Customize your JAMA Network experience by selecting one or more topics from the list below.
Donaldson AI, Heavner KS, Zwolan TA. Measuring Progress in Children With Autism Spectrum Disorder Who Have Cochlear Implants. Arch Otolaryngol Head Neck Surg. 2004;130(5):666–671. doi:10.1001/archotol.130.5.666
Copyright 2004 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2004
To quantify progress after cochlear implantation for children with autism spectrum disorder (ASD).
Retrospective review of speech and language and speech perception test scores of children with autism who have received a cochlear implant at our center.
University of Michigan Medical Center, Cochlear Implant Program.
Six children, ages 3 to 16 years, who received cochlear implants at the our center. All children were diagnosed as having ASD by a neuropsychologist, either before or after receiving a cochlear implant.
Main Outcome Measures
Children participated in preoperative and postoperative speech and language and speech perception testing. A survey was administered to parents to evaluate subjective impressions of cochlear implant benefit and quality of life before and after implantation.
Improved scores were recorded for children on whom standardized expressive and receptive vocabulary testing was possible. Children who could not complete standardized tests demonstrated improvement in raw scores. Improvement on the Meaningful Auditory Integration Scale was noted for the 4 of 7 children who completed the scale preoperatively and postoperatively. Survey results suggested changes in responsiveness to sound, interest in music, vocalization, and eye contact following implantation. Five of the 6 families indicated that they would recommend a cochlear implant to another family in a similar situation.
Gains made by children in our study were small compared with the general implant population; however, when compared with themselves preoperatively, these children did demonstrate progress. Improvements in behaviors and interaction point to a quality of life benefit following implantation that is difficult to quantify.
Our program has noted a recent increase in the number of pediatric cochlear implant candidates and recipients who have a diagnosis of autism spectrum disorder (ASD). Possible reasons for this increase include implantation of children at younger ages, less strict candidacy criteria, and an increase in the prevalence of autism in young children.1-3 In the United States, recent reports estimate prevalence to be between 3.4 and 6.7 cases per 1000 children.4,5 Studies from Europe and Australia have similarly suggested that the prevalence of autism in young children is on the rise.1-3 An epidemiologic review in Australia conducted in 1997 reported a 200% increase in diagnosis of ASD,3 citing an increase in awareness of the disorder as well as an improved ability to diagnose the disorder in young children as possible reasons for this increase.
The terms autism spectrum disorder (ASD) and pervasive developmental disorder (PDD) are used to describe a spectrum of related characteristics.6 There are 3 main characteristics of the disorder: impairments in social interaction, impairments in verbal and nonverbal communication, and restricted and repetitive patterns of behavior.7 Children with a diagnosis of ASD may vary widely in the severity and presentation of the disorder. A diagnosis of "autism" refers to a specific prototypical disorder, while a diagnosis of ASD or PDD may include diagnoses such as Asperger syndrome, Rett syndrome, or childhood disintegrative disorder. These other diagnoses, or a diagnosis of PDD "not otherwise specified," indicate disorders that share characteristics of autism, but may be different in severity, behaviors affected, or age at onset.6
The prevalence of ASD among children with hearing impairment is relatively unknown. Carvill8 suggested a higher prevalence of psychiatric disorders in the hearing-impaired population than in the normal-hearing population; however, an exact correlation of autism and hearing impairment was not documented. Jure et al9 reported that 4% of children at a school for deaf children met the criteria for a diagnosis of autism. Of their hearing-impaired population, 13% were hearing impaired due to maternal rubella, which is known to cause neurological disorders in addition to hearing impairment and may have accounted for the high number of children with ASD in their study.
To date, 475 children have received a cochlear implant at the University of Michigan. Of these children, 8 have been diagnosed as having ASD. Thus, the prevalence of ASD among our population is 1.7%. Among these 8 children, the severity of diagnosis varies from very mild to severe and includes children with a diagnosis of autism as well as PDD.
Only a few published studies have discussed the performance of children with ASD when using a cochlear implant. Waltzman et al10 reported results for 29 profoundly deaf children with documented impairments in addition to deafness who received cochlear implants. Two of the children from their sample were diagnosed as having ASD: one child had a diagnosis of PDD, and the other child had a diagnosis of autism. Speech perception scores were not available for the child with autism at any time interval. Scores were available for only the child with PDD 5 years after implant activation and indicated slightly improved speech perception scores. Similarly, Hamzavi et al11 described a group of multiply-handicapped children with cochlear implants, and included one child with a diagnosis of autism. This child demonstrated mild expressive speech gains 3 years after implantation. Although neither study specifically addressed the issue of cochlear implants in children with ASD, both concluded that the presence of additional disabilities does not always preclude a child from being considered for a cochlear implant.
Evaluating children with ASD for cochlear implant candidacy, and programming such patients postoperatively, presents the clinician with a unique set of challenges. In this study, we present results of speech perception and speech-language measures for 7 children who had a diagnosis of ASD and received a cochlear implant at our facility. Subjects included children diagnosed as having PDD and autism, both of which fall under the broader diagnosis of ASD. Additionally, a questionnaire was developed and administered to parents of these 7 children to survey their opinion regarding several issues related to their child's use of a cochlear implant and their diagnosis of ASD.
Subjects included 7 children with ASD who received cochlear implants at the University of Michigan between 1998 and 2003. Demographic information for the subjects is provided in Table 1. This study was approved through the University of Michigan's institutional review board, and written informed consent was obtained from the parents for all of the patients. Of the 7 children, 5 were diagnosed as having autism and 2 were diagnosed as having PDD by neuropsychologists. Four children were diagnosed before implantation, and 3 were diagnosed after implantation. The mean age at implantation and the mean age at ASD diagnosis for the children in this group were both 4.7 years. The average time of cochlear implant use was 25 months. Sound field audiograms obtained on all subjects postoperatively indicated a functioning cochlear implant device.
Speech perception tests reported for the children in this study include the Meaningful Auditory Integration Scale (MAIS),12 which uses an interview format to generate parental report of performance; the infant-toddler version of the MAIS (IT-MAIS),12 which is similar to the MAIS but for younger children; and the Glendonald Auditory Screening Procedure for words and sentences (GASP-W and GASP-S),13 which is an speech recognition test using familiar words and phrases that is given in an open-set format (ie, no choices are given to the child). Receptive vocabulary was measured using the Peabody Picture Vocabulary Test (PPVT-III),14 a test of listening comprehension and receptive vocabulary presented through spoken language in standard English. To measure expressive vocabulary, the Expressive Vocabulary Test15 was used, which is a norm-referenced assessment of expressive vocabulary and word retrieval for children and adults 2½ to 90 years of age. The MacArthur Communicative Development Inventory16 was also used, which uses parental report to assess early language skills, as well as play behaviors. The tests were administered whenever possible, although most children could not complete testing due to their communication abilities.
A questionnaire was given to parents of all 7 children. Topics surveyed included mode of communication, view of success with the implant, behaviors before and after implantation, and family interactions before and after implantation. Five parents completed the survey in the office and 2 parents completed the survey at home and returned it via mail. A copy of the questionnaire is available upon request from the authors.
Standardized measures of speech and language development were not possible in many cases due to the developmental level of the children. However, testing was performed whenever possible. Use of standard scores is not informative in some subjects since many of the children in our sample performed far below age expectations. Both standardized scores and raw data for the speech and language measures are reported in Table 2.
The MacArthur Communicative Development Inventory is used to assess comprehension of spoken words in the home based on parent report. This test was completed for 5 of the 7 children. Four of the 5 children demonstrated an increase in their comprehension of spoken words after implantation. Subject 6 demonstrated no comprehension of spoken words preoperatively or at the 12 months postactivation interval.
Two children participated in receptive vocabulary testing with the Peabody Picture Vocabulary Test. One child demonstrated little difference between the score obtained preoperatively and 6 months postoperatively (subject 1). The score obtained by subject 2 increased from a standard score of less than 40 preoperatively to a score of 72 five years postoperatively.
The Expressive Vocabulary Test was successfully administered to 3 of the 7 children. One child (subject 2) demonstrated strong improvements. The standard score obtained by subject 2 increased from 54 at the 24-month postimplant interval to 81 at the 60-month postimplant interval. The other 2 subjects did not show any difference in their standard scores, but both demonstrated increases in their raw scores.
Results obtained on the speech perception measures are displayed in Table 3. Many of the children in this study were unable to participate in tests routinely performed with other children in our clinic because of the severity of their developmental delay. The MAIS was administered to 5 of the 7 children. Comparison of preoperative and postoperative scores is only possible for 4 children. For these children, the average preoperative score was 42% and the average postoperative score was 72%. All 4 children demonstrated higher scores after implantation.
Administration of the Glendonald Auditory Screening Procedure for words and for sentences was only possible for 2 of the 7 children. Both children demonstrated improved scores following implantation on these 2 measures. Subject 2 achieved a score of 100% correct on both words and sentences 2 years after implantation. This child, whose diagnosis is mild PDD, demonstrated speech-language and speech perception scores similar to children who are implanted who do not have ASD.
Parents were asked to indicate which methods their child used to communicate with others. Responses are summarized in Table 4. Most parents (6/7) indicated that their child used gestures, while many (4/7) indicated use of word approximations or vocalizations. Three of 7 parents indicated their child uses sign language. Only 1 parent indicated that her child uses spoken language (subject 2).
Parents were asked to rate their child's behaviors before and after implantation. A rating of 1 represented "never" while a rating of 5 represented "always." Results are summarized in Table 5. The largest difference between preoperative and postoperative ratings was reported for enjoyment of music. The mean score for the group increased from 1.5 preoperatively to 3.8 postoperatively. Mean scores for the group increased more than 1 point for each of the following: reaction to sound, vocalization, eye contact, use of sign language, and response to requests. All 7 children were reported to vocalize, react to sound, and enjoy music at least sometimes following implantation.
Parental ratings for family interactions before and after implantation are summarized in Table 6. Minimal differences between the preoperative and postoperative conditions were reported. There were, however, some notable individual findings. When asked if others expressed an interest in the child's progress, the ratings of 2 children went from "sometimes" preoperatively to "always" postoperatively (subjects 1 and 3). Subject 5 was reported to have an increase of 2 points in compliance to family routine following implantation, but 2 others (subjects 2 and 4) were rated lower in their compliance to family routine following implantation. Both of the children who demonstrated a decrease in their compliance to family routine after receiving the implant were diagnosed as having ASD after implantation.
Table 7 summarizes parents' rating of several categories following implantation. Ratings were given on a scale of 1 to 10, with 1 representing "most impacted by implant use" and 10, "least impacted by implant use." Of the categories, communication was ranked the highest with a mean rank of 3.8. Great variability was noted for parents' ratings, suggesting different goals and outcomes for each individual child.
Lastly, parents were asked to rate their child's overall success with the cochlear implant and their willingness to recommend the implant to a family in a similar situation. Results are listed in Table 8. When asked about their child's overall success with the implant, parents' responses appear to be related to their child's time of diagnosis. All of the children who were diagnosed as having ASD after implantation reported that their child did worse than expected with the cochlear implant. Three of the 4 parents whose children were diagnosed with ASD before implantation indicated that their children were performing "as expected" or "better than expected," while 1 parent reported she had "no expectations." Five of the 7 parents reported they would recommend the cochlear implant to another family in a similar situation.
This study represents a preliminary quantification of success among a small group of pediatric cochlear implant users with ASD. The incidence of ASD in our population is 1.7%, suggesting that ASD among severe to profoundly hearing-impaired children may be more common than in the general population. Based on our sample, the age of diagnosis of ASD may also be affected by hearing impairment. The average age of diagnosis of ASD of the children in our sample was 4.7 years, later than 3.4 years reported by Bertrand et al4 in 2001.
Examination of speech and language and speech perception data reveals improvement for all subjects postoperatively on measures that could be completed. These data also underscore the need for more standardized measures of progress for children with ASD who have cochlear implants. Only a few standardized measures could be completed on a small number of children. With the exception of the MAIS, a missing test score was due to the subject's inability to complete the test. All measures that could be completed by the child were administered at each evaluation interval. The MAIS was not completed on all children primarily because of time constraints: time usually spent evaluating the child was instead spent counseling the parents, working with the child toward development of a con ditioned response to sound, and performing objective measures to aid in programming of the speech processor.
The results of this survey suggest that several behaviors may be affected when a child with ASD receives a cochlear implant. As demonstrated by the responses summarized in Table 5, these include increases in vocalization, eye contact, use of sign language, reaction to sound, and response to requests. One limitation of this survey is that it asked parents to rate behaviors retrospectively. To better examine changes in behavior, future surveys will be given to parents of children with ASD preoperatively and at set intervals following implantation.
Although most children who receive cochlear implants do so to facilitate spoken language, only 1 of the 7 children in our study (subject 2) uses spoken language to communicate. This child's diagnosis is mild PDD. It is believed that his family's commitment to oral education, his length of device use, and the mild form of his disorder have contributed to his success. He is a noted outlier, and his case does not necessarily suggest that all children with ASD can develop oral communication. Based on our experience, oral communication is not likely to be a realistic goal for implanted children who present with a diagnosis of ASD.
Despite the low incidence of spoken language after implantation in these children with ASD, most families surveyed indicated they would recommend the implant to other families in a similar situation. In the survey, parents reported positive benefits of implantation, which included changes in behavior, changes in communication, and increased awareness of the environment. The parents in our study reported increases in reaction to music and sound, vocalization, eye contact, use of sign language, and response to requests. These findings suggest areas that may be examined closer in studies that can prospectively analyze behavior before and after implantation. Goals and expectations for performance are different for each child with ASD who receives an implant and these should be discussed for each child before implantation.
In our sample, the cochlear implant did not appear to impact the behaviors most closely associated with ASD. Families reported little or no change regarding the child's interaction with siblings and other children, and regarding the child's compliance to family routine. It is important that this is discussed preoperatively with families when considering a cochlear implant for a child diagnosed as having ASD. Parents must realize that intervention of the child's deafness with a cochlear implant may have little or no effect on the diagnosis of ASD or its severity.
Unique challenges exist when a child is diagnosed as having ASD after receiving a cochlear implant. Expectations for performance are very different for children with ASD. Not surprisingly, parents of children who were diagnosed with ASD after implantation reported that their children had done worse than expected with the implant. Preoperative counseling, particularly when done for very young children, should include information about the possible impact of undiagnosed disabilities such as ASD on performance.
Overall, the results of this study suggest that cochlear implantation can be beneficial for children with ASD. Many areas of research remain, including a longitudinal study of specific behaviors of children with ASD before and after implantation as well as development and evaluation of objective tools to measure progress in children with ASD. Counseling parents to facilitate development of appropriate expectations and the use of a team of professionals is strongly recommended for this population.
Corresponding author and reprints: Amy Isaacs Donaldson, MA, 475 Marketplace, Bldg 1, Suite A, Ann Arbor, MI 48108 (e-mail: firstname.lastname@example.org).
Submitted for publication September 17, 2003; final revision received January 26, 2004; accepted January 28, 2004.
This study was presented at the Ninth Symposium on Cochlear Implants in Children; April 25, 2003; Washington, DC.