Neonatal Hearing Screening: Have We Taken the Right Road? Results from a 10-year Targeted Screen Longitudinally Followed Up in a Single District

Abstract

10-year cohort born in the East London District of Redbridge from 1990 to 2000, was enrolled into a targeted neonatal hearing screen (TNHS). Selection was based on the risk factor criteria of the American Joint Committee on Infant Hearing. The screen was typical of most other selective neonatal screens that were being undertaken in the UK in the 1990s. The cohort was followed up until the children had entered primary school. From the cohort of 32,890 neonates, 1148 (3.5%) were considered at risk of deafness and of these, 1054 received a neonatal Auditory Brainstem Response (ABR) air conduction test. There was a yield of 17 (5.2/10,000) with a moderate or worse bilateral permanent childhood hearing impairment (PCHI). By primary school entry the prevalence of moderate or worse PCHI had risen to 15.2/10,000 with 12.7/10,000 having a congenital deafness. Although 40% had been identified by the neonatal screen it could have been optimized to give a yield of 6.7/10,000. The potential results were compared to the model for targeted screens constructed by the US Preventative Services Task Force (USPS TF). There were important differences between the actual and constructed models. These differences are further discussed in the paper. The East London screen was more efficient but even the optimized screen was less effective than the estimates provided by the hypothetical model, with only half the predicted number of cases being confirmed before the model's target age of 10 months. The USPS TF model has been used to measure the potential added value of Universal Neonatal Hearing Screens (UNHS). However, it does not reflect the selective screens that have been undertaken in the UK, and substantiating evidence will not be available from this source. The conclusion from the reported longitudinal cohort study is that the hypothetical model overestimates the worth of targeted screening and by implication underestimates the potential added value of universal screens. Without a universal newborn hearing screen early detection of congenital deafness is not achievable for the majority of deaf and hearing impaired children.