Aetiologies and Outcomes in Children with Comorbid Autism Spectrum Disorder and Severe-Profound Hearing Impairment

Background: Autism Spectrum Disorder is a high incidence disability found in approximately 5% of hearing impaired children, with common aetiological pathways. Previous policies advised against cochlear implantation of children with these co-conditions.  There is little published outcome information as evidence base for treatment expectations and parent counselling.

Objectives: To study aetiologies and outcomes of children with co-morbid severe-profound hearing impairment and Autism Spectrum Disorder followed from 1999 to 2010.

Methods: Children with severe-profound hearing loss who received cochlear implants were enrolled through the Royal Victorian Eye and Ear Hospital programme following multidisciplinary diagnosis of Autism Spectrum Disorder.  Assessment used DSM-IV criteria and specialist instruments, the ADOS and ADI-R. All 19 children had non-verbal cognitive assessments, and 17 had post-implant speech and language outcome data. Tests were selected appropriate to each child, mostly WPPSI-3/Griffiths-R and PLS-4.

Results: The children’s mean age of cochlear implantation was 31.7 months (range 9-101), and mean age of Autism diagnosis was 50.8 months (range 24-105). Hearing impairment aetiologies were known for 14 children: very preterm (1), Cytomegalovirus (3), Connexin 26 (4), nonspecific genetic/familial (3), and vestibular aqueduct syndrome (3). Of five children with unknown aetiologies, two had abnormal MRI brain imaging. Communication outcomes were associated with cognitive ability. The five children with cognitive ability in/above Average (within/above 1 standard deviation of the mean) had speech and language outcomes Average and above. All 12 children with below Average cognitive scores, including 9 with intellectual disability, had communication outcomes in the Disability range (greater than minus 2 standard deviations below the mean), with 10 Moderate-Severe (greater than minus 3 standard deviations). Children were taught communication modes including speech, sign and picture exchange. All children received early intervention for hearing impairment, and 12 also received early intervention for developmental delay and/or Autism.  At school-age, children with normal cognitive ability (within 2 standard deviations of the mean) were in inclusive mainstream education.  

Conclusions: Universal infant screening has meant that hearing impairment is normally diagnosed before symptoms of Autism Spectrum Disorder are recognized. As children normally first enter services for hearing impairment and continue in shared care, professionals in the hearing impairment field need training in screening and interventions appropriate to children with these co-conditions. There was no dominant aetiology for hearing impairment in the study group. Communication outcomes were encouraging. They showed that implanted children with severe-profound hearing loss can develop functional oral language, provided their cognitive ability is within the normal range for age.  Even more encouraging was the finding that children with Average and above Average cognitive ability can develop formal oral language appropriate to ability. Children with ability within the normal range can be accommodated within mainstream education. Further study of the profiles, treatments and outcomes of these children is needed in order to provide an adequate evidence base for treatment expectations and parent counselling.