Cochlear Implants

What the research shows: After any residual hearing in an ear is destroyed during cochlear implant surgery, that implanted ear provides speech understanding, on average, like an ear with a 100-110dB hearing loss (at best, an equivalent of about 77dB hearing loss, at worst no speech perception). Speech intelligibility is also on average like that of a person with a 100-110dB hearing loss. Cochlear implants are therefore, by my reckoning, worth considering for children with hearing losses greater than 110dB. Powerful hearing aids seem to offer about as much speech/language benefit as implants to other profoundly hearing-impaired children.

Complications occur in between 1% (according to the first author in Chapter 9, who appears be ignoring his own data) and 10% (according to Article 9D) of implant patients, depending on what you call a complication. Prior to March 1998, data provided by two manufacturers presented indicate a 10.36% complication rate in Clarion implants in children, and a 14.68% complication rate in Nucleus implants in both children and adults. Complications included healing problems, dizziness, pain, facial nerve stimulation, extrusion of the device, migration of the electrode array and electrode array damage. 3% of adults and 4.7% of children with Clarion implants required re-implantation (equivalent Nucleus figures are not quoted). Article 9A reports 'no mortality, meningitis or facial palsies were registered in our series' (p176).

The critical period for language-learning appears to be drawing to a close at around age 6. This, I would suggest, allows time for 2-year-old children to try out hearing aids for an extended period, and still get an implant well within the critical period, if hearing aids provide no language benefit.

This book is dismissive of sign language, and many authors seem to be trying unconvincingly to substantiate the use of 'oral' education methods. Only one author (in chapter 14) refers to another text which discusses the body of research indicating that natural sign language (e.g. Auslan here in Australia, ASL in the US) actually assists deaf children with their oral language and literacy skill development.

Article 12B discusses a 1996 parent survey in the US and Canada, which was sent to 2626 parents of children with cochlear implants, with 1008 replies. 49.3 of the children still using their implants (9 children were not) were enrolled full-time in mainstream classes; 25% were in special classes for hearing-impaired children. 75.3% were in the same class level as other children their own age. The other 24.7% of children had school delays from 1-6 years, averaging 1.5 years. 53.1% of children used auditory/spoken language as their primary means of communication.

What the book says the research shows: A lot of the book's research into language outcomes of cochlear implants is poorly designed and controlled, uses questionable measures (e.g. speech perception tests done in an audiology suite, but no information about a child's actual communication skills at home or at preschool) and inflated 'improvement/growth/rate of learning' figures rather than children's actual test scores. These design flaws are generally glossed over or dismissed e.g. 'The usual confounding issues of device/strategy, age at implantation and length of deafness did exist' (p204). The large quantity of badly designed research proves very little, but there are a few delightful 'findings' such as 'children in both the early CI and late CI groups identified lexically easy words with significantly greater accuracy than lexically hard words (P<0.001) (p253) i.e. all the children found the easy words easier than the hard words; and 'Although children 2 to 5 years of age at the time of implantation showed the greatest improvements from pre- to post-implantation intervals, children 6 to 12 years of age generally had better spontaneous speech skills' (p296) i.e. preschool children learn language faster than school-aged children, but school-aged children are better at talking. Well, we certainly needed expensive scientific research to tell us that.

Regardless of their data, many authors make inflated claims for cochlear implants in their conclusions. For example, Article 14B is about expressive vocabulary development in children with cochlear implants. Children studied were aged 2-8 years. Before getting an implant, seven children used an average of 14 spoken words during videotaped play sessions. After their implants 6 months later, they used an average of 27.4 spoken words. This was not significantly different from the spoken words measured in a group of children without implants, and thus could have occurred because of normal development. Post-implant, children were still using signs with their speech attempts. From this, the authors come to the astonishing conclusion that 'cochlear implants accelerate vocabulary acquisition and its spontaneous expression in young deaf children' (p288). Articles of this type are the rule, rather than the exception, in the sections of this book discussing language research.

The book is also full of spelling, punctuation and grammar errors, and editing or proofreading mistakes, such as missing or wrongly labelled tables, and referencing errors. These made me wonder about the authors', editors' and publisher's general level of attention to detail and accuracy.

I would recommend this book to any family considering a cochlear implant for their young deaf child. They should take it to someone who knows about language research design and statistics, and ask that person to explain what the data actually mean. Then compare this with the claims many authors are making. They will find some very interesting contrasts.
December 19, 2000