From Public Health and Epidemiology Report Ontario, Volume 13, #9

Ontario's Infant Hearing and Communication Development Program

SOURCE
Dr. Martyn L. Hyde, Ph.D.
Director of Research and Development, Otologic Function Unit, Mount Sinai Hospital, Toronto
Professor of Otolaryngology and Public Health Sciences, University of Toronto

CONTACT

Marlene Stein, BA, DSPA, S-LP(C), Reg. CASLPO
Program Consultant Speech Language Pathology and Audiology
Population Health Service
Public Health Branch, MOHLTC

Background

Significant, permanent childhood hearing impairment (PCHI) is one of the most common of the major, congenital disorders¹. It is more prevalent than all the hemoglobinopathies and errors of newborn metabolism combined². PCHI is not in itself life threatening, but it has significant consequences for the individual, the family and society at large, especially if it is not detected early³. Despite its clear importance, population-based screening for PCHI has not been widely implemented until recently. This omission has been a serious public health deficiency in any society professing concern with optimal development, quality of life, and educational achievement of its children ^4,5.

There are many, diverse sequelae of undetected hearing impairment in early infancy. An infant with significant hearing impairment cannot hear the vast array of environmental sounds, and especially cannot hear speech, song, or any other part of the normal acoustical exchange between parent and child. This is a significant disadvantage in and of itself, but it also has deleterious outcomes. For example, there is clear evidence from animal and human studies that auditory deprivation in early infancy can lead to significant changes in the structural and functional organization of the auditory system, up to and including the cerebral cortex^6,7. It is also clearly established that hearing impairment that is not addressed promptly can lead to delays in the acquisition of speech and language^3,8. Further impact in areas such as cognitive development, socialization and later, academic achievement have been reported⁹.

In the absence of systematic screening of newborns for PCHI, the patterns of detection, diagnosis and provision of hearing and communication development services are cause for great concern. A large-sample study from Ontario¹⁰ revealed that mean age of diagnosis in unscreened children referred by physicians because of concern about hearing was 2.8 years in the period 1991-95. This is contrasted with 5.7 months for a screened group of children at risk. For children with moderate impairment and no risk factors, the mean age was about 5 years! It is very probable that most of the undetected impairments were present since birth.

The reasons for this poor performance are many. First, it is impossible for parents or health professionals to detect impaired hearing reliably in the first year of infancy by casual observation or by informal, behavioural tests. Indeed, even computerized behavioural tests with controlled stimuli in the hands of hearing experts have proved to have poor sensitivity or specificity¹¹. Second, in the young infant, PCHI may be confused with, or overlaid by, transient impairment associated with middle-ear disorders. Until recently, reliable distinction between transient and permanent components of impairment has been problematic. Third, there is much variation in normal development of overt response to sound, as well as in early vocalization. Apparent delay in milestone achievement, even if suspected by the family, all to often continues to elicit the notorious 'don't worry, (s)he'll grow out of it' response from health professionals. Indeed, there is evidence that inappropriate professional response has long been a significant source of identification delay¹².

Over the last 15 years, the situation has changed dramatically as a result of technological advances, especially in the detection and quantification of hearing impairment in the newborn and young infant. It is now possible to screen newborns and young infants quickly, accurately and objectively, using computerized, portable test equipment operated by personnel with limited training. It is also possible to quantify hearing status in detail, both accurately and comprehensively, due to advances in diagnostic assessment procedures. In short, detection and quantification of hearing impairment are no longer an obstacle to early initiation of effective services for maximizing hearing ability and/or initiation of communication development strategies.

These developments have prompted major efforts in several countries to improve hearing and communication healthcare for newborns and infants. While screening of infants at high risk for impairment has been advocated for many years, universal newborn hearing screening (UNHS) programs have proliferated in the last five years, especially in the US. Concurrently, many agencies and organizations have endorsed universal newborn hearing screening (UNHS). In compliance with the well-established principle that population screening should not be undertaken in the absence of accessible and effective services to address the needs of those children and families identified by screening, the most common, current terminology for the overall system of care is 'Early Hearing Detection and Intervention' or EHDI programs. Bodies that have endorsed UNHS/EHDI include the interdisciplinary US Joint Committee on Infant Hearing (JCIH)³, the American Academy of Pediatrics (AAP)¹, consensus development conferences from the US National Institutes of Health¹³ and the European Economic Community (EEC)¹⁴, as well as from several major audiology associations. A large program has been initiated in 20 UK health districts and currently, UNHS is legislated or otherwise mandated in at least 40 US states; over 60% of newborns in the USA are currently screened, usually prior to discharge from the hospital birth admission¹⁵.

There is a directive overtone to the use of the word 'intervention' that is not consonant with the important theme of family-centered health services, so the EHDI term is not considered appropriate in the Ontario context.

A Health Canada national survey in 1999 revealed only a patchwork of local initiatives in newborn hearing screening, covering only 25% of newborns and mainly directed at high-risk groups¹⁶. However, in May 2000, Ontario took a national leadership position in announcing funding for a major new initiative: the Infant Hearing and Communication Development Program. Before describing the key elements of this important advance, the basic epidemiology and related issues underlying the design of a high-quality program will be outlined.

Prevalence of Permanent Childhood Hearing Impairment

The prevalence of PCHI in the newborn and infant is now understood reasonably well, due to regional and national ascertainment studies in the UK¹⁷, as well as to large UNHS programs that have been in place for several years in the US^18,19. The prevalence is a function of many variables relating to the definition of the target disorder, especially the severity, frequency range, type and laterality of the impairment (one or both ears). In summary, the prevalence of congenital impairment of 'moderate' degree or worse is about one per thousand live births. As a reference point, such an impairment is more disabling than that which a hearing adult would perceive when wearing typical earplugs and earmuffs combined. For so-called 'mild' impairment, comparable to that experienced with earmuffs alone, the prevalence doubles, and if unilateral impairment is included in the target, it increases further by about 50 to 100%. This yields a prevalence of about 3-4/1000 for at least mild impairment in one or both ears. It is important to detect unilateral impairment for several reasons. For example, a child with PCHI in one ear may acquire a major hearing disability due to otitis media in the previously normal ear, or may be at risk for progression to bilateral PCHI.

PCHI prevalence is about 8-10 times greater in at-risk newborns than in those without any risk indicators. About 15% of newborns may have a known risk indicator. The most common indicators are neonatal intensive care for more than 48 hours, familial PCHI, and major or minor craniofacial anomalies. However, about 40-50% of newborns with PCHI manifest no known risk indicator, and therein lies part of the rationale for universal newborn screening, as opposed to targeted screening of only high-risk newborns^{20, 21}.

Screening Tests

It is well-established that casual observation of response to sound or informal behavioural tests of hearing sensitivity are worse than useless in newborns and young infants. The two major, innovative screening technologies in current use are automated otoacoustic emissions (AOAE) and automated auditory brainstem response (AABR)³. OAEs are minute sounds that are actively generated in the normal inner ear (cochlea), in response to external sound stimuli. The OAEs radiate out to the external ear canal and may be detected by a miniature microphone in the canal. Any hearing impairment of moderate degree or greater will abolish the OAE, thus yielding a binary screening test ('pass' or 'refer' result). However, the test is not a perfect indicator of hearing: some infants with hearing impairment of mild degree will yield an OAE, as will the small proportion who have impairment arising from dysfunction at a more central location in the auditory system than the site of OAE generation in the cochlea. Conversely, some infants have normal hearing but their OAEs are abolished by minor middle-ear conditions. Furthermore, it is not possible to estimate hearing threshold sensitivity quantitatively with OAEs, which are essentially either present or absent for a range of acoustic stimuli. By their nature, OAEs yield a reasonably accurate binary test for hearing impairment of mild-to-moderate degree or greater.

ABRs are minute, gross (whole-nerve) neuronal action potentials that are generated in response to external sounds. They originate in the neurons of the afferent auditory pathways in the brainstem. They can be recorded using EEG electrodes on the head, after computer extraction from the spontaneous neural activity of the brain. Automated response detection algorithms render the test completely objective. ABR presence is equated with auditory perception, thus yielding a binary screening test of hearing sensitivity. The stimulus is usually a rapid series of clicks, presented at a low intensity level. In contrast to the OAEs, ABRs can yield fairly accurate, quantitative estimates of hearing thresholds. Thus, AABR screening is a simplified type of ABR testing, wherein a single stimulus level is chosen such that failure to record an ABR implies presence of the target disorder.

Both AOAE and AABR screening are straightforward. AOAE screening is done with a hand-held device and takes 3-5 minutes. ABR screening is done with a laptop computer and hand-held attachment, and takes typically 5-10 minutes. Both screens require a reasonably quiet environment, and best results are obtained in a sleeping baby. Because of its greater instrumentation expense and slightly greater skill requirements for testers, ABR screening is commonly reserved for high-risk newborns, or as a second screen in babies who refer from OAE screening.

Initial screening is usually done during the perinatal hospital admission. It is best done just before discharge home. If it is done within 12 hours of birth, the false-positive rate increases significantly, mainly because of fluid in the middle ear space and possible debris in the external meatus. AABR screening is less affected by minor middle-ear conditions, but its false-positive rate is likely to increase if it is done in newborns under about 34 weeks gestational age.

AABR screening is generally preferred for infants at risk for hearing impairment, because the higher prevalence in that group justifies a test with very high sensitivity, despite increased expense. Also, AABR screening can detect auditory neuropathy (AN), which is a recently-recognised cluster of disorders that are thought to originate in the cochlear inner hair cells or the cochlear nerve²². OAE screening cannot detect AN, which may comprise 5-10% of all cases of PCHI in early infancy. Most babies with AN are graduates of neonatal intensive care units.

Screening Test Performance

For at least moderate hearing impairment, pre-discharge AOAE screening is believed to have a sensitivity of about 80% and a specificity of about 90%; predischarge AABR screening has a sensitivity of at least 85% and specificity of 90-95%^23,24. Performance is better than this for more severe impairment. These sensitivity estimates are likely to be biased negatively, especially for the ABR, because the only way to determine sensitivity is to adopt a cohort approach in which true hearing status is determined later for all infants screened. Such experimental studies are very demanding, both in terms of sample size requirements and cohort attrition rates. The earliest developmental age at which reliable reference audiometric data can be obtained longitudinally is 8-9 months, and the so-called 'definitive' hearing tests may be inaccurate in infants with any of several concurrent morbidities. Also, there may be intercurrent or progressive impairment that is not present at the time of screening. For specificity, on the other hand, because PCHI is relatively uncommon, cohorts are easier to acquire and screening referral rates approximate the false positive rates closely. Large UNHS programs that have been in operation for several years in the US have provided solid evidence on specificity but cannot determine sensitivity in an unbiased manner because only babies who refer from screening are followed up.

Because PCHI has a prevalence of only about 3/1000, very high screening specificity is desirable in order to achieve a reasonable PPV and keep the number of families experiencing anxiety due to screening refer outcomes to a minimum. One way to achieve this is by maximizing use of AABR, including using it as a series second screen in infants who refer from AOAE. The series sensitivity is upper-bounded by that of AOAE, and the series specificity is lower-bounded by that of AABR. Actual high-quality UNHS programs, easily better than a guideline target maximum rate of 4% proposed by the AAP. By adding a further series AABR re-screen after about a month, the net referral rate can drop to as low as 1-2%²⁵. This is likely to be accomplished with little loss of net sensitivity, because the intrinsic false negative rates of the best AABR response detection algorithms are less than 1%, and the main effect of re-screening is to eliminate false-positives that arise due to transient, perinatal middle-ear conditions.

Overall effective sensitivity

A refer result from screening does not, of course, mean that the infant has PCHI, but that the risk of PCHI is very high. The purpose of screening is essentially to deliver all infants who are at very high risk of having a genuine PCHI to comprehensive audiologic assessment. Failure to deliver vitiates the purpose of doing the screening. Thus, there are three major facets of the overall process sensitivity: screening coverage of the target population, the net sensitivity of the complete screening protocol, and the linkage of screening referrals to the definitive audiologic assessment. Evidence from the large UNHS programs in the US suggests that screening coverage of over 95% is achievable, given commitment and good program design. Logistical difficulties tend to arise not from sheer volume, as in well-baby nurseries, but in unpredictable discharge and transfer that are typical in NICUs. Screening protocol sensitivity has been outlined earlier. The rate-limiting step is usually tracking and referral compliance for audiologic assessment, and a success rate of over 80% for that step is unusual²⁶. Because the net sensitivity is the product of the case throughput rates for all the component stages, it follows that great attention must be paid to case tracking and compliance for referral. The key ingredients are the information system that records and flags infants requiring assessment, and a careful process of family information and instruction that promotes compliance yet is sensitive to family anxiety levels.

Definitive Audiologic Assessment

A widely endorsed goal of early identification programs is to initiate enhancement of hearing and/or communication development by six months of age³. This requires that wherever feasible, audiologic assessment should be completed by about 3-4 months. In infants less than about 8 months developmental age, the only method in widespread use at present is electrophysiologic estimation of hearing thresholds, using the ABR. Such methods do not require overt behavioural response, and rely upon the detection of electrical activity in the auditory system that is a correlate and statistical predictor of actual perception of sound. In combination with auxiliary, physiologic tests of middle-ear function, brainstem neuro-myogenic reflex arc integrity, and cochlear hair cell function, ABR estimates of hearing sensitivity can be reliable and accurate²⁷.

In most infants with a developmental age over about 8 months, it is feasible to conduct accurate, behavioural testing of hearing using Visual Reinforcement Audiometry (VRA)²⁸, a method that relies upon operant conditioning of a head-turn response to controlled sounds. However, in infants with cognitive delay or other neurodevelopmental problems, behavioural methods may remain unfeasible and there must be continued reliance on electrophysiological methods.

Both electrophysiologic and behavioural assessment of hearing in infants requires careful technique and considerable expertise, to obtain accurate results. Many errors are possible and these can have damaging consequences. It is essential that well-designed test protocols be used consistently, and that testers have access to adequate training and continuing education, as well as that they have sufficient case load to maintain and develop skills.

Hearing and Communication Development Options

There are many options for improvement of hearing ability and/or for enhancement of communication in infants with hearing impairment ^3,29. It is important that all these options be accessible and that families of infants with PCHI be familiarized with them in a manner that is sensitive, culturally appropriate, evidence-based and unbiased. Psychological support for families is also important throughout the process of understanding the implication of hearing impairment and developing a positive and proactive role in selecting a course of action and promoting its success.

One option is to pursue maximization of hearing ability and normal development of speech and oral language. This option is commonly, but not necessarily, chosen by families of infants with less than profound impairments. Hearing aids and related assistive devices, if appropriately selected and fitted, improve hearing in most cases and can lead to successful speech and oral language outcomes. In infants with severe and profound PCHI, cochlear implants are often effective, if hearing aids prove insufficient. In this situation, as well as for some infants with lesser degrees of PCHI, it is important to combine the assistive technology with training in auditory perception and speech development that may be extensive and prolonged.

An alternate route to successful communication development is manual communication, which methods include American Sign Language as well as various manual signing aids to speech and oral language. ASL is a distinct language in its own right, not an approximation to spoken English. Families, especially those with affinity for the culture of the Deaf community, may prefer to eschew auditory-oral approaches to communication or to develop ASL as the primary language of their child and consider auditory-oral options at a later stage. The evidence with respect to comparative effectiveness of various approaches to auditory-oral communication development is inconclusive, but there is an increasing evidence base that language stimulation by any means is important in early child development³⁰.

Ontario's Infant Hearing Program (IHP)

Funding for enhancement of services for newborns and infants with impaired hearing was announced in May 2000. The IHP complements and substantially augments existing provincial services for Preschool Speech and Language Development. The key elements of the program are:

1. Secured funding to develop and deliver all necessary program services and supports.

2. Universal newborn hearing screening during the perinatal hospital admission, augmented by follow-up community screening services.

3. Comprehensive services for confirmation of hearing impairment and diagnostic assessment sufficient to inform medical and non-medical processes, as elected by the family.

4. Family psychological supports and provision of comprehensive, unbiased information about hearing and communication development options.

5. High-quality services for provision of hearing aids and other assistive technologies.

6. Expanded provision of comprehensive and diverse communication development services.

7. Auxiliary sub-programs for detection of emergent hearing impairment in at-risk infants and in those who develop post-natal risk indicators for PCHI.

8. Educational and informational programs for families and hearing healthcare professionals.

Infrastructural elements that facilitate provision of high-quality, integrated services include, but are not limited to: optimal, evidence-based protocols for screening, audiologic assessment and provision of assistive technology, comprehensive information systems to ensure reliable tracking of infants and recording of outcomes, extensive training programs for personnel involved in screening, assessment, assistive device provision, family support and communication options provision, communications development for families and professionals, administrative systems to ensure necessary operations including supplies and services, and a substantial subprogram for evaluation and quality improvement.

The overall IHP model is one of centralized planning, development and quality management, combined with regionally adapted implementation that reflects local variations in need and infrastructure. The training and services enhancement processes are nearing completion, systematic screening of infant at high risk is now widely implemented, and universal screening has been implemented in many regions of the province. Full functionality is anticipated in early 2003.

Some 30,000 newborns have been screened to date, and assessment systems are fully functional. A net referral rate to audiologic assessment of about 1% is already among the best achieved in large programs in North America. There is every indication that the Ontario IHP is securing an exemplary position globally in terms of high-quality healthcare for newborns and infants with significant, permanent hearing impairment.

References

1. American Academy of Pediatrics. Newborn and infant hearing loss: detection and intervention. Pediatrics. 1999;103(2):527-30.
   
   Finitzo T, Crumley W. The role of the pediatrician in hearing loss. From detection to connection. Pediatr Clin North Am. 1999;46(1):15-34.
   
   Joint Committee on Infant Hearing. Year 2000 Position Statement: principles and guidelines for early hearing detection and intervention programs. Pediatrics. 2000;106:798-817.
   
   Mehl AL, Thomson V. Newborn hearing screening: the great omission. Pediatrics. 1998 Jan;101(1):E4.
   
   Bamford J, Davis A, Stevens J. Current topic. Screening for congenital hearing impairment: time for a change. Arch Dis Child:Fetal & Neonatal Edition. 1998;79(1):F73-6.
   
   Sininger Y, Doyle K, Moore J. The case for early identification of hearing loss in children. Pediatr Clin North Am. 1999;46(1):1-14.
   
   Hardie NA, Shepherd R. Sensorineural hearing loss during development: morphological and physiological response of the cochlea and auditory brainstem. Hear Res. 1999;128(1-2):147-65.
   
   Yoshinaga-Itano C, Sedey A, Coulter D, et al. Language of early and later identified children with hearing loss. Pediatrics. 1998;102:1161-71.
   
   Karchmer M, Allen T. The functional assessment of deaf and hard of hearing students. Am Ann Deaf. 1999;144:68-77.
   
   Durieux-Smith A, Whittingham J. The rationale for neonatal hearing screening. J Speech Lang Pathol Audiol. 2000 Jun;24:59-67.
   
   Davis AC, Wharrad H, Sancho J, et al. Early detection of hearing impairment: what role is there for behavioural methods in the neonatal period? Acta Otolaryngologica Suppl (Stockh). 1991;482:103-9.
   
   Lyon M, Lyon D. Early detection of hearing loss:a follow-up study. Can J Public Health. 1986;77:221-4
   
   National Institutes of Health. Early identification of hearing impairment in infants and young children. NIH Consensus Statement 1993;11(1):1-24.
   
   Grandori F, Lutman M. Clinical Focus: grand rounds. The European Consensus Development Conference on neonatal hearing screening. Am J Audiol. 1999;8(1):19-20.
   
   National Centre for Hearing Assessment & Management (NCHAM): www.infanthearing.org
   
   Brown D, Dort J, Sauve R. Newborn hearing screening programs: A truly Canadian perspective. J Speech-Lang Pathol Audiol. 2000;24(2):48-58.
   
   Fortnum HM, Summerfield AQ, Marshall DH, et al. Prevalence of permanent childhood hearing impairment in the United Kingdom and implications for universal neonatal hearing screening: questionnaire based ascertainment study. Brit Med J. 2001 Sep 8;323(7312):536-40.
   
   Mehl AL, Thomson V. The Colorado newborn hearing screening project, 1992-1999: on the threshold of effective population-based newborn hearing screening. Pediatrics. 2002 Jan;109(1):E7
   
   Prieve B, Stevens F. The New York State Universal Newborn Hearing Screening Demonstration Project: Introduction and Overview. Ear Hear. 2000;21(2):85-91.
   
   Vohr BR, Widen JE, Cone-Wesson B, et al. Identification of neonatal hearing impairment: characteristics of infants in the neonatal intensive care unit and well-baby nursery. Ear Hear. 2000 Oct;21(5):373-82.
   
   Cone-Wesson B, Vohr BR, Sininger YS, et al. Identification of neonatal hearing impairment: infants with hearing loss. Ear Hear. 2000 Oct;21(5):488-507.
   
   Starr A, Picton TW, Sininger Y, et al. Auditory neuropathy. Brain. 1996 Jun;119:741-53.
   
   Thompson DC, McPhillips H, Davis RL, et al. Universal Newborn Hearing Screening. J Am Med Assoc. 2001(Oct);286:2000-2010.
   
   Norton SJ, Gorga MP, Widen JE et al. Identification of neonatal hearing impairment: summary and recommendations. Ear Hear. 2000 Oct;21(5):529-35.
   
   Keren R, Helfand M, Homer C et al. Projected cost-effectiveness of statewide universal newborn hearing screening. Pediatrics. 2002;110(5):855-64.
   
   Prieve B, Dalzell L, Berg A, et al. The New York State universal newborn hearing screening demonstration project: outpatient outcome measures. Ear Hear. 2000 Apr;21(2):104-17.
   
   Stapells DR, Oates P. Estimation of the pure-tone audiogram by the auditory brainstem response: a review. Audiol Neuro-otol. 1997;2:257-80.
   
   Widen JE, Folsom RC, Cone-Wesson B, et al. Identification of neonatal hearing impairment: hearing status at 8 to 12 months corrected age using a visual reinforcement audiometry protocol. Ear Hear. 2000 Oct;21(5):471-87.
   
   Knott C. Universal newborn hearing screening coming soon: 'hear's' why. Neonatal Network. 2001 (Dec);20(8):25-33
   
   Ruben RJ, Schwartz R. Necessity versus sufficiency: the role of input in language acquisition. Int J Pediatr Otorhinolaryngol. 1999;47(2):137-40.