Statistics from Altmetric.com
I have read with great interest the original article from Wake and colleagues1 and I would like to acknowledge and compliment their valuable efforts in such a difficult research area. I felt nevertheless quite concerned with the conclusions of this study and their possible repercussions. Diagnosis and management of childhood deafness is one of my areas of interest and I have also been actively involved in the setting up of NHSP in my local district In the UK, the NHSP is in its final phase of implementation and hopefully there will be no going back. In other areas of the globe, however, where professionals may still be pondering about the importance and need of such a programme, outcomes of research studies like this one may help to tilt the balance in the wrong direction.
Research into deafness and especially childhood deafness is extremely difficult, a real minefield. Severe and profound deafness is relatively rare and the number of variables to take into consideration is huge: age of diagnosis, age of hearing aid fitting, consistent use of hearing aids, cochlear implant, age at start of other forms of intervention such as speech and language therapy, educational input (type of specialist intervention programmes, bilingual versus oral-only programmes), cognitive ability, parents’ hearing status, parents acceptance and cooperation with professionals...the list is enormous.
Only a study involving very large populations would allow for improved variable control and still achieve comparison samples large enough to be treated statistically. This would require huge human and financial resources and is usually beyond the possibility of most research centres.
The present study did attempt to control some of these variables, but the inclusion of hearing losses from mild to profound (or even hearing losses above 40 dB HL) may have skewed the results. Severity of hearing impairment is in itself such a stronger predictor of language outcome that it compensates for many other variables including age of diagnosis.
Deaf children with a hearing loss of around 60 or 70 dB HL, may, with consistent use of well fitted hearing aids, achieve enough amplification to be able to hear and discriminate spoken language, essential for spoken language acquisition. A profoundly deaf child with >90 dB HL loss or more will never be able to achieve that much. If comparisons between severe and profound deaf children already cause difficulties, what to say when moderate hearing losses are also included?
I believe this is one of the reasons why, in this study, age of diagnosis did not help to predict language outcome and therefore the conclusion that early diagnosis may not be an important factor in improving outcomes for deaf children may not be correct.
Other factors may also have influenced outcome in this particular study. There is very little information about intervention programmes and since children came from different areas and schools, these may be very different and have significant impact on progress. Also, there is no mention of use of sign language and I wonder if this is not used at all by the children in the sample or just spoken language progress was considered.
I would like to finish with a parent’s reply when asked how she felt at the time of her child’s late diagnosis (at 9 months of age): “We were too relieved. We should be upset or shocked but, having battled with someone for five months, it was just a relief that someone believed”. However, later on, she would say: “I was angry, I was very angry, I don’t know I will ever get over the anger”.