Deaf Children with Autism Spectrum Disorders
- First Online:
- Cite this article as:
- Szymanski, C.A., Brice, P.J., Lam, K.H. et al. J Autism Dev Disord (2012) 42: 2027. doi:10.1007/s10803-012-1452-9
- 1.9k Views
Epidemiological studies investigating the prevalence of autism have increased in recent years, within the United States and abroad. However, statistics as to how many of those children may also have a comorbid hearing loss is lacking. The prevalence of school-administrator reported diagnosis of autism spectrum disorders (clinical diagnosis [DSM-IV] and/or IDEA classification) among children with hearing loss in the US was estimated from the 2009–2010 Annual Survey of Deaf and Hard of Hearing Children and Youth conducted by the Gallaudet Research Institute. Results indicate that during the 2009–2010 school year 1 in 59 children (specifically 8-year olds) with hearing loss were also receiving services for autism; considerably higher, than reported national estimates of 1 in 91 (Kogan et al. in Pediatrics 124(4):1–8, 2009) and 1 in 110 (CDC 2007) for hearing children. Significantly more children with profound hearing loss had a comorbid diagnosis of autism than those with milder forms of hearing loss. These results are discussed, while highlighting the need for increased awareness and research in a population that has thus far received little services or attention.
KeywordsAutismHearing lossDeafAnnual Survey
During the 1970s blindness, deafness or “auditory deprivation” and ear infections were suggested as possible causes of autism (Hayes and Gordon 1977; Collins and Carney 2007; Smith et al. 1988). This belief was often rooted in older research that suggested deafness or hearing loss led to social isolation, emotional distress, psychological disorders, and language difficulties (Myklebust 1960). A possible relationship between autism and deafness was hypothesized due to striking similarities in children with rubella who often had a concomitant hearing loss and those with autism who did not have hearing loss (Chess 1977). Others have suggested that hearing loss may account for the language deficits and attentional challenges in children with Autism Spectrum Disorders (ASDs) (Smith et al. 1988).
Higher rates of autism have been reported to occur in children with hearing loss when compared to their typically developing hearing peers (e.g., Jure et al. 1991; Gordon 1991; Rosenhall et al. 1999) and their developmentally delayed peers (Taylor et al. 1982; Klin 1993; Tas et al. 2007). Rosenhall et al. (1999) conservatively estimated the prevalence of profound hearing loss in children with autism to be 10 times higher than the general public (0.1–0.2%) and Jure et al. (1991) found autism often to be overlooked in groups of children with hearing loss. Jure et al. (1991) also provided data that suggested a lack of any correlation between a child’s severity of hearing loss and the severity of autism. Instead they concluded that the severity of autism was linked to severity of cognitive deficit, not a child’s hearing loss (Jure et al. 1991).
Children with hearing loss are generally known to be at risk for additional disabilities. Research conducted by the Gallaudet Research Institute estimates that approximately 40% of deaf children had additional disabilities during the 2009–2010 school year. Often, research attributes the high degree of secondary disability in children with hearing loss to “neurological risk factors” associated with the cause of the child’s hearing loss (e.g., Marschark 1993). Neurological risk factors are considered to be medical conditions that are severe enough to cause a child’s deafness, and, due to the severity of the neurological condition, often cause additional disabilities for the child such as vision loss, intellectual disability, brain damage (Vernon 1969), or autistic-like characteristics (Marschark 1993; Marschark and Hauser 2008; Jure et al. 1991; Gordon 1991). Examples of conditions that have been linked to both severe hearing loss and autism (or characteristics consistent with an ASD) include rubella, cytomegalovirus (CMV), herpes, prematurity, toxoplasmosis, CHARGE, meningitis and measles (McCay and Rhodes 2009; Jure et al. 1991; Chess 1977).
Genetics continues to be implicated in both hearing loss (over 80% of children have a form of genetic hearing loss [Arnos and Pandya 2003]) and autism (Muhle et al. 2004), therefore making it difficult to suggest that deafness causes autism when both have their own documented genetic influences. Furthermore, to suggest that hearing loss causes autism would imply that a much larger number of deaf individuals should have autism, which has not yet been documented (Jure et al. 1991). Sensory deprivation, whether vision impairment or hearing loss, as causal of autism has been refuted in numerous studies (Zafeiriou et al. 2007). Instead, research shows that the severity of autistic symptomatology is often related to the severity of the cognitive impairment and not the severity of hearing loss (Rosenhall et al. 1999; Jure et al. 1991). This would make sense as research shows that hearing loss itself is not a risk factor for behavioral difficulties, but instead children who have hearing loss and poor language competence are at the most risk for displaying disruptive behavior (Stevenson et al. 2010). Poor language development has been documented in literature pertaining to cognitive impairments (Warren and Abbeduto 1992), behavioral disturbances (Donno 2010), difficulties with attention (Im-Bolter et al. 2006), socialization (Durkin and Conti-Ramsden 2007), as well as stereotypical behaviors (Edwards and Crocker 2008).
Intervention research for children who are deaf and have an ASD is only beginning to emerge. Educational services for deaf students with multiple disabilities are scarce, problematic, and not always appropriate (Guardino 2008; Bruce et al. 2008; Ewing and Jones 2003; McCay and Rhodes 2009). Documented behavioral interventions, educational strategies, and social skills courses for children who are deaf or hard of hearing and have autism do not yet exist. Jure et al. (1991) reported evidence of reading and overall academic deficits in children who are deaf and have autism. Rosenhall et al. (1999) comments that when both autism and deafness are present the communication and linguistic deficits are compounded, thus resulting in more severe problems. McCay and Rhodes (2009) also emphasize how communication is uniquely impacted for children with hearing loss or autism.
Cochlear Implants, which are commonly made available to deaf children without multiple disabilities, were discouraged for deaf children with multiple disabilities until 1992 (Edwards 2007). Research shows that cochlear implants in deaf children with an ASD may help the child recognize sound and produce vocalizations more often than without implantation (Dammeyer 2009; Edwards 2007). However, research also shows there is no noticeable improvement in characteristics of autism (e.g., social, behavioral, or communication) for the child (Dammeyer 2009; Edwards 2007).
Previously, only Jure et al. (1991) and Rosenhall et al. (1999) had conducted studies that specifically examined the prevalence of autism in children with hearing loss. Both studies consisted of small clinical samples, making inferences regarding national prevalence difficult. While they and others have hypothesized higher rates of autism in children with hearing loss, a national data sample to support such a hypothesis has lacked. Today, due to efforts by Gallaudet University’s Research Institute, demographic information pertaining to deaf children in the United States and its territories are collected annually using the Annual Survey of Deaf and Hard of Hearing Children and Youth (Annual Survey), allowing for a better understanding of prevalence rates of children with hearing loss and autism to be explained in this paper. Results from the 2009–2010 Annual Survey and some discussion of existing research with considerations for future research into the field of deaf children and autism follows.
Annual Survey of Deaf and Hard of Hearing Children and Youth
Initiated in 1968, the Annual Survey of Deaf and Hard of Hearing Children and Youth (Annual Survey) has been conducted nearly annually by Gallaudet University’s Research Institute. The Annual Survey is designed to identify characteristics of Deaf and Hard of Hearing Youth throughout the United States. Information pertaining to characteristics of individuals and schools, trends in educational progress, demographic information, and additional disabilities are collected and de-identified prior to analysis. Constructed as a national population-based survey of special and private schools as well as public school systems, the Annual Survey remains the “only national database on Deaf and Hard of Hearing Children and Youth in the United States” (Gallaudet Research Institute 2008). In 2004, autism was added to the Annual Survey as a possible “additional condition” indicating that a child was receiving services for both hearing loss and an ASD. The Annual Survey was not collected during the 2008–2009 school year.
School administrators and/or school psychologists who have access to a child’s educational file, and therefore the most up to date and most accurate information about the child voluntarily complete the Annual Survey for every child in their school or district who has a hearing loss. Other “additional conditions” as outlined on the Annual Survey are congruent with IDEA classification codes (e.g., visual impairments, blindness, traumatic brain injury, specific learning disability) and other childhood disorders such as Attention Deficit Hyperactivity Disorder (ADHD). Children with an autism code on the Annual Survey or any other “additional condition” may have either a clinical (DSM-IV) diagnosis or an IDEA classification. The Annual Survey does not distinguish the different ASDs. Instead, the clustering code of autism on the Annual Survey from Gallaudet University is meant to be interpreted as children who are both deaf and have a previous clinical or medical diagnosis of an ASD that guarantees school-based services.
Methodologically the reporting of children with comorbid hearing loss and autism in this study best mirrors data collection practices of Kogan et al. (2009). While Kogan et al. (2009) relied on parental report to determine the prevalence of autism, the current methodology relies on school district administrators and/or school psychologists to report relevant information about children who have a previous diagnosis of an ASD and comorbid hearing loss. Children who are no longer considered to have autism (as defined by individual schools) are not included in the following data analysis.
This method does not involve direct diagnostic testing as conducted by Chakrabarti and Fombonne (2001) and Fombonne (2005), or strict record reviews as conducted by the CDC (2007), but has been shown to be an acceptable method for estimating reported prevalence of ASDs. Therefore, reported estimates should be interpreted with caution until a larger national study can be completed using direct testing or extensive record review, providing more definitive prevalence estimates.
Characteristics from the Annual Survey of Deaf and Hard of Hearing Youth and Children for children with an ASD
Deaf children with an ASD
Total deaf students with an ASD
Parental hearing status
Both parents hearing
Both parents deaf or HoHa
One deaf or HoH; other hearing
One deaf or HoH; other unknown
One hearing; other unknown
One parent unknown or student adopted
Ethnicity (multiple responses were allowed)
Black or African-American
American Indian or Alaska-native
Multi-ethnic or other
Family members regularly sign in the home
Data not available
Cochelar implant still used for information
Summary of prevalence rates of ASD in children who are deaf or hard of hearing from the Annual Survey for Deaf and Hard of Hearing Youth (2004–2010)
Respondents in the Annual Survey (with total known information for additional relevant conditions)a
Children with autism classification
Overall prevalence rate
Prevalence rate for 8-year olds with autism and deafness classificationsb
Trend of Autism and Hearing loss, 2004–2010
Reported here are the prevalence rates of ASD in deaf and hard of hearing children for the years 2004–2010. During the 2009–2010 school year, the overall prevalence rate of ASD in deaf children as reported by schools was 1 in 53. When specifically examining 8-year olds with an ASD, prevalence rates were estimated to be 1 in 59.
Etiology of Deafness and ASD
Etiologies of deafness as reported in the 2009–2010 Annual Survey of Deaf and Hard of Hearing Youth
All children from the Annual Survey
N = 37,828
Children from the Annual Survey with an ASD
N = 611
Presence of Additional Conditions
Comparison of deaf children with ASD and without when considering the presence of additional conditions
Deaf children with an ASD
N = 611
Deaf children without an ASD
N = 31,723
Additional relevant conditions to a child’s hearing loss
Specific learning disability (LD)
Attention deficit disorder (ADD/ADHD)
Traumatic brain injury
Other health impairments
Other additional conditions
Total additional conditions to a child’s hearing loss
No additional conditions
One additional condition
Two additional conditions
Three additional conditions
Four additional conditions
Five additional conditions
Six additional conditions
Seven additional conditions
Severity of Hearing Loss and Presence of Autism
Intellectual Disability, Autism and Deafness
Historically, autism has been thought to co-occur with mental retardation or intellectual disability. Estimates of intellectual disabilities in children with autism has been reported to range from 15% (Gillberg 1998) to 70 or 80% (Fombonne 1999). According to the Annual Survey, a relatively small number of deaf children with an ASD also had a diagnosis of intellectual disability (15.5% or n = 95). However, of those children with both autism and intellectual disability, 42% had a profound hearing loss.
Educational placement information
Deaf children with an ASD
N = 611
Deaf children without an ASD
N = 31,723
Primary educational setting
School for the deaf
Self-contained class room in a mainstream school
Number of hours integrated with hearing peers
More than 26 h/week
Primary mode of communication in the school
Sign and speech (total communication)
Spoken language with cues
While the Annual Survey cannot be representative of every deaf child within the United States and its territories, it remains the largest national database available of children with hearing loss. Researchers at the Gallaudet Research Institute speculate that the Annual Survey consists of approximately two-thirds of Deaf and Hard of Hearing Children throughout the United States and its territories (S. Hoto, personal communication, 2010). The overall prevalence of school-administrator reported diagnosis of ASD among children with hearing loss from the 2009–2010 Annual Survey of Deaf and Hard of Hearing Children and Youth was 1 in 53, or more specifically prevalence estimates suggested 1 in 59, 8-year old children had a comorbid diagnosis of hearing loss and an ASD, considerably higher than current reported prevalence rates of autism for children without hearing loss.
The occurrence of intellectual disabilities in Deaf and Hard of Hearing Children with an ASD (15.5%) is lower than many estimates for hearing children with an ASD (Bryson and Smith 1998; Byrd 2002; Fombonne 1999). The existence of diagnostic overshadowing (Reiss et al. 1982) may provide some explanation for the disparity of reported comorbid ASD and intellectual disability in the current study. Deafness or hearing loss has implications for a child’s language development (Paul 2009), social development (Moeller 2007) and behavior (Stevenson et al. 2010; Easterbrooks and Handley 2006). Psycho-social characteristics of deafness may include delays in language acquisition and comprehension (Marschark 1993), theory of mind competence (Lundy 2002; Jackson 2001) and social knowledge (Suarez 2000). Similarly, psycho-social characteristics of autism often include delays in language acquisition (Landa 2007), difficulties with theory of mind tasks (Baron-Cohen 2000), and struggles to master social situations (Baron-Cohen 1988). Intellectual disabilities also have many of the exact same characteristics as previously mentioned. Therefore, clinicians may be inadvertently attributing some characteristics to one diagnosis when in fact they are evident of another, evidence of diagnostic overshadowing. Additionally, given the known difficulties in distinguishing children with profound intellectual disabilities from those with autism, the ability to make an accurate diagnosis of both in a child with hearing loss is further complicated.
Observation, parent report, and one-on-one play based interactions may be adequate for an initial diagnosis of autism. This allows a broad range of clinicians the ability to make a diagnosis of autism, even if they do not have expertise in the unique cultural and linguistic background that children with hearing loss bring to an assessment. However, when making a clinical diagnosis of intellectual disability, clinicians must posses a different skill set, one heavily rooted in knowledge of linguistic, cultural, and medical implications that a child’s hearing loss may have on their psychological and intellectual presentation. Due to the scarce number of clinicians with this knowledge and expertise, it becomes possible to suggest that children with hearing loss and autism may have intellectual disabilities that have not yet been diagnosed because of a lack of knowledgeable clinicians.
Data from the Annual Survey does suggest a disproportionate number of profoundly deaf children having a coexisting diagnosis of an ASD (35.4%) than those children with milder forms of hearing loss. This trend was also observed over time (see Fig. 2, Annual Survey results from 2004–2010). Results of statistical analysis also show that children with a reported diagnosis of an ASD had more severe forms of hearing loss than those without a reported diagnosis. However, given the dichotomous nature of other possible predictor variables for the relationship observed (e.g., presence of other disabilities such as vision loss, intellectual disability), one cannot control for such possible confounding variables. Nonetheless the existence of a significant difference between the number of children with autism having a profound hearing loss, than those with and without other disabilities, in addition to the trend observed over time, lead us to believe that a relationship may exist. Previous research samples (Jure et al. 1991; Rosenhall et al. 1999) may have been too small and therefore inadvertently overlooked the relationship observed in the study. Additionally, with the advent of new technology it is likely that more specific estimates of severity of hearing loss exist today than when previous research was conducted. The Annual Survey does not ask for severity of intellectual disability or severity of ASD, therefore there remains no way to determine if severity of intellectual disability is linked to diagnosis of autism as suggested by Jure et al. (1991).
The overall percentage of deaf children with an autism classification in the 2009–2010 Annual Survey (1.9%) remains lower than the percentage estimated in a clinical sample by Jure et al. (1991) of 5.3%. One explanation for the discrepancy may be the differences in samples selected. Rosenhall et al. (1999) sample consisted of children with a previous diagnosis of autism referred for an audiological evaluation, while Jure et al. (1991) performed chart reviews of children referred for an evaluation by a developmental pediatrician. Those samples consisted of children with a previous ASD diagnosis or other suspected disabilities, thus not a true representation of the entire population of children with hearing loss. The current study utilizing the Annual Survey provides a much larger normative population sample of deaf children with and without disabilities and likely is a more reliable estimate of the prevalence of autism in children with hearing loss, explaining the discrepancy found.
When considering the total sample of children within the Annual Survey (N = 37,828), more children with a diagnosis of autism (n = 611) had pregnancy related causes of hearing loss than those children without autism. Complications during pregnancy such as fever, CMV, and meningitis have been reported to be associated with hearing loss, as well as reported to be associated with characteristics of autism. This coupled with the finding that there were fewer reported genetic or familial cases of deafness in children with comorbid hearing loss and an ASD further supports previous research that neurological risk factors may best explain the comorbid existence of hearing loss and autism and not that of causal relationship of a sensory impairment on autism (Zafeiriou et al. 2007).
While Roper et al. (2003) and Garreau et al. (1984) report similar characteristics between deaf and hearing children with autism, what is not clear is to what severity those deficits must first be observed within a child with hearing loss to warrant such evidence of deficits in communication, socialization, and behavior that are necessary to satisfy diagnostic criteria for an autism diagnosis. Complexities of diagnostic practices for children with hearing loss can be further highlighted in the lack of utility of some of the current DSM-IV criteria needed to make an autism diagnosis. For example, DSM-IV TR criteria for Asperger’s disorder prohibits the existence of speech delays; unfortunately, many deaf children have delays in spoken language due to their hearing loss, suggesting limitations in making or even suggesting an Asperger’s disorder diagnosis in children with hearing loss. Additionally, the communication cluster of the DSM-IV is heavily weighted on the use of spoken language mastery to rule out autism, something that many typically developing deaf children do not have. This makes the cluster nearly impossible to utilize for children who are deaf and have a suspected autism diagnosis.
Furthermore if no differences between autistic behaviors exhibited by deaf or hearing children and youth exist (Roper et al. 2003; Garreau et al. 1984), making a diagnosis of autism in a child with hearing loss should be similar to making a diagnosis of autism in a hearing child. However, diagnostic and psychological measures used in making an autism diagnosis, such as the Autism Diagnostic Observation Schedule (ADOS), specifically state they should not be used in standardized ways with children who are deaf, often speculating that test items and understanding of characteristics will be different. This notion appears to directly contradict findings of Roper et al. (2003) and Garreau et al. (1984). Nonetheless, clinicians are currently left without any instruments that are suitable or helpful in making an autism diagnosis in children with hearing loss (Szymanski and Brice 2008). Without specific instruments, numerous overlapping characteristics of hearing loss and autism, and lack of expert clinicians who can adequately evaluate children with hearing loss, a differential diagnosis for children with comorbid hearing loss and suspected autism remains challenging. This difficulty is further compounded by the likely role of diagnostic overshadowing in making a diagnosis; be that autism masking hearing loss or intellectual disability, hearing loss masking autism or intellectual disability, or the presence of intellectual disability masking both.
Questions remain in the literature as to the best educational setting for a child who is deaf and has an ASD. In 1991, there were no schools in the US specifically designed for deaf autistic children (Jure et al. 1991). Since that time a few schools have slowly begun to admit children with ASDs as well as provide services. The National Deaf Academy in Florida has a program specifically designed for deaf children with autism, often with intensive behavioral challenges. Several other states (e.g., Vermont, CA, TX, PA) are initiating programs to address needs of these children, but still lack the resources and professionals knowledgeable in deafness and ASDs. A school for the deaf may provide the visual and language input that a child needs while lacking knowledge or resources for the best interventions for autism. A school or program designed for a child with autism will likely have a well-established program with several protocols ready to meet the needs of various students. Unfortunately, these schools have little to no experiences working with children who are deaf or understanding the language needs of the child.
In the current study, children with hearing loss and an ASD were more likely, than their peers without an ASD, to be educated in a schools or programs specifically designed for the deaf, as well as educated with exposure to sign language. Hearing children with autism, when exposed to sign language have shown improvements in communication, behavior, and socialization (Bonvillian and Nelson 1976; Bonvillian et al. 1981). A placement in a school for the deaf may provide a naturalistic sign language environment, thus prompting language, behavioral and socialization improvements which researchers such as Delprato (2001) suggest is most influential in language learning. Sign language communication strategies for children with autism may be good for encouraging early vocabulary learning as signs can be easily prompted and done “hand-over-hand” for a child who is struggling to produce language (Goodstein 2002). This may suggest that children with an ASD and hearing loss who are immersed in an environment with constant exposure to sign language (such as a school for the deaf) may also have better communication, behavior and socialization skills than their peers not educated in such an environment. Further research is needed before conclusions can be made.
While previous research (Roper et al. 2003; Garreau et al. 1984) stated that autism in children with and without hearing loss is similar, it still appears that one will face difficulties and challenges when making an autism diagnosis in children with hearing loss. The challenge in making a differential diagnosis should contribute to caution in interpretation of the current study that utilized school-based reporting to estimate prevalence of ASDs in children with hearing loss. Until a study that conducts a nationally based diagnostic study such as Chakrabarti and Fombonne (2001) and Fombonne (2005) or strict record reviews as conducted by the CDC (2007) an exact prevalence remains beyond the scope of this paper and unfortunately beyond the scope of the current state of clinician abilities. As previously discussed, the lack of appropriate psychological measures to diagnosis autism in children with hearing loss, a clear understanding of the presentation of autism in children with hearing loss, and the existence of experts who understand the linguistic and cultural impacts that hearing loss and sign language may have on making on an autism diagnosis contribute to this deficit.
This large national population based study of children with hearing loss allowed the researchers the opportunity to better estimate the prevalence of autism in children with hearing loss as well as begin to understand demographic characteristics of autism in children with hearing loss. In preparing for this research, we were alarmed as to the paucity of available literature pertaining to best practices for deaf children with autism in the areas of intervention, education, diagnosis, language acquisition, and general quality of life concerns. Concerns are further amplified given the high prevalence rate of comorbid hearing loss and autism in children reported in the Annual Survey. It is our hope that in the future, research can be established to help parents, providers, and educators better understand and work with children who are both deaf and have autism. Essential to future progress will be designing a psychological measure that is applicable to children who are both deaf and have autism (the ADOS states that it should not be used with children with hearing loss) and establishing an early intervention curriculum that incorporates both the needs of the child’s hearing loss and autism.