence of ASD in children with sensory impairment. When children experience multiple conditions and disabilities, as is often the case in children with sensory impairment, the timing and severity of these co-morbid conditions and interactions between the conditions also must be considered in order to avoid diagnostic overshadowing.

Consider Neurological Stability

Children with sensory impairment within the context of complex neurological histories may experience changes to their medical status that can contribute to changes in cognitive and behavioral functioning. When determining whether a diagnosis is appropriate, there should be reasonable certainty that the child’s symptoms of ASD remain clinically significant despite changes in neurological status. If the child’s symptoms are expected to be ameliorated with a medical intervention, the presentation is likely not consistent with ASD. This can be particularly challenging in the differential diagnosis of ASD in very young children as a timely and accurate diagnosis is vital. One possible option in this case is to provide a provisional diagnosis of ASD in order to promote access to early intervention in the case that the child has ASD. This reduces the risk of missing the critical opportunity for early intervention should ASD be a persistent diagnosis; however, this approach should be used carefully and necessitates intentional conversation with the family and medical team to continue with medical workup as appropriate.

Know the Limitations of Diagnostic Tools

Use of diagnostic tools that have been examined in or developed for BVI and DHH individuals is ideal; however, given the dearth of tools validated in special neurological populations (e.g., those with sensory impairment, motor impairment, substantial attention problems), this may not be possible. In this case, planned and thoughtful modification of administration and/or interpretation may be warranted, coupled with careful behavioral observation and clinical expertise. Sometimes qualitative information gathered though behavioral observation may be most informative and quantitative assessment tools may have little utility. In these situations, expertise with the special population of interest or to consultation with someone with such expertise is critical. Furthermore, utilizing a multi-method and multi-informant (i.e., caregiver, teacher, therapist) approach will likely be most helpful when well validated tools are limited.

Furthermore, best practice for the evaluation of ASD includes a valid assessment of cognitive functioning (Hyman, Levy, & Myers, 2020) given that symptoms present differently based on developmental level. Determining general cognitive level may be particularly difficult in working with BVI and DHH children given that appropriate cognitive measures are limited, especially in the earliest years of development. Options may be even more limited for children with broader neurological complexity who may have significant motor deficits. Despite these limitations, clinicians are encouraged to determine the most appropriate method in estimating general developmental level for use in diagnostic decision making, which may or may not include standardized methods (see Hill-Briggs, Dial, Morere, & Joyce, 2007, Lueck, Dutton, & Chokron, 2019, and Nicholas, 2020 for considerations).

Re-Evaluation Throughout Development

Although much research has explored the behavioral markers of ASD in idiopathic ASD, less is known about the early behavioral markers of ASD within the context of sensory impairments, and certainly within the context of sensory impairments and complex neurological conditions (Roberts, Bradshaw, Will, Hogan, McQuillin, & Hills, 2020). As such, ongoing follow-up in children with sensory impairment is vital, whether or not a diagnosis has been made. This is especially important for those children with co-morbid cognitive impairment, where problems with social communication may become more obvious as social demands increase with development. Furthermore, while the developmental trajectory of ASD symptoms in BVI and DHH populations has not been systematically studied, research in children without sensory impairment has shown there is a subset of children with ASD who present with subtle symptom presentation early who are not diagnosed until early school age (Davidovitch, Levit-Binnun, Golan, & Manning-Courtney, 2015; Ozonoff et al., 2019). While this has not been systematically studied in BVI/DHH populations, it seems reasonable to hypothesize that children with sensory impairment who present with more subtle symptoms early on, would be even more difficult to detect early in development lending additional support for ongoing assessment.

Future Directions

As medical advances are made, and more children survive life-threatening and complex neurological conditions, research is needed to improve ASD diagnostic accuracy to meet the needs of children with sensory impairment. Thus, the development of new assessment tools or adaptation and validation of existing tools in concert with ongoing training for clinicians working with the BVI and DHH populations is necessary. Beyond diagnosis, increased precision in the evaluation and monitoring of ASD symptoms within these groups will foster clinical trial readiness and inclusion in clinical trials from which they may receive benefit. Therefore, the ongoing development of assessment tools and clinical expertise in clinical populations with sensory impairment is an urgent public health venture.