David H. Marshall

Prevalence of permanent childhood hearing impairment in the United Kingdom and implications for universal neonatal hearing screening: questionnaire based ascertainment study.

Abstract Objective: To estimate the prevalence of confirmed permanent childhood hearing impairment and its profile across age and degree of impairment in the United Kingdom. Design: Retrospective total ascertainment through sources in the health and education sectors by postal questionnaire. Setting: Hospital based otology and audiology departments, community health clinics, education services for hearing impaired children. Participants: Children born from 1980 to 1995, resident in United Kingdom in 1998, with permanent childhood hearing impairment (hearing level in the better ear >40 dB averaged over 0.5, 1, 2, and 4 kHz). Main outcome measures: Numbers of cases with date of birth and severity of impairment converted to prevalences for each annual birth cohort (cases/1000 live births) and adjusted for underascertainment. Results: 26 000 notifications ascertained 17 160 individual children. Prevalence rose from 0.91 (95% confidence interval 0.85 to 0.98) for 3 year olds to 1.65 (1.62 to 1.68) for children aged 9-16 years. Adjustment for underascertainment increased estimates to 1.07 (1.03 to 1.12) and 2.05 (2.02 to 2.08). Comparison with previous studies showed that prevalence increases with age, rather than declining with year of birth. Conclusions: Prevalence of confirmed permanent childhood hearing impairment increases until the age of 9 years to a level higher than previously estimated. Relative to current yields of universal neonatal hearing screening in the United Kingdom, which are close to 1/1000 live births, 50-90% more children are diagnosed with permanent childhood hearing impairment by the age of 9 years. Paediatric audiology services must have the capacity to achieve early identification and confirmation of these additional cases. What is already known on this topic The prevalence of confirmed permanent childhood hearing impairment (>40 dB HL) in the United Kingdom has been estimated to rise with age to 1.33/1000 live births among children aged 5 years and older It has been predicted that only an additional 16% of children will remain to be detected in the postnatal years, given current yields from universal neonatal hearing screening What this study adds The prevalence of confirmed permanent childhood hearing impairment (>40 dB HL) in the United Kingdom has risen with age to at least 1.65/1000 live births (and may be as high as 2.05/1000 live births) among children 9 years of age and older If the current yield from screening is sustained, then an additional 50-90% of children will remain to be detected in the postnatal years

Effects of simulated reverberation on the use of binaural cues and fundamental-frequency differences for separating concurrent vowels

A computational simulation was used to generate impulse responses between points in a rectangular room and two points on opposite sides of a spherical “head”. Sounds were convolved with the impulse responses to generate stimuli with which to study the effects of reverberation on the ability of listeners to use differences in fundamental frequency (Δ Fos) to separate concurrent vowels. Experiment 1 verified the suitability of the simulation by showing that it produced (i) appropriate percepts of lateralization, (ii) a larger contribution to lateralization from interaural differences in timing than level, and (iii) no effects of reverberation on lateralization. Experiments 2–5 measured masked identification thresholds for synthetic harmonic “target” vowels in the presence of masking sounds. In Experiment 2, listeners identified targets against pink-noise maskers. The experiment established a spatial geometry and a degree of reverberation for which listeners did not benefit from binaural cues arising from the spatial geometry of the sources. Experiment 3 demonstrated that the same arrangement did not undermine the ability to use Δ Fos to separate targets from vowel-like maskers when both had static Fo contours, but did prevent listeners from using Δ Fos carried on coherently changing Fo contours. Experiment 4 showed that a modulation width of ±1.45% was sufficient to reduce the benefits of Δ Fos, but that the benefits were not eliminated until the width of modulation exceeded the Δ Fo. It is argued that these results are compatible with existing models of the ability to use Δ Fos to separate concurrent vowels and that reverberation undermines the ability when the Fos are changing by diffusing the periodicities of the competing sources. Finally, Experiment 5 demonstrated that reverberation had no effect on the ability to separate a modulated vowel from pink noise. Thus, reverberation may have its detrimental effects in these experiments by diffusing the periodicity of the masking sounds rather than the targets. Overall, the experiments demonstrate that Δ Fos can be more robust cues for separating concurrent sounds than binaural cues. The relevance of these results to the perception of natural continuous speech is discussed.

Dichotic pitches as illusions of binaural unmasking. II. The Fourcin pitch and the dichotic repetition pitch.

The predictions of three models are compared with respect to existing experimental data on the perception of the Fourcin pitch (FP) and the dichotic repetition pitch (DRP). Each model generates a central spectrum (CS), which is examined for peaks at frequencies consistent with the perceived pitches. A modified equalization-cancellation (mE-C) model of binaural unmasking [Culling and Summerfield, J. Acoust. Soc. Am. 98, 785-797 (1995)] generates a CS which reflects the degree of interaural decorrelation present in each frequency channel. This model accounts for the perceived frequencies of FPs, but produces no output for DRP stimuli. A restricted equalization-cancellation (rE-C) model [Bilsen and Goldstein, J. Acoust. Soc. Am. 55, 292-296 (1974)] sums the time-varying excitation in corresponding frequency channels, without equalization, to form a CS. A central activity pattern (CAP) model [Raatgever and Bilsen, J. Acoust. Soc. Am. 80, 429-441 (1986)] generates a CS by scanning an interaural cross-correlation matrix across frequency. The rE-C and CAP models yield inaccurate predictions of the perceived frequencies of FPs, but predict the occurrence of the DRP and its correct pitch. The complementary predictions of the mE-C model compared to the rE-C and CAP models, together with the evidence that the FP is clearly audible for the majority of listeners, while the DRP is faintly heard by a minority of listeners, suggest that the mE-C model provides the best available account of the FP, and that the DRP is produced by a separate mechanism.

Paediatric cochlear implantation and health-technology assessment

Cochlear implants are provided to children on the basis of the hypothesis that short-term outcomes in auditory receptive skills will translate via a cascade of medium-term outcomes into greater social independence and quality of life. The medium-term outcomes include: (i) enhanced engagement and integration in primary education, leading to greater scholastic achievement; (ii) enhanced social versatility and robustness, permitting a successful transition to secondary education; and (iii) enhanced educational qualifications, allowing greater opportunities in further education and employment. A sufficient number of children have used implants for long enough for it to be feasible to establish whether the first two medium-term outcomes are being achieved and, if so, at what cost in the provision of health care and education. The first part of this paper discusses alternative research designs that could address these issues. Although a prospective randomised controlled trial would provide the most powerful evidence for or against the hypothesis, it is implausible that adequate compliance with randomisation to treatments could be sustained to give such a study sufficient power. The most powerful realisable design would be a large-scale cross-sectional comparison of implanted children and matched groups of their non-implanted peers. The second part of the paper describes the results of a speculative cost-benefit analysis that seeks to identify the cost to society of providing implants to children. The analysis is based on measured costs of health care, but on estimates of costs and cost-savings in other domains. It indicates that paediatric implantation could be cost-neutral in the UK, provided that implantation saved pound sterling 3000/year in the cost of education, pound sterling 1000/year in other domains, and permitted an increase in personal income of 25% of the national median household income. These savings might be realised if implantation permitted sufficient facility in spoken language to allow every implanted child to enter mainstream education.

Health-service costs of pediatric cochlear implantation: multi-center analysis

OBJECTIVE Pediatric cochlear implantation (CI) entails surgery followed by lifetime maintenance, and hence incurs both initial and ongoing costs. Previous assessments of these costs were either undertaken early in the evolution of services, or were based on single hospitals, or estimated costs largely from hospital charges. The aim was to overcome these limitations by conducting a multi-center evaluation of the costs of providing unilateral CI to children in the United Kingdom (UK). METHODS Annual numbers of implantations in all UK pediatric CI programs were monitored prospectively from 1991. Resource use was measured in 12 programs in 1998/1999 and retrospectively back to the year of inception of each program. The profile of outpatient and outreach visits was assessed in the 12 programs. Together these variables were used to estimate health-service costs for four phases of management: pre-operative assessment, implantation, tuning, and subsequent maintenance, using economic micro-costing methods. Costs were subsequently estimated for all children implanted in 1998/1999 (N=199) and were aggregated over 1, 15, and 73 years following implantation. To assess the robustness of cost estimates, parameter values were varied over plausible ranges and costs re-estimated. Total UK health-service costs were also estimated. All costs are presented in euros (1=US dollars 0.98= pound 0.65, 1st July 2002), inflated to 2000/2001 financial-year levels, and discounted at 6% per annum. RESULTS Per-child average costs were 42972 (1-year), 73763 (15-years), and 95034 (73-years). Cost estimates were not overly sensitive to the value of any one cost component nor to the relative cost of outpatient and outreach visits. When these parameters were varied, costs ranged between 30000 and 47000 (1-year), 61000 and 83000 (15-years), and 82000 and 108000 (73-years). The total UK health-service cost of unilateral pediatric CI was estimated to be 14 million in 2000/2001 and is predicted to rise to 23 million in 2015/2016, if the present model of service-delivery continues. The cost of maintaining implanted children was estimated to account for 22% of the total in 2000/2001 and is predicted to rise to 63% by 2015/2016. CONCLUSIONS Ongoing costs of maintaining implanted children and their implant systems are significant and should be factored into resource-allocation decisions.

Lip-reading the BKB sentence lists: Corrections for list and practice effects

Two groups of 21 adult subjects with normal hearing viewed the video recordings of the Bamford-Kowal-Bench standard sentence lists issued by the EPI Group in 1986. Each subject viewed all of the 21 lists and attempted to write down the words contained in each sentence. One group lip-read the lists with no sound (the LR:alone condition). The other group also heard a sequence of acoustic pulses which were synchronized to the moments when the talkers vocal folds closed (the LR&Lx condition). Performance was assessed both by loose (KW(L)) and by tight (KW(T)) keyword scoring methods. Both scoring methods produced the same pattern of results: performance was better in the LR&Lx condition; performance in both conditions improved linearly with the logarithm of the list presentation order number; subjects who produced higher overall scores also improved more with experience of the lists. The data were described well by a logistic regression model which provided a formula which can be used to compensate for practice effects and for differences in difficulty between lists. Two simpler, but less accurate, methods for compensating for variation in inter-list difficulty are also described. A figure is provided which can be used to assess the significance of the difference between a pair of scores obtained from a single subject in any pair of presentation conditions.

Dichotic pitches as illusions of binaural unmasking.

Dichotic pitches are binaural perceptual phenomena which occur when broadband noise with specific interaural phase/time relationships is presented simultaneously to the two ears. Four dichotic pitches have been described: Huggins’ pitch; the binaural edge pitch; the Fourcin pitch; the dichotic repetition pitch (DRP). This paper shows that the perceived pitch and timbre of the first three dichotic pitches are correctly predicted using a model of binaural unmasking [Culling and Summerfield, J. Acoust. Soc. Am. 98, 785–797 (1995)]. The model detects across‐frequency variations in interaural correlation; the resulting central spectrum shows peaks which correspond to the frequencies listeners hear. These pitches are probably, therefore, illusions produced by the mechanism of binaural unmasking. The DRP is not predicted by the model. The DRP could be produced through binaural cross talk, but experiments using insert earphones which produce minimal cross‐talk confirmed that the phenomenon persists when cross talk is effectively eliminated. The DRP is probably, therefore, produced by a separate mechanism from the other dichotic pitches.

Non-use of cochlear implants by post-lingually deafened adults

Abstract Objectives To document the occurrence of medical/surgical complications in the provision of cochlear implants to a cohort of post-lingually deafened adult patients; to compare cumulative mortality in the cohort with cumulative mortality predicted from an aged-matched control population; to relate the occurrence of medical/surgical complications to the patients position in the case series; to identify risk factors for elective non-use of implants; and to examine the impact of complications, mortality and elective non-use on the cost of creating a successful user of a cochlear implant Design Longitudinal prospective observational study. Setting Nine hospitals in the UK, which received central funding between 1990 and 1994 to participate in a programme of adult implantation. Participants All 313 post-lingually deafened adults who received multichannel cochlear implants in this programme. Main outcome measures Status of patients in July 1998 (dead, an elective non-user or a user); duration of use prior to death or elective non-use; costs of creating and sustaining a successful user; ability to identify environmental sounds; self-reported benefit. Results Cumulative mortality after 7.5 years was 6.3% (95% CI 2.5−10.1%) and was no different from cumulative mortality predicted from life expectancy in the general population. Cumulative elective non-use was stable at 6.3% (95% CI 3.6%−9.1%) between 4 and 7 years after implantation but rose to 11.0% (95% CI 1.7%−20.3%) at 7.5 years after implantation. Major medical/surgical complications were experienced by 25/313 patients (8.0%). The occurrence of such complications declined to ca. 1% over the first 50 cases in a hospitals case series. For patients who were users, self-reported benefit did not differ between patients who had experienced complications and those who had not. Older age at implantation, greater duration of deafness prior to implantation or an early position in a hospitals case series were not significant risk factors for elective non-use. Experiencing a major complication was a marginally significant risk factor (odds ratio = 3.2, 95% CI 1.0−10.6). Low auditory performance (odds ratio = 8.2, 95% CI 2.1−31.9) and low self-reported benefit (odds ratio = 19.6, 95% CI 4.6−84.4) were significant risk factors. Medical/surgical complications raised the cost of creating a successful user by less than 1%. Elective non-use raised the cost by between 5% and 10%. Conclusions Medical/surgical complications were generally managed successfully without detriment to benefit; life expectancy in the general population can be used to predict the length of time for which patients have the potential to use cochlear implants; patients who experience little benefit are most at risk for elective non-use; given the high cost of cochlear implantation, benefit should be monitored to allow early rehabilitative intervention in cases of low benefit; patient cohorts such as the present one should continue to be followed up with a view to identifying pre-operative risk factors for elective non-use.

Early Detection of Hearing Impairment: What Role Is There for Behavioural Methods in the Neonatal Period?

A survey of the use of behavioural methods for neonatal hearing screening in 1985 (1) concluded that the future for automated methods was quite promising. Since then several studies have assessed the two main automated behavioural tests: the Auditory Response Cradle (ARC) and the Crib-o-Gram (COG). As a screen targeted at neonatal intensive care unit (NICU) babies and other high risk groups (at present the most cost-effective form of neonatal hearing screening), the ARC is shown to have low sensitivity, even for severe hearing impairments, and the COG has an unacceptably low specificity. Any future for behavioural testing during this period must therefore rely on new implementations flowing out of a fundamental understanding of (a) the way in which neonates respond to sound and (b) the ways in which a behavioural test might complement screening with Auditory Brainstem Responses (ABR) or Evoked Oto-acoustic Emissions (EOAE). A clearer understanding of the relative benefits of detecting different degrees of hearing impairment at birth in both the NICU population and the unrestricted population is urgently needed. To determine what role should be played by specific screening programmes such benefits need to be balanced against the total costs of screening assessment and rehabilitation, in which false positives (low specificity) play a large part.

Prediction of hearing thresholds in children using an automated toy discrimination test

A previous paper described the development of the prototype of a semi-automated, sensitive and accurate version of the McCormick Toy Discrimination Test. In this report we describe a further development of the hardware, and demonstrate that results obtained from the automated test provide a basis for estimating the mean elevation of pure-tone threshold in the childs better ear. The correlation between speech and pure-tone results is high. The average of the better ear pure-tone thresholds at 0.5, 1 and 4 kHz can be predicted from the word-discrimination threshold obtained with the toy test, with a 95% confidence interval of +/- 11 dB.